AtuwaDa A t7.es Singapore S *M.9S Malaysia M «14.2S New Zealand NZ fO-BS Hong Kong H (79.00 Sweden 30:-SEK $2.95 USA 0S-9 Atari Amiga Mac S-50 6800 680» 68*08 6U000 68010 68030 68030 Tlio Magazine for Motorola CPU Devices F *r Over a Decode! This Issue "C" User Notes pg. 6 100 lb Software User Notes pg. 1 1 FORTH p. 16 Macintosh - Mac Watch pg. 28 FLEX 6809 Diagnostics p. 36 Til* Bit Budwl - New* RcImms, Lctttrc, Updates ttc & DO Continued aC»D0S Atari Anagi And Lots Mare! U5*9 FLBX Midnlod) A tlur Comriiutor frumat The problem wilh most real-lime operating systems is simple, they're not an integrated solution. You end up dealing with a multitude of suppliers for languages, compilers, debug- gers and other important development tools. And when something does go wrong, it can be a frustrating experience hying to straighten out the mess. Why Not Try the Microware One-Stop Total Solution? Microware's OS-9 Real-Time Operating System is a total inte- grated software system, not just a kernel. VVfe offer an extensive set of development tools, lan- guages, I/O and Kernel options. And this total integrated solution is entirely designed, built and supported by the same expert Microware team. MJrroWiV¥$Vtinn\ColHff.HKifi C5-9 14 .1 (Mttmutk id MitrtfWjrr Ll NIK b j IMduiviik ol At M VAX iw l>.xk<mi/i u( IXC The OS-9 Success Kit A Total Integrated Solution for Your Next Project Development Tools: Languages: C Source level Do bugijer c- Symbolic Debugger Bask System Slaw Debugger Pascal uMACS Ten Editor Fortran BeclromcMall Ada" Communications Assembler' Super Shell Kernel Options: I/O Options: SCSI SASI&SMO Disks MMU {Security Protection) SuppOil J-.5-. »-inch Diskettes Mitfl Coprocessor Support Magnetic Tape ■ RtsdBit or uMX verslent smote atiemet-TCP/tP ••VAXHosW Aiaiet- OS-9 Net Modularity Lets YOU Choose fust What You Need. The modular design of OS-9 allows our Operating System to adapt as your requirements change. OS-9 can support a complete spectrum of applications — from embedded ROM- based code in board-level prtxJucts all the way up to laige- scale systems. Support is Part of the Package. Microware is proudly setting the indusliy's standard for cus- lorrter support. \bu'll find pro- fessional and comprehensive technical documentation and a Customer Hotline staffed by courteous and authoritative soft- ware engineers. So slop messing with sim- ple kernels and independent suppliers. Call Microware today and find out more about the "One-Stop Integrated Solution" with OS-9! -~7nictowMc^' OS-9 Microware Systems Corporation 1900 N.W. 114th Street Des Moines, Iowa 50322 Phone: 515/224- 1929 Western Regional Office 4401 Great America Parkway Santa Clara, California 95054 Phone: 408/ 980-020 1 Microware Japan Ltd. 41-19 1 loncho 4-Chome Funabashi City Chiba 273, Japan Phone: 0474 (22) 1747 Muataog-020 Miutsuig-Oe Benehmarks !BMAT7300X«naSyO AT*T 7300 UNIX PC 690)0 D6C V AX 1 1/780 UNIX tWIuoy 4 1 □eCVAXU/WO ' WrX» OS-» UK s Mhi 68000 054 WK 10 Mha MVSTAHC-M «*0M OJ-»«» f « WAi MVST*NG4aaima os-e tst is wu UUiTAUGJUOUeV UCiiUl Um&UtX M Mkt 32 Ml Ra (totes mleaer Lueg 9.7 7.2 4J 3.6 3.2 3.1 3.7. 3SjO 9.0 6.5 44 M <J Z2 0J3 t-t 122 Ma*« for (i.0: 1 < 999999-. .-.i). EMiratcd MIPS - MUffrANC-ttlO _. *i MIPS, Dura ul.ll MIPS: Motorola Seats OS-9 OS.9 Probuioul Vo> stvuw •ti»Jv"fc.CC^n»k. BaskO? 300 SO CGsnpUa* SOO.OO aeOOO OoaasemHm (wAwrm „i± 1100.00) loom FanrsoT? 730 00 MoomfKil WD 00 OrapwA Pwal 900 A) Styio-Orsr* 495.00 s<y>e-sa>u ) 95.00 SrjtoMe.se 173.00 SijIo-Qbob-SP* M°»I» 695.00 PAT »/C auv 229.00 JUST w/C mm 79.95 PAT/JUST Canto J49J0 Jcillf»«» M balm) 995.00 COM 123.00 UniFLKX VtuFlXXIoMlOwj) MSO.00 Sclera Saner 150.00 •SUf 'Mni 200 no G A S dOf«CanQ^ 300.00 *C CtJSTpkava)* 3» no coaot 750.00 CM006M wAuara. 100.00 TM006M .Auoftii 100.00 X-TALK'e»Ad) 9955 C~rra A^esnhfcer SOjOO Fortran?*) 450jOO Sailpiorr (•» Wow) 995 jOO Strain* MUJTANGflOT* shipped 115 Mhz. A<Mtc»16J5Mh«6»a20 mxt> Aodfoj)6J5Hfli.t«BJ] 375 DO ADJf.»S>Mhi<»»T3XAM 730,00 16Pon«pR5.732 i >> in CUaumal a 2 A*i<pt<x Canal balm JUU2 AJopKr 163 00 usta aud pbhj*" * aratrrkral tee p«*l» <ioul of 36 asrtal ports eusyuueel 1 60lmePsnlialVOcanl 198.00 In 3 6*230 Iraoetsce/Ticnr cfcsps, 6 jroupi of S lp« aacfa. vpvsn H*j*ci dnoioa ooneroj for each trown. I Tl^LafVwa D04Ju «•«* fcr boOj dip «od PGA dwkm A ■ pt^wired VBD7 «« ■ Op u> 512K ORAM. 4?5j00 SBC- AN ftgCyST M_U£hd «*»»»- [nuniiiit I ftN.fiWar*^" ! 11 ^* 1 -call. LAK eA«»7B— '(*» 120.00 Bit^iwbMcfcvi^t/OQHBriMqfkka SI 99,00 Spot mJ far* i ^irm M\JSTMtG-(W* tj*WB*Wf*vw ■ Scolptor* $«SJXLSAVBJmX00 Softwm rt DCnL« u AD MUSTAN&02CP* « diKSVra en ill ltftM* «DtWan: 10-70* t •mw (or tj-o-ftp- DtfunuD pp»f»)r «Ac*r 0» win * The P R O f Mustang Specifications 123 Mhz (optima) 16 6 Mhz available) MC68020 full 32-bil wide psdi 32-bii wide data and eddies* buses, n on- multiplexed on chip instruction cache object code com pal ibtc with all 68XXX family preevsors oxhaiiaed instruction ad- malh so- processor interface 68881 mathbi-armd (loatmgpoinl o^jjroreaaor (optional) direct extension of fill] 68020 uuflsucfion act full suppot IEEE P7S4. draft 100 tranzoBToicrtlal and olher aocmifi c malh fuocumu 2 Magabytc of SIP RAM (312 i 32 bit caganization) up to 256K byte* of EPSOM (64 i. 32 bita) 4 AjynchjonouJ achal I/O porta flandard c^tiana! to 20 aerial ports aiandatd RS-232 interface optional network interface buffered 8 bit parallel pott (1/2 MC68230) Ccntrrauca type pinout expanaiofi cartnoctor far 1/0 dovricot I6bitditi pith 256 byte addresa tpacs 2 mtorupi inpiu clock and control xigTiiU Motcrola t/O Qianncl Modules lime of day clock/cal aidar w/baucry backup eaiunllcr for 2, 5 1/4* floppy disk diivc* single of double aide, tingle or double dentil y 35 to 80 track •elccuhls (48.96 TF1) SAS1 inlcrfaco programmable periodic interrupt gcnaaior inlcrrupt rata Cram raicrr>iecand» Loaeconos Highly accurate time base (5 PPM) 5 bit acme switch, readable by the CPU Hardware autgle-rtep capability ItietDhiiraaad 68020 ayalenu an pre* cully working at NASA, Atomic Energy Cacnmuvton. Goverron ail Agencies as well as Uruvcmuea. Susinou. Laba. and other Criucal Application! Centers, wcntdwide. where speed , mlth cruncliing and cnulu -uicr. mulu. tasking UNI X C level V compaubllity and low cost is a minx Don't be mislead! ONLY Data-Corn p delivers the Super SfVSTANG-020 Only the "PRO" Version of OS-9 Supported! This is HEAVY DUTY Country! Fori isTtrM ilnwin wii oftw a *400 tiaoa-gi on youTH oldooXXXSaC Must b* wnrKa^g propMrr and " L-BrTiT(ait» wnh a1 aortwara. ubust and ooainrnmtMm , Can for mom information Price List: Mustang.420 SRC J249O.0O Cabinet w/swuching PS J299.9J 5* 80 track noppy DS/DD J269 95 Roppy Cable $39 95 OS 9 68K Ptofcsnooil Version $850.00 CCompilcrtJJOO Value) N/C Winchester Cable $39 95 Wtncheeier Drive 25 Mbyte $895.00 Haul Dui Coo trailer $39500 Shipping USA UPS $7J0O UniFLEX Una $100.00 MC6888! 1/p mslh prorew Add $27501) 1667 Mhi MC68020 $375.00 16.67 Mhz MC6888] $375.00 20 Mhz MC68020 Sya $750.00 Note all 6888) craps wok with 20 M)iz Sya Total: $5299 80 NEW LOWER PRICES 25 Mbyte HD $3299311 $3749.80 85 Mbyte HD $524&gH $4548.80 f Note: Only Professional OS-9 Now Available (68020 Version) Includes ($500) C Compiler - 68020 A 68881 Supponcd - For UPGRADES Write or Call for Professional OS-9 Upgrade Kit Data-Comp Division A Decade of Quality Service" Tj V^ ~nSBr*' ' Systems World-Wide Computer Publishing, Inc. 5900 Cassandra Smith Road Teteprtooe 615 842-4601 - Telex 510 600-6630 Hixson, Tn 37343 68 rvfero Journal April '89 A Member of the CPI Family 68 Micro Journal 10 Years of Dedication to Motorola CPU Users 6800 6809 68000 68010 68020 The Originator of "DeskTop Publishing™" Publisher Don Williams Sr. Executive Editor Larry Williams Production Manager Tom Williams Office Manager Joyce Williams Subscriptions Cheryl Hodge Contributing & Associate Editors Ron Anderson Dr. E.M. "Bud" Pass Ron Voigis Art Wellcr Doug Lurie Dr. Theo Elbert Ed Law & Hundreds More of Us Software UscrNolcs FORTH Logically Speaking LMac-Watch FLEX Diagnostics - Disk Drive Test, ROM Test, RAM Test 36 Korpi Hit Bucket "" Classifieds 68M1CROJQ (^Contribute Nothing - Expect Nothing' DMW 1986 D m^T ^ r7 <rz-~. x- COMPUTER PUBLISHING, INC. "Over a Decade of Service" I'v-y <ic vifi].Lv^*^m>. ■World Wide" 68 MICRO JOURNAL Computer Publishing Center 5900 Cassandra Smith Road I'O Box 84'i Hixson, TN 37343 Phone (615) 842-4600 Teles 510 600-6630 Copyrighted © 1987 by Computer Publishing. Inc. 68 M icro Journal is the original" DeskTop Publishing" product and lias conlinously published since 1978 using only micro-computers and special "DeskTop" software. Using first a kit built 6800 micro-corn puicr, a modified "ball" typewriter, and "home grown" DeskTop Pub lulling software. None was commercially available at that lime. For over 10 years we have been doing "DeskTop Publishing"! We angi noted what has become traditional "DeskTop Publishing " f Today 68 Micro Journal is acknowledged as the "Grandfather" of "DeskTop Pub- lishing" technology. 68 Micro Journal (ISSN 0194-5025) is published 12 times a year by Computer Publishing Inc. Second Class Postage paid at Hixson. TN. and additional entries. POSTMASTER: send address changes to 68 Micro Journal. POB 849. Hixson. TN 37343. Subscription Rates 1 Year $24.50 USA, Canada & Mexico $34.00 a year. Oihers add $12.00 a year surface, $48.00 a year Airmail, USA funds. 2 years $42.50, 3 years $64.50 plus additional postage for each additional year. Items or Articles for Publication Anicics submitted for publication must include authors name, ad dress, telephone number, dale and a statement that the material is original and the property of the author. Articles submitted should be on diskette. OS-9, SK»DOS, FLEX. Macintosh or MS-DOS. All printed items should be dark type and satisfactory for pholo-rcproduc tion. No blue ink! No hand written articles - please! Diagrams o.k. Please -do not fo mat withspacesany text indents, ch rts. etc. (source listing o.k.).Wewi edit in all for matting Text should fa I flush left and use a carriage return only to indicate a paragraph end. Please write for free authors guide. Letters & Advertising Copy Letters to the Editor should be theoriginal copy, signed! lyeuers of grip as well as praise are acceptable. We reserve the ighl to reject any letter or advertising material, for any reason we deem advisable. Advert is ing Rales: Commercial please contact 68 Micro Journal Advertising Department. Classified advertising must be non-commercial. Mini- mum of $15.50 for first IS words. Add $.60 per word thereafter. No classifieds accepted by telephone. April '69 66 Mcro Journal r JUST *\ PAT t With 'C Source $229.00 stemni! FiSaiDnM A Division of I y^ Media 5900 Cu usiki Smith Rd Hiuoa. Tn 37343 TckrttoM 613 842-4809 Tck>310 6OO-6<3Q I14M2I ^ PAT FROM S. E. MEDIA » A FULL FEATURED SCREEN ORIENTED TEXT EDITOR with all the best of PIE. For those who swore by and loved PIE, this is for YOU! All PIE features & much more! Too many features to list. And if you don't like ours, change or add your own. C source included. Easily configured to your CRT terminal, with special configuration section. No sweat! 68008 - 68000 - 68010 - 68020 OS-9 68K $229.00 COMBO Special $249.00 JUST JUST from S. E. MEDIA - - Text formatter written by Ron Anderson; for dot matrix printers, provides many unique features. Output formatted to the display. User con- figurable for adapting to other printers. Comes set-up for Epson MX80 with Graflex. Up to 10 imbedded printer control commands. Compensates for double width printing. Includes normal line width, page numbering, margin, indent, paragraph, space, vertical skip lines, page length, centering, fill, justification, etc. Use with PAT or any other text editor. The ONLY stand alone text processor for the 68XXX OS-9 68K. that we have seen. And at a very LOW PRICE! Order from: S.E. MEDIA - see catalog this issue. 68008 - 68000 - 68010 - 68020 OS-9 68K ^ With 'C source $79.95 J 68 Mcro Journal April 69 ©IL@ @ot ©iPi^^niiiL SCULPTOR From Hie world's oldest & largest OS- 9 software house! cits ric4K.ic\>nii\c- time up to 80% 6809/68000-68030 Save 90% SCULPTOR-a 4GL ■ Only from S.E. Media at these prices. OS 9 levels one and two (three GIMIX) 6809, all 68XXX OS-9 standard systems. Regular SCULPTOR versions 1.4:6. One of if not the most efficient and easy to develope DBMS type systems running under OS-9. A system of llexible keyed lile access that allows extremely fast record and data retrieval, insertion and deletion or other programmed modifica- tions. Access by key or in ascending order, very fast. The system provides automatic menu generation, compilation and report generation. Practically unlimited custom input format and report formatting. A rich set of maintenance and repair utilities. An extremely efficient development environment that cuts most programming approximately 80% in development and debugging! Portable, at source level, to MS- DOS, UNIX and many other languages and systems. Standard Version: 1.14:6 6809 - $1295.00 68000 $1295.00 68020 $1990.00 r Due to a "Special One Time" Purchase, We Are Making This Savings Offer. Quantities Limited! Once this supply is gone the price goes back up! System OS-9: 6809/68000-68030 • Regular s 12 9 5.00 ONLY ^ $ 99.95 S.E. MEDIA POB 849 5900 Cassandra Smith Road HIXSON, TN 37343 61 5 842-4601 Telex 510 600-6630 FAX (615)842-7990 AVE - WHILE SUPPLIES LAST! April '69 68 Micro Journal %i$) 842-4600 Sou tfi "Las t Media Tefex; 510b00hbA0 OS-9, 'Um'JCEX, 'JCLX, S%*DQS '/a*: ($15)842-7990 Sculptor Full OEM & Dealer Discounts Available! n^ soi; n i. ik v.k mm a ■**** tjn ht to tiara*) null In* horn rut package AN ESTABLISHED LEADEfl afta pMooli rappr*! prwralo' dtHfyMHuv J evp* mitv h* chfwm ■ni chm, math laadbarh (rcmr* an nr wcti— nj tuitoflw baa*. Sculptor Kim bam rrtnad and mhwcWl no bacorTW <**<Atw itMBU riwawhif int. f^afauwafl rabbi* -vilrrttt M ih* Najajl anJ*v SYSTEM EMJEPENDCNCE Scutpani ■ ruaahl run many dtfhftfw rrm-hirwi arid a* mtfaOltlft^PfftoM, ingtrirM WS IX rv Unii.lUnujimli VMS th mfti rt w b* ■.* wmwaanJ hmhuwtinafi hm *nwlp«rHfliJC4'npvWft, * ,h3 »"J h rtluta - ™« wan™ 141*1 larga flwand maaniranMii Snip** an tta Far nt Facts DATA DICTIONARY Ejk h Ur may hiw on* at iwh m (*J Ivpr* ijWk rrbrd I '*fkk may haw» • hwtx- -rwaclvu fvp< vaJh UtW<n fcu F uLd | bom HIMI fltMtr-ny [mMitl _, DATA RLE STRUCTURE ■ ■■■■■■■ : ! Mrrfi^th¥tfdmkniv*4('iji9vnfvi* l « ■■■■■■■■■■■■■■■■ ! ! Dam****!** inl*g« nl*» »Mir**n INE»E XING TECHNIQUE Sc utpscir trauma*** B Br* ndn *►* **f H tlttl Nr ftt^Mfi V4C 4* ■**> *flry rHJmb*TI 1 J nJltfTEuiBuc lr*W *<^ Irn ru- coded tn*o ow ctftw At INPUT DATA VALIDATION input rial* m#y hi v**ii*hr>f n *»** RELATIONAL OPERATORS fatal lo UwituJi L«h than 01 equal u Gaataaf than en *nuaJ * 1 Nunmiwlm I (JflUKl* nraalasa* ARITHMETIC OPERATORS Untiui SPECIAL FEATURI* l: 1 .iii. 1 APPLICATION PORTAB1I ITY PRVHM Sculptor for 68020 OS-9 & UniFLEX $995 nwn Scjpd.t w Vatican' -njPlhfiiionilar^myftvtfcri-fi^ifHi F™ fun a%#r baler* ■ ■ ** tyarm MMhrnoh *pp»ddrv*iwi pyp Aral ■ SPEED AND EFFICIENCY Sculptor una* a lariand prwoi md*ung MhrKM wrfMrb SoA** * him* frrmtfr* ■ <w»v9> 1* ««pM *>* compact iimiiiwirnn fed* *ha-h **#nja« »** bnnHR ISTERNATTONALLY ACCEPTED By <W*Q * Mmpk HTt*if^m*fli wi 1. aim.'. V u^* » <«n iwvwdl V*<nlrwla#>OMa||f»rKltipmi*MK«y»Mjr» V wF Th*n ■bnU thaAnwvdAM^'ll utt^jw nxwttPT *Kh* i»*iiW Amn. - SnJpfcn * ahvadv at «™* ftrniaar*arf Addnon r \ cS L^ \ jrf JSjjii*- I f rsi ^^ tii 1 ^^^ MAXIMA AND MINIMA Mirtimum Wv t™jfrh 1 1 Miumum k#v bri-Hjcfi :. '* 1 b Marumum wt nfd L*yth 3 b Haurmim fpcok! t#n^h Ji?767 b HfltlrTMiiim rwlihfMTfri.mJ JT7fc7 MulrtHfm rn™h prr Mt lb mi Munurn Urt p#r ptripam I f > Uuffvtaia^- . Krt if^Uri Ut ! RvtiuJd ht*J*» AJwr l*i»qi>agr *rn ! ii»w ii kumI PROGRAMS . CrrjBhrtwtii mdnHfraV ■ tnWrff*tr lUnMKiwn fcwm rm-^am ' {WiwraM tunlard wpon rtiafarn ' CnjAa^aV *mwi lo*m pfi*a"ans Ci ifnpaV tvpLai (mj^rwn : Kapnrt-pm^wnKnhrTWrlrr SCREEN FORM LANGUAGE : [VmJunrnnwT , |rtr*[ il*»n an. 5 H*jhlj4r ranijfirtt atii [W4autl *.rt {*i4att*rwi c« rpoofi eondHLim : Primrttul «rrti* -a* ifinji ,V(^nl PalAKMH : imubafi*xT» Jnf^ts 11* munjib Fadfefy lo«l a^9 f*»^»m*4ntl (wraAng Maawi rnfnmflBidk : r t «iJffliwwJi*a*rTTwr«h • Irwfa ptfraVi j <J awnwnal Mpr r" MUSTANG-020 U»rs - Ask For Your Special DLscount! n MUSTANG-020 *$1.990 $398 $795 *$1,295 $259 $495 PC/XT/ AT/MSDOS $695 $139 $299 L MUSTANG-08 "$ 1.295 $259 $495 Call or write for prices on the following systems. XENIX SYS III ft V. MS-NET. UNIX SYS III & V. ATARI OS 9. 68K. UNOS. ULTKIX/VMS (VAX. REGAL). STKIOE. ALTOS, AI'KICORT, AKETE. AilM STRONG. BLEASDALE. CIIA«1>;S I0VEKS. GMX. CONVERGTECH. DEC. CIKER. EQUINOX. GOULD. III'. HONEYWELL. IliM. INTEL. MEGADATA. MOTOROLA. NCR. NIXDOKJ''. N.STAR. OUVETH/AT&T, ICL. I»ERKINS EIJUEK. i'HILLll'S. HXEL. I'LESSEY. MJCXUS. ItJSlniON. PRIME. SEOUEYI SIEMENS. SWTPC. SYSTIME. TANDY. lORCli. UNISYS. ZYLOG. E1C. * For SFKCIAI. M1W SCLIJTWH prlcivs csocvially for BSOB^riKXXX <IS-» Syswms • St-c Spe- cial \d thiw Iwuc. Itt'm«*mb«*r, M When they art* ifont* lht> prltn* ^m". bat*k up a** :ilio% <■:" ... Sculptor Will Run On Over 100 Other Types of Machines ... ... Call for Pricing ... !!! Please Specify Your Make of Computer and Operating System !!! J * Full Devdopement Package •• Run Time Only *** C Key CTIc library AOlltWIK} Ltfiolb O • OS-*, S ■ SK-DOS r . FLO, . UOIED CO . Ctfor Ca^Hiw OS-* CO> . Oaior O m fut tr KLtl South Tost 'Media $900 Cnssambu Smith 3fo£ - Jftat on, Tn. 3 7343 Ttltpfvmt: (61$) 842-4&0 TtlvQ 5106WC&0 •• Shipping ** .. KU J» MSA. (ml» X1J0 Rnlp Serfca Add 5* For«ltli Alraull Add !«• Or CAA SKlppinj On*/ *Q5.* Ii ■ Tr»dwwt rf Mkrowri Hid Motenib-'KLIX .Bd UoH.m f « I r»«ttrmkJ J lichnkJl Sy*» CormdMB-'ilCDOS \s % 1>«dMimt «f »ir.K Saltwm S^^n. I jiry. 68 Micro Journal April '89 The C Programmers Reference Source. Always Right On Target! C User Notes A Tutorial Series By: Dr. E. M. Bud" Pass 1454 Laua lane N.W. Conyers,GA30207 404 483-1717/4570 Computer Systems Consultants INTRODUCTION This chaplcr begins with a siatcmcnt by Dennis Ritchie concern- ing the ANSI C standards committee. He placed il on USENET, with the usual disclaimer that the opinions expressed arc his, and not those or his employer. Il is reprinted here verbatim, without editing. For those who do not recognize the name, he is the "R" of"K"and"R". For the tenth anniversary of the original publication of the K & R book on the C language, there is a new edition, encompassing the new fcalurcswhich have been added to the language in the last ten years. Although i tat tempts tocovcr the new ANSI C standard, be aware that the standard is still changing and some features noted in the book have been deleted from the standard. This includes the "noalias" keyword. The "volatile" keyword may also be modified in meaning before the standard is finalized. Following the siatcmcnt by Ritchie arc commentaries (also from USENET) by others concerning various aspects of the C lan- guage. Their arguments show that the C language definition needs additional specification in several areas. The example program produces text suitable for input to a text processor, such as nro, from plain text not containing the neces- sary commands. EVALUATING ANSI C . Dennis Ritchie Since the subject has come up, this is a good time to record my thoughts on ANSI C and the work that X3J 1 1 has done. In brief, 1 think the result is commendable. I concur in the belief of those who watch such things that X3J1 1 managed to improve on what they started with, and that this is an unusual accomplishment for a large committee that lasts for 5 years or so. In particular, they successfully resisted the usually inexo- rable pressure to add features and options. The committee had ccitain explicit goals, among them to bring tlic language specification up to date with reality, since much had changed in the 10 years since thcoriginaldclimtion; to add a very few things deemed necessary (function prototypes); and to spec- ify things more completely than K&R did. They also wanted to standardize the library, something that has become important not only because the drift in the System V vs. BSD worlds, but even more because of the vastly increased use of C on non-Unix systems. Perhaps as importantas all the rest was a goal that was not stated explicitly, namely to supply the stability, legitimacy and gcncial cachet that is possible only with all the legalities required to gel the result accepted as an Official Standard by the appropriate Official Bodies. The need for this can be questioned, but the fact that many people want il to happen, and that it lends to occur regardless of desires, can't be denied. I have only a few worries about the specification of ihc language. First, the introduction of the new function declaration mecha- nism. The new scheme is belter than the old, bul the change is going to cause trouble, and the need lo accommodate both ways confuses Ihc language specification and will confuse users. 1 made my peace with the change some years ago; il is better to do il than not. Second, 1 don't think type qualifiers have been fully digested. 1 never did object very strongly to volatile, even though I maintain that il is not onerous to live without it. The remaining qualifier, const, suffers from tension between conflicting views of what il is supposed to mean. One possible view is ihat const things arc real constants. If il prevailed, then one would expect that things of type const could appear where constants arc expected (case labels and so forth). Perhaps even the notion of pointer inconstant would become suspect. This was not the view thai ihc committee eventually adopted, although (perhaps unconscious) sentiment for it remains; instead, a more implementation-minded approach was taken, namely lhal const means something that can be put in ROM (hardware or writc-protcctcd memory). Nevertheless, the tension remains, and figured heavily in the committee's argu- ments over seemingly unrelated things (including noalias). Even now, questions of whether a pointer lo const can be assigned to an ordinary pointer, or whether these two types arc compatible (and when compatibility mailers) arc lively issues. Il is clear lhat the rulcson qualifiers arc to some extent artificial — they could be decided in several ways. Good rules arc not arbitrary, they arc forced by the logic of Ihc design. April 89 68 Micro Journal Some of the things Utat people arc complaining about were completely necessary, in particular ihc insistence that only a fixed set of library routines is defined, that initial _ is reserved for the implementation, and Utat all other names arc reserved for the user. It is just not possible to write a standaid that perm its the im- plementation to intrude into Ihc user's name space (by letting it define names internal to its library that the user might use by accident), or to give a defined, standard way for the user to replace system routines. Ihc latter would constrain implementa- tions loo much. At the same lime, the consequences of such rules ate misunderstood. A program that calls "read" is not defined by the standard, but I assure you that the supplier of your Unix system will arrange that it works. X3JI I did miss some opportunities. Perhaps the most obvious lack in the language isa scheme for variadic arrays. However, ihc proposals 1 have seen arc awkward and don't fit smoothly, so it is not sui prising that nothing was done. Another thing that must await the future is a genuine rethinking of integer arithmetic, not just the fiddles thai they uid. All in all, I think X3J11 did an excellent Job. When they began, several years ago, I was somewhat apprehensive about what would result, but also decided ihat I did not have the stamina to become involved in their activities; 1 would have to irust their good sense. I have never regretted the decision, and I'm pleased with the outcome. Whether this value implies wriiability is a random parameter, hence the variable nature of the bug. The comma is now a com ma operator, since it is outside the function brackets. Assignment takes precedence over the comma operator, so Ipfd is indeed assigned the return value from openO. but this entire assignment now constitutes the left side of a comma expression, so it is the valucofO_WRONLY that is compared against 0!O_WRONLY is 6efined as a positive constant, so the test always fails and attempted execution of the wriicO becomes inevitable. The lint program did not report an argument count cnor for opcnO It so happens that openO is one of the few system calls thai can take a variable number of arguments (depending in Ihis case on whether the 0_CREAT flag is included in the second argu- ment). Hence lint's argument count check is suppressed by the / "VARARGS2*/ modifier in llib-lc. It look them a very long time to find the problem, as could be imagined. He noted that C's overloading of certain operators (in this case, the lowly comma) increases the probability that an erroneous code fragment will be legally inter prciablc, a situation somewhat analogous to having too many words in a spellchccking diction- ary. This lead directly to the compiler not locating the logic error, since ii was not considered a syntax error GOTOS FURTHER DEBATED OVERLOADING C OPERATORS Ihc following commentary is due to Frank Byrum of BROOKS Financial Systems, Inc. They recently experienced an exceptionally insidious error in their production code, which was causing sporadic write failures oncxecution. This error successfully eluded detection by a DOS compiler as well as by a full liming under Unix System V Release 3. The intent was to have the following C code: if ((Ipfd* op«n(print«r_»t ring (PORT) , 0_WROKLY) » < 0) (fatal trior Mingi) ... writ* (lpfd ) ; In English, if the return value from opcn()ing ihc primer as assigned to Ipfd is not a valid file descriptor, quit; else use it Unfortunately, what was actually coded was the following: if ((Ipfd - op«n(pi:int«r_»tring(POB»)), 0_«ROMLY) < 0) (fatal «croc maaaaga) ... writ* (Ipfd ) ; Take careful notice of the change in parenthesizing. These statements constitute perfecily legal C. Notice first that openO is called wiih an insufficient number of arguments. The stream's status will be based on a random value taken from the stack! The following section was written by William Wells of Proxim- ity Technology of Fort Lauderdale. Florida. As usual, his opin- ions arc his own, not his employer's. Under ordinary circumstances, there is exactly one place where a human C coder might use a goto. Ihis is to implement multi- level breaks and continues. I say this, having managed (and wi iticn huge chunks of) a 1 7,000 line software system (and that is only the part we sell, and docs not include development tools). 1 have programmed inC for six years now and have NEVER used a goto. We have uncounted megabytes of C code written in-house. None of it (to my knowl- edge) contains a goto. Ihc closest thing we have to a goto is setjmp/longjmp, used to implement a multi-level return (and that is a recent change, one whose contemplation caused much debate). With that aside, let mc explain why the goto discussion is really fruitless. People have observed that gotos arc used in a lot of bad code. From this it is concluded ihat gotos are bad. This is really bad logic. Try this: programmers have baenobscrved to write bad code: therefore, programmers arc bad! There is nothing wrong with goto. How do I reconcile wilh my stale menis above? Wait and see... The thing Utat is wrong is the control structures being implemented with the gotos. 68 Micro Journal April -69 The whole point of the structured programming debate is this: every program has a control structure; some of these control structures arc better than others. Whether you use golos or some other language feature to implement the control structure docs not change what the control structure is nor does it affect the goodness of the control structure. Thcqualilyofyour program is strongly influenced by the quality of its control structures. Furthcimorc, you want that control structure to be obvious and understandable to the reader of the program. This implies that you use language features that make your use of a control structure obvious. So, the first question should be: what arc the good control structures? The second question should be: given a particular language, how should the control structures be implemented? Ok, so what makes a control structure good? Well, the basic answers are: a control structure is good if it is 1 ) appropriate to solving problems. 2) easy to write. 3) easy to understand. 4) easy w maintain. 5) ... add your own as long as they do not contradict the above. There arc obviously lots of control structures that meet these requirements and you do not have to use all of them. In fact, you should pick a set of those which are most appropriate for your programming environment and use them. This set should be, in somesense, a minimum one; for example, if you ha vetwoconirol struc lures wh ich can accomplish the same thing, bulonc is easier to use than the other (for you), pick the easier one and forget the other. All other things being equal, a smaller number of control structures helps make your program easier to understand. Now, I hope my claim about our C programs is understandable. But if not, here is what it amounts to: I have chosen a set of control structures which isappropriatc to programming in C, for the kind of programming tasks that I do. It happens that, while my set of control structures includes multi-level breaks and continues (which would be implemented w itha goto), I have never had need of one. Given the amount of code I write, it seems to me that one might never need to use an explicit goto in C code. Now that I think of it, here is a reason to avoid naked gotos in C code: for all other constructs, the control structure being imple- mented is obvious from the keywords employed. This is not true for goto. Therefore, supposing that you have found a control structure that you have to implement using golos in C, you should dress the goto up. As an example, suppose that you arc using the state machine control structure. I normally code it as the following: state = STATB_INIT; while (atate >- STATE_DONE) ( switch (atata) ( cut STATE INIT: 1 However, this is not the most efficient way to do it. You could also implement it as the following: I* Wherever you see the macros stale and ncxlsialc being used, you will be seeing a stale machine. The stale macro defines the start of a state. The ncxlsiate macio causes a transfer of control to another stale of the same machine. A slate machine statts at a 6cfinc of slatcpref and ends with slatc(DONE). */ 6e£ina dummy <x) x 6efine atata (x) £efine nextst&te. (x) 4fdef statapraf Ondef atatapraf lndif 6efina atatapraf dummy (atatapraf) x goto dummy ( atatapraf )x STATE state (INIT) : (coda for thia atata) naxtatata (DONE) ; (mora atataa with appropriate code) atata (DONE) : (coda after atate machine) I am aware that not all preprocessors will do what I want here; for real portability, you would explicitly write the prefixes. Also, this method fails for nested slate machines, something I have occa- sionally had need of. Somcof you will no doubt be thinking: but why should I go to all this effort when I could just use the goto directly? Well, if this was all you did with goto, I don't really see any reason why not (but I do think your program should include a comment saying that you use goto for state machines and describes how you structure it). If, however, you have more than one way of using goto, you should clothe the golos somehow so that the reader of the program knows what control structure your goto belongs to. After all, a while is just a disguised goto. WHITE-SPACE IN C SYNTAX Almost without exception, white-space may be included or excluded from C programs without affecting the logic of the program. The obvious exceptions arc within quoted strings, identifiers, keywords, numbers, and multi-character symbols, where while-space may not be included without splittingonc unit into two. April 89 68 Micro Journal A problem in some of ihc earliest C compilers, which is described in the original K A R book as bcingan anachronism, was ihe older form of Ihc binary-opcrator/assignment statement. Rather than being coded as i -= 1, it was coded as i =- 1. leading to the possibility of an ambiguity if the spaces were removed, since i=- 1 could represent a binary-opcrator/assignment statement or an assignment siatcmcnt. U nfortunatcly . some of the curren tC com- pilers accept ihis syntax, cither silently or with warnings. Certain implementations of the C language, especially older ones, had dependencies and peculiarities which made white- space significant in unexpected contexts. Forexamplc, one C compiler required no white-space between an array name and ihc open bracket introducing the first airay dimension. This situation generated only syntax errors and some grumbling on the pait of the programmers attempting to use the compiler. AnothcrC compiler required no white-space between a function name and the open parenthesis introducing the function argu- ments. Again, this was primarily a nuisance. The same C compiler required no white-space between a par- ametric define name and the open paicnthcsis introducing its arguments. Unfortunately, if it found no adjacent open parenthe- sis, it assumed that the arguments were all null and expanded the define. Then, it encountered the open parenthesis and evaluated the arguments of the define as expressions separated by comma operators. If the programmer were lucky, the compiler generated a syntax error. However, in many cases, ihc compiler silently generated incorrect code because it mis-inicrprcicd Ihc state- ment. C compilers confoiming to the ANSI C standard must ireat comments syntactically as white-space. Many older compilers, and several current ones, treat comments as if they were not present. The difference is probably academic, but may cause minor portability difficulties. There is one ambiguity in the C language definition which has been present since essentially the beginning and continues into the ANSI C standard. If a divide operator is immediately fol- lowed by the asterisk of a dereferencing pointer operation, the result will be indistinguishable from the "/*" multi-character symbol introducing a comment. This will cause a problem in any C program containing this sequence, but it will be exceptionally severe with current C compilers which do not support nested comments, as Ihc resulting comment will encompass all code up to and including the next comment. Ine easiest manner in which to avoid most ambiguity is to use white-space liberally. This assists both the human and the com- puicr in interpreting the program correctly. C PROBLEM Following is a puzzle on unsigned promotions, posed by Sieve Friedl of V-Systcms, Inc. It illustrates a portability issue with sign extension. Can any of you help with a puzzle? I've been trying to understand the unsigned vs value pcrscrving rules of various C compilers. and I 'm afraid I ' ve run into acasc 1 don't understand. I have pored over the dpANS on this but am still confused. Note that this question applies to all compilers, not just dsANSI ones. Oxffff /* what la printed? V 6a fin* HXNDS_ONE main () ( unaignad ahort aval; long lvall, lval2; aval - HINOS_OHE; lvall - - aval; lval2 - - (unalgnad abort) HINDSONE; prlntf <"lval 1 - 71d 2 - Tldn", lvall, lval2> ; I Tba normal anawar I gat la: lval 1 - -65535 2 - -65535 but on the HP9000 I *•• : lval 1-1 2 - .65535 For what it's worih. the HFXXX) has sizcof(short) = 2 and sizeof(int) ■ 4, and 1 get Ihc same results when I 6efinc MINUS.ONE to be (-1) or Oxffff. My specific questions: (1) is this a case of questionable unsigned-ness? (2) if 1 have a vendor who asserts "this is a valuc-picscrving compiler", what is the necessary value of "lval"? (3) if 1 have a vendor who assens "ihis is a unsigned-preserving compiler", what is the necessary value of "lval"? (4) if 1 have a vendor who asserts "this is a dpANS-conformani" compiler", what is the necessary value of "lval"? 68 Mtoo Journal April 89 EXAMPLE C PROGRAM Following is this monih's example C program; it converts text files toa simple nro format It assumes that text begins in the fii« column and lhai nontext (tables, etc.) begins in other than ihc first column. By changing the commands somewhat, the program could be modi- fied to output commands for another text processor. 4ncluda <»tdlo.h> 4ncluda <ctyp«.h> char *p; char string [2S6] ; lnt 1; lnt n lnt a lnt t main (argc, argv) lnt argc; char *argv[ ] ; ( whlla (fgata(atrlng, 256, atdln) ) ( If <*«trlng > ' ') < If (1) 1 - 0, fputa (".fin", atdout) ; If (t) t • 0. f put a (". In On*, atdout) If (a) prlntfC.ap Sdn", a), a - 0; fputa (string, atdout) ; • !■• for <p » atrlng, n - 0; laspaca(*p); ++p, ++n) ; If t*p) < if on 1-1, fputa (~ .nfn", atdout) If (t !- n) printf (".in 8dn". t * n) ; If (a) prlntfC.ap Sdn", •).»■= 0; fputa (p, atdout); » FOR THOSE WJ HMHhl I) TO KNOW 68 MICRO JOURNAL TM 10 April 89 68 Micro Journal SOFTWARE A Tutorial Series By : Ronald W Anderson 3540 S turbridge Court Ann Arbor, MI 48105 USER From Basic Assembler to HLL's NOTES Centcolmnal This is my lOOlh column, though not all 100 have appeared In 68 Micro Journal. Way back when. 1 started out to publish a newsletter, advertised in "Kilo- baud Microcomputing" via letter. I had printed, duplicated and dis- tributed 8 newsletters when Don Williams phoned one evening to ask if I would like to switch to writing a column for '68' Micro. I said that would be fine, but I had committed myself to ^newslet- ters, and had collected a small fee that would Just about cover cost of duplication and mailing. Don said that he would send four Issues of '68' to my list of about 25 Initial subscribers to fulfill my obligation. Of course, my "col- umn" was then read by a few thousand hobbyists rather than just 25 of them. That means that I have written some 92 columns for 68 Micro Journal, and I've missed more or less 4 limes (Including two recently). That means that It has been just about 8 years that I've been here In this spot. My early columns included a lot of software listings, mostly utilities for the 6800 FLEX09 operating system. When the 6809 came along. I did the same for that. 1 had little choice, since there really wasn't much to talk about In software available for those early microcomputers. Technical Systems Consultants had an ad very early in Kilobaud, selling some little games that would run on a KIM-1 board with terminal attached. The games were fun but truly primitive by today's standards. There was HANGMAN, of course, and RE- VERSE in which you tried to order ten digits that had been randomized by reversing some fraction of them from a given position on to the end. The KIM-1 was a 6502 based single board computer with 256 BYIES of RAM. Before the addition of extra memory and a terminal it ran little games on its 6 seven seg- ment LED displays. It had a keypad with which you could explore memory or enter pro- grams a byte at a time. Programs could be saved to audio tape and loaded from it as well. There was a KIM-1 newsletter in which were published little games, such as a horse race in which the three horizontal bars of the seven segment displays were illumi- nated In turn advancing from left to right based on a random number generator. The ilrst one to make it across the 6 digits was. of course, the winning horse. Eventually 1 found a Tiny Basic for it. and attached a 4K memoiy board, adapting its S-100 bus to the KIM bus. About January 1977 I or- dered my Southwest Technical Products 6800 box and a terminal that consisted of a keyboard and a logic circuit board that included a character generator etc. I attached an old television set and the combination could be used as a 40 column terminal! SWIPC used a cassette as the software medium at a whopping 300 baud transfer rate. I remember starting "16K BASIC" loading to the computer and breaking for sup- per. When I came back either BASIC was running or I had an error message and had to start over again. TSC offered their games In 6800 version and 1 bought those again. Then they offered an assembler and a line editor. I bought both (to save money) in source code form and typed in the HEX DUMP to get them running. That was an exercise in frustration, searching for a wrong byte here and there in the code until it all worked cor- rectly. Later I typed in the source code and got it assembled. SWTPC had a printer to go with the computer. You guessed it. it 68 Mkro Journal April '89 11 printed 40 column on roughly 4 Inch wide "adding machine" tape to match the 40 column display. A year or so later. I bought a genuine ADM-3A terminal (with the option to allow both upper and lower case), and then a pair of 5 1/4 Inch disk drives which came with a primitive disk operat- ing system (that worked line. I might add). Six months down the road from then. SWTPC sent all the customers who had bought floppy disk systems a copy of Mlnlllex. the first REAL operating system for the 68XX processors. I remember that they had stuck the disk in the middle of their printed manual and sent It In a 9 by 1 2 envelope which the mall man nicely folded In half the long way. to fit the mailbox. I improvised and found that I could remove the magnetic medium from the disk Jacket by cutting one edge of the jacket open. Then I did the same with a good Jacket and moved the lnsldes to that. I could then read the disk and copy it to a good disk and 1 was running Mlnifiex. TSC offered their line editor EDIT and their assembler ASMB again, and this time I bought them on disk. TSC supplied printed source code. They did the same for their text processor called simply PR. I bought that as well. There was a veiy early compiler by Jack Hemmenway called Strubal (STRUctured BAslc Language). I can't say much more for It than to say lhat It worked. I typed in a 6 or 7 page program and the com- piler output overflowed memoiy! (I had 32K at the time). Lucldata itleased their 6800 Pascal Pcode compiler that was many times more efficient at code genera- tion... (Not a knock for Hemmen- way. Just progress happening). Shortly. TSC came along with their DEBUG program that let the user single step through a pro- gram, set break points, etc. I wondered at the time how I had ever debugged an assembler program without It. A year or two alter that I upgraded by buying a 6809 processor board and a lot more memory for the system. It origi- nally had 16K but I had previ- ously upgraded to 32K and now I went to 48K and then 56K (the maximum possible in that com- puter system). FLEX09 came along for the 6809 and went through a few upgrades. Then started the era of fast and furious development in the area of soft- ware. Lots of software was devel- oped by hobbyists and then sold through '68' Micro Journal and other sources. If 1 am correct, only Technical Systems Consult- ants and Percom. and perhaps SWTPC itself supplied soltware that was specifically written with profit in mind, at least early on. There was TSC's original Text Editor and Processor combination (updated to 6809 FLEX version), their Assembler, Lucldata's Pascal compiler, followed by a couple of "C compilers. The first, from a little company that will remain nameless, was a disaster. The author was not a great program- mer, but worse the company hadn't the vaguest Idea how to test software. There were so many bugs that a program com- piled under it seldom ran cor- rectly. Later Wlndrush Microsys- tems In England released a "C compiler by James McCosh that was as good as the earlier one was bad. McCosh is a good programmer, and Wlndrush knows how to test software and respond to bugs reported by users. I still use that "C" now and then when I fire up the 6809 system. It is. in fact a direct ancestor of Mlcroware's present OS9 68K "C* compiler. The folks from Mlcroware might say that it has come a long way since then, and I'd agree heartily to that. Other soltware that came from that era was the PL9 compiler. Stylo Text Editor. OmegaSoft Pascal. PIE text editor, a couple of implementations of FORTH, and much more. About the end of the 6809 era I received a COBOL compiler for evaluation. Just about eveiy new piece of software that came along (most or all of them were written in Assembler) showed a greater understanding of how to program a 6809. The compilers each generated better and smaller code than their predecessors. One peculiar language came along that was very good at doing things I don't normally do in my programs and terrible at doing things that I usually do. I let someone else review it and it got a fair write up. Then, along came IBM with their PC and PC -XT and the big switch began. Many of the 6809 users remained loyal and we wrote much of our own software, but it became evident that there would be no more commercial software available for these systems. I wrote PAT. an editor which was about a year's worth of spare time effort. I figure I've probably cleared about a dollar an hour for that effort, but it wasn't done to get rich. I really did it to prove that good software could be written In a higher level language (PL9). and that it could be sufficiently small to leave a reasonable edit buffer, and run fast enough to be adequately usable. I did end up coding about 3% of it in Assembler as 12 April 89 58 Merc Jounal ASMPROCs In order to speed up some veiy processor Intensive operations such as string searches. (I am at the point now with the 68000 systems when.- I was back about a year after the 6809 came along. Software Is limited but coming. I've managed to get enough together to do whatever I want to do. and I've filled in with some "homemade" utilities In assembler, all of which have been published here at various times). Somewhere along the era of all the new software arriving for the 6809. I began to slant the column more toward program- ming In higher level languages, and in reviewing the various offerings from the suppliers. I got In deep trouble with a few readers by saying once that I would choke (or worse) if I saw yet another COPY or DIRectoiy utility for the 6809 published. My comments by pure coincidence appeared in the column Just about the time a reader sent me a disk with some nice things on it, but including a directory utility. The writer took personal offense and didn't believe that my comments were not aimed at him specifically! I got a veiy nasty letter and wrote a couple letters of explanation which must have been transferred directly to the garbage, because there was never another re- sponse. (About then I thought seriously about quitting this column). With regard to the comment about Copy or Directory utilities, what I had in mind was that computers had to serve some useful purpose beyond allowing their users to create more utility programs for the operating sys- tem. One "hacker" (I don't use the term In any sort of derogatory sense) wrote and told me that he didn't intend ever to do anything useful with his computer! I've since adopted a similar attitude toward the more typical computer uses. I told someone that I would quit my Job if I ever had to work with or on Database software or a Spreadsheet. I do appreciate the value of such software and com- puter uses, but they are not for me. I am an Electrical Engineer by education, and I Immediately saw applications in the field of instrumentation and control of machines. It Is In that area where my interests lie. So much for histoiy except to note that the name of this file is NOTES 100. Since 1000 months Is 83 years and 4 months. I'm sure I won't have to shorten the name after the 999th column! As you may have noticed, I'm back into hacking again, writing new utilities for the 68000 and SK'DOS. Should any of you think that after ten or twelve years, it must be a total breeze for someone to write programs, let me tell you about last night. A couple of days ago I decided that a rewrite of my text processor JUST was long overdue. I first wanted to set it up to read a printer configuration file so that the text processor software cold be fixed and constant. The user could then prepare his own printer configuration file using only printable characters to represent the control sequences. That turned out to be a relatively easy one-evening plus, project. Then I decided to tiy to clean up some of the procedures, which had "grown" from the initial "keep It simple" concept to have fea- tures added and more features added. I was successful In getting It to correctly center and right Justify a line In double wide printing mode, and decided that It would be nice to fix it so It could Justify a line with double wide printing In it also. Having done similar things before. I made a careful backup of what I had that worked correctly, and set off to make the necessary changes. That was about 9:00 In the evening. Things went from not working to still not working to not working as well as before, to total garbage. At one point. I actually was printing out eveiy other letter In the line with a blank space between! Finally about 1:30 AM, I conceded defeat and threw away my new effort and reloaded my backup. In the four hours of effort. I had found and fixed one bug. so I added the fix to the original file. I guess I'll stare at a listing and let that project sit for a while. The time was not wasted, since I now have gotten back into the program and have a much better feel for how I had done it several years ago. A rewrite will be based on what I learned. My procedures were all much too Interdependent. I'd change one line and something that ought to be totally unrelated would no longer function prop- erly. When I was a lot younger and living in Chicago. I had a friend named Dick Nellsen. Dick always had a "shortcut" to get somewhere. Taking Dick's short- cut usually resulted In getting lost and taking twice as long to reach the destination. Among Dick's friends we began referring to doing something the hard way as a "Dick Nellsen Shortcut". Some- times a poorly thought out short- cut In tiying to write a program is Just that. i.e. doing It the hard way. When that happens. It is usually best to think about the problem a little more and then try 66 Mora Journal April '69 13 again. I had a few good Ideas and a couple of nights later I started again and made at least some progress. I now have a couple of new versions of JUST both work- ing and doing more than the old one. but they are lacking In some of the features that I want to add. I am obviously too much of the "start programming and get something working, and then clean it up later", school. I think some programmers go too far In the direction of getting It abso- lutely light on paper and expect- ing It to work on the first try. I've said that there is a middle ground that should result In minimum overall programming and debug time, and I observe that when 1 spend about equal amounts of time programming and debug- ging. I seem to get the Job done fastest. I maintain that If I were to spend three times as long writing the program. I would have about 1/3 as many bugs, so that the program plus debug time would be about 3 1/3 units. Equal program and debug time would add up to 2 time units. Similarly. 1/2 the programming time would result In 2 times as much debug time, so that the minimum is about wheie the program and debug time are equal. I realize that this is not something that can be derived mathematically nor proven rigor- ously with logic propositions, (nor for that matter anticipated before the debug is done) but it would seem to be verifiable by experi- ence. If I find that I spent too long debugging the last program, I spend more time writing the next, etc. Based on my experience of the other night. 1 conclude that I didn't understand the problem thoroughly enough to be able to program well. In many Instances when I write a program. I don't know exactly what It has to do until I take a good stab at it and try It. Then what it has to do sometimes becomes very obvious, and turns out to be significantly different from the original con- cept. This Is particularly true in the area of machine control and instrumentation, though I am sure it carries into other program- ming areas as well. If the prob- lem is REALLY spelled out well, that is a different case, but I don't believe that happens In practice nearly as often as the theorists would have us think. Let me give you an example of this. I have been working on a word processor again, so I'll use that. How do we read text from a file with ragged right margin, not necessarily with every line filled as full as possible, and print it out fully justified, i.e. each line the same length (by Inserting spaces in the line). A first ap- proach might be to read charac- ters from the Input file until you have LENGTH characters. If you come across a CR change It to a space. Now back up through the partial word at the end of the line (If it Is partial) and past the space that separates It from the previ- ous word. Of course you count the number of places backed up. and then add that many spaces within the line. You put a CR at the end of the line and then print It. (How you distribute the extra spaces Is another problem we'll assume you have solved). Now move the partial word lelt over from that line to the start of the next line and repeat the process until end of file. If you do that, you will, of course have run the whole file together into one big paragraph. Titles and headers are not part of a paragraph. Of course you must have a way to tell when a para- graph has started and only treat the text within a paragraph as described above. Some authors of text processors choose to make a paragraph one long line. That is, the editor is set up to handle formatting on the screen without any CRs In the text. That simpli- fies the identification of a para- graph, but the file is hard to list using an ordinary list utility. I chose to use a signal Inserted in the text to mark the start of a paragraph. ",p" all by Itself on a line signals that a paragraph starts on the next line. The end of a paragraph Is detected by two consecutive CRs (a blank line after the end of the paragraph) or another ,P signaling the start of a new paragraph. When the para- graph is ended, a partial line might remain in the line buffer. Of course it has to be printed too. and It might as well not have extra spaces In it since It will probably not be a full line any- way. Do you see what I am getting at? Unless you are an absolute genius (or at least a lot smarter than i). you won't think of all the conditions and exceptions without making a stab at a way to solve the problem and tiylng it to see what happens. Trying to write a program that is correct the first try is like trying to think up all the conditions and exceptions without trying what you have done so far to see If It works. More experienced programmers at least think of how a process like that described In the previous paragraph must start and stop, and perhaps they catch a few 14 Aprt'89 60 Micro Journal more of the exceptions, but to tiy to catch them all without "experi- menting" would seem to me to be a great waste of time. Epson "Enhancement"? A few months ago I saw an add in Computer Shopper for a new ROM to upgrade an Epson MX-80 among other moaels. 1 have a veiy old one that still works like new. I had upgraded It a long time ago with GRAFTRAX. the Epson upgrade, but of course it still didn't have "NLQ" mode. The ad for an upgrade kit called "Dots-Perfect" indicated that the upgrade would add letter quality and control of printing mode via the Online. Formfeed, and line- feed buttons, more or less like the newer Epsons. I sent for the upgrade and It has arrived. 1 followed the In- structions and it all worked pretty much as advertised on the first try. but the "NLQ" looks pretty bad. In fact not as good as just printing in Emphasized mode as 1 had been doing with that printer ever since 1 bought It. Certainly It doesn't compare with the LX-800 in either of its NLQ modes In which the characters look veiy full and smooth even under a magnifying glass. Vertical lines are quite obviously zigzag. It appears that the NLQ is achieved by simply shilling the carriage and overstriking, The "upgrade" cost $69. In my opinion. $20 would be more in line with what It does, primarily based on the poor NLQ. The Epson MX-80 is capable of the same graphics resolution as the new LX-800. so there is no reason that a better NLQ font couldn't be done, other than, perhaps, lack of ROM space for the font character definitions. The upgrade does allow selection of fonts and print styles using the On Line. Formfeed and Linefeed buttons as advertised, and those are also selectable via soltware compatible with the latest Epson LX-800. It even has a "small print" mode that Is simply superscript mode, condensed mode, and half line feed all combined. That's nothing you couldn't do on any Epson with true half size superscripts and subscripts under software control. I suppose if 1 didn't know how to write or configure word processing software. I'd be more Impressed. In would think you'd be better off spending that $70 as a significant fraction of the cost of a new Epson LX-800 or even a Citizen 180-D which can be had for around $170. It is a nice little printer but Its NLQ is a little thin compared to the LX-800. CRASH! 1 just found a way to crash my editor PAT. I'll mention It so you can avoid it if you are a PAT user. 1 accidentally typed ESC A L. Of course ESC 65 A L would set the line length to 65. but with no parameter the line length was set to 0. 1 didn't notice that, and tried to format a paragraph to zero width. The "BUSY" notice came on. and stayed there until 1 ran out of memory, at which point there was an error message and an exit to the monitor. Of course all my edit was lost. I'll trap the no-parameter case and make it not change the current line length. Meanwhile, PAT users be warned. FOR THOSE WHO 68 MICRO JOURNAL TM 68 Mkro Journal April SB 15 Forth A Tutorial Series By: R D Lurie 9 Linda Street Leominister, MA 01453 OF-Link from Sardis Technologies At last, a way to combine FLEX and OS-9 Level II! Sardis Tech- nologies has a real winner, here. If you have been reluctant to switch to the Color Computer and OS-9 because of a heavy Invest- ment In FLEX software. OF-Link could be the answer to your prayers. Now you can have all of the advantages of OS-9 without losing that hoard of FLEX soft- ware you have accumulated over the years. I don't want to spend any time going over the virtues of OS-9 vis- a-vis FLEX, as I am sure that you have heard enough to be tired of that argument. Rather. I want to describe my reaction to running FLEX programs In this new environment. First, let me say that not every program will work. 1 could not find any software In my admit- tedly small Inventory of "regular" FLEX goodies which would not run completely as expected under OF-Llnk. However, the copy of STYLOGRAPH which I use is the one especially tailored to the CoCo FLEX sold by South East Media, and 1 was not able to finagle the patches to make It run under OF-Link. Similar troubles would arise with any software which does not use the standard FLEX hooks. The instructions which come with OF-Llnk give help with some popular software which has a problem, but this Is still for "regular" FLEX programs. There are stringent hardware requirements, also. You must have a CoCo3 with 512K of RAM and two disk drives: otherwise, you cannot Install the system. However, once OF-Link has been Installed, you then can manage with only one drive, although it will be inconvenient. The reason that you MUST have two drives is that you must Install D. p. Johnson's SDISK3 before you install OF-Link. Contact DP. Johnson at 7655 SW Cedarcrest St.. Portland. Oregon 97223. 1503) 244-8152. See D.P. Johnson's ad on page 59. If you have a DMC "no halt" disk controller from Sardis. you have no problems, since you already have SDISK3. If. like me, you have a Tandy, Dlsto. or other controller, then you must separately obtain and Install SDISK3. I followed this route with both the original Tandy controller and the Disto SCII. and built a working system on SSDD35 drives, so you can. loo. Just follow the excellent direc- tions exactly and literally, and don't free-lance! Once you have a working system, you can use modpatch or the public domain dmode to change to any number of tracks you wish. I easily converted from SSDD35 to SSDD80 by way of dmode. Unfor- tunately. I could not get any sort of double-sided drives to work: this may have been the result of conflicts with my particular hardware. Also. I could not get more than 2 physical drives to work, even though the manual gave me the impression that I could. However. I had no problem with as many as 4 logical drives, so this may not be a practical limitation. There is one "gotcha" which comes lrom the peculiarities of OS-9 record locking, and not from OF-Llnk. In any case, each active FLEX device window must have its own virtual system file. In other words. If you have 3 win- dows running with FF9. you must have 3 separate virtual system files, one to support each device window. I solved this "problem" by having all but one of the virtual windows contain only FF9 and the absolute minimum of other files, such as FLEX. COR, etc. Only one of the virtual system files contains a full complement of FLEX utilities. This saves physical disk space. By naming the 3 files (remember that these are really OS-9 files In disguise) as fvl. fv2. and fv3. 1 can call each one as I setup /wl . /w2. and /w3. There is a way around this problem of record locking and it is described In the manual: but the technique I described above is the one rec- ommended as being a lot safer and saner! 16 April 89 68 Micro Journal Be veiy careful when you start sharing FLEX files, since It Is veiy easy to mess up the links between sectors and the free chain. If you should have this sort of accident, you are In for a real hassle recov- ering from It. Be warned!!! Utilities, etc. Several very useful utilities are Included with OF-Llnk which make It easy to Install and use. The ones I used the most, and I expect everybody will have about the same experience, are dpoke and newcrc. Dpoke lets you change an OS-9 file and newcrc updates the crc without the ne- cessity of explicitly moving that file lrom and back to the disk. There Is also a simple monitor which will help In debugging FLEX programs. Unfortunately, this was not fully operational on the review-copy of OF-Llnk which I received, so I can't say more than that It certainly looks good. FF9 under OF-Llnk So much for the preliminaries, now for the Important stuff! ! I think that the first question answered should be. "Why bother?" Well, that Is the easiest question to answer. With OF- Llnk, you gain: 1. 80-column screen 2. windows 3. graphics 4. color Ok. if OF-Llnk Is so good, what do you lose? The only lost feature of FF9 that I have been able to Identify, so far. Is the direct Interface to the machine. How- ever, this Is a characteristic of OS-9. and not solely of OF-Llnk. If you use your FLEX software for machine control, then you must either choose between the great conveniences of OF-Llnk and the necessity for writing OS-9 device drivers or go back to your old system. No. I have not yet written a successful device driver from scratch — something I have to learn. It may be possible to use the OS-9 call F$MAPBLK, as Is done in FOKTH09. but I have not worked that out. either. The 80-column screen Is a bigger Improvement over the 5l-column screen than I had ever expected. As of now. 1 have put aside my SS50 and separate terminal In favor of the CoCo3. This. In my not -so-humble opinion, alone Is worth the cost of the software! I may resurrect the SS50 at some later date, but I can't Imagine when or why that would be. Windows let me have several copies of FF9 going at once, so that I can flick from one program to another Just by pressing the <CLEAR> key. Again. I am sure that you have been Inundated with praise for this feature of OS- 9. so I won't spend time on the hard sell. By graphics. I don't mean Just the crude cartoons to the excellent art that Is possible, but also the full business and engineering-style graphics that can be generated by the lowly EMIT . Color Is available, as expected. The options and limitations nor- mally available under RSDOS or OS-9 are all there. If you have the hardware, you can access them. Melding FF9 to OF-Llnk OF-Llnk gives you the option of using the normal hardware FLEX disks or the virtual FLEX disks which actually are resident as OS-9 files; FLEX Just thinks that they are the usual disk. For use under OF-Llnk. the virtual disk Is to be preferred. This Is particularly true If you have a hard disk; you can easily partition It Into FLEX and OS-9 segments. OS-9 would think that there were only OS-9 flies pres- ent, and all of the OS-9 file utili- ties would work. However, don't try to use a FLEX file utility on one of the OS-9 virtual FLEX files; this Is fatal! You can boot FLEX from the virtual disk on the hard disk, so you don't have to keep a floppy around Just for that. The virtual FLEX disk has 64 sectors per track, so you must modify the FF9 disk parameters accordingly. Assuming that you have formatted your virtual FLEX disk to have 24 tracks (to produce 368 screens), you would modify FF9 by entering: 24 * TRK/DRV >BO0Y ! 64 ' SEC/TRK >BODY ! 1 * BASCTRK >BODY ! According to my calculations, a single virtual disk could have as much as 2032 screens. If the disk is large enough. Of course, a physical disk can have more than one virtual disk! Summary Do I like OF-Llnk? You bet I do!! Do I recommend OF-Llnk? You bet I do!! Sources: OF-Link, $49.00 (+$5.00 SAH) Sardis Technologies. 2261 East 1 llh. Ave., Vancouver, B.C.. Canada V5N 1Z7 SD1SK3. $29.95 (+$1.75 SAH) D.P Johnson. 7655 S.W. Ccdarcresl St., Portland. OR 97223 FORTH09 ver. 1.02 Dan Johnson has addressed my complaint about FORTH09 shadow screens which I expressed In my review last time. He now allows a command line option *-n" which eliminates any call to shadow screens, so you are not forced to provide for them. Thanks, Dan, I appreciate that! 68 Micro Journal Apnfl '89 17 FORTH09 has some quirks which take a bit of getting used to. These quirks are forced by OS-9 and are not something to be changed on a whim. The most obvious difference with a FOKFH operating under OS-9. as opposed to another environment. Is that ALL absolute addresses are Illegal. Therefore, you cannot legally use Jump vector tables. My observation that the form shown In Screen #3 of my last column would work resulted from a peculiar set of circumstances. This win only work If you compile the screen each time you use It; you cannot run from a SAVEd or a SAVESYSed binary file, since the absolute addresses would then be wrong! Dan also said that this would not work under Level I. so I did not bother to try it. The proper way to do vectored execution in FORTH09 is through the MAYBE ...THENAGAJN... MAYBE NOT construct (1 love Dan's choice of names). A work- able definition is : EXPERIMENT ( n - ) MAYBE EXO 1 EX1 2 EX2 1 EX3 MAYBE NOT One can easily make OS-9 calls directly from FORIH09. and Dan has provided most of the words one could ever expect to use. Furthermore, he has provided enough information so that defini- tions can be quickly written to cover the obscure calls that he did not bother with. Among these calls are F$ALLRAM , F$MAPBLK . and F$CLRBLK . which can be used to access directly any RAM within reach of the MMU. These words can be used to access any RAM block, so that FORIH09 is limited only by the available memory. I wonder if this kind of thing could be used to control a latch which bank- switched additional chunks of RAM, so that the CoCo3 could go even beyond the present 512K? I don't plan to try it. but I am cuilous. Another way to access OS-9 Is illustrated In the definitions contained In Screens 46-48. Actually, several features of FORIH09 are illustrated in these three screens. This definition will accomplish the same result as the set of dellnl- tlons and commands I had shown in Screen #3. Remember, the method shown in Screen #3 is still valid for other environments, Just not for OS-9. OS-9 «ccet» from FORTH09 Here is another example of how much I still have to learn about OS-9. I had read all the way through (tough going) the manual which came with my CoCo OS-9 Level D, but the significance of the few lines on manipulating the MMU did not register through all of the other noise. However, the documentation on FORI"H09 finally woke me up. 1 won't spend much time on the first two screens, since they were discussed In some detail in Sep- tember. 1987. The only really Important points, here, are in the use of VAR , instead of VARIABLE . to define ARGC (line 3 of screen 46). Line 5 of screen 46 and line 7 of screen 48 are identical in showing how to place a value into a VAR . There is no FORFH need to use the loading command twice, but these screens come from my new FORrH09 utility file, so that screen 46 can be used with a reasonable default value in ARGC . when It is loaded wilhout screen 48. On the other hand. ARGC is loaded in the definition of HARD Just in case I might have used ARGC somewhere else, so that the value got changed. This is simply cheap Insurance against an annoying crash! Screen 48 shows the use of CR" and SHELL . two words which are unique to FORI"H09. CR" places the address of the CR-teiminated string on the Data Stack. The string literal must end with a " . SHELL then uses this address to pass Ihe string along to the OS-9 shell for execution and then return control to FORIU09. As you can see, this causes the cuirent date and time to be printed along with the screen listing. Hardware control through FORTH09 under Level n I may well be the only one in town who did not realize that, with FORTH09. I could directly access the hardware located at physical addresses $FFOO-$FFFF. In other words, this command line: fload forth -n 148k gives me 8K. starting at logical $COOO to play with through F$ALLRAM . etc. and lets me write directly to the hardware 1/ O. such as the casette motor relay, or anything else located at the physical addresses $EOOO- $FFFF. I tried to do this hardware access from BASIC09, but I could not find a way to make it work in Level II. Undoubtedly, it must be illegal, since It Is so desirable and easy! In any case, the whole thing is perfectly straight forward in FORTH09. For example: : MOTORON 4 6S313 C! ; : MOTOROFF 52 65313 C! ; Speaking of timing tests. I hope to include some benchmark results with FORIH09 in the near future. I am impressed with the speed of most of the operations: most of the ones that I have checked to date are faster than can be ex- plained by the simple increase in processor speed! 18 April '89 68 Micro Journal Screen 46 \ ARGC HELP-SCREEN \ A part Of PARSE -COMMAND- LI(E VAR ARGC \ expected i 2 TO ARGC I HELP-SCREEN ( - ) CR CR Two input parameters are required 1 . f irat-acr#" CR 2. laat-scr»" CR CR In no caae, can the value of " firat-acr»" ASCII * EMIT be greater than the value of * laat-acr»." ASCII * EMIT October 22, 1988 19:19 :30 RDL102088 10 11 12 »r of input parameters 13 14 15 16 \ RDL062587 17 18 1:" CR 19 110 111 ASCII "* EMIT 112 CR 113 ASCII " EMIT 114 CR ; 115 Screen 47 \ ARGV-OK PARSE-03«AND-LINE \ Requires previous screen I nl n2 - ) < \ first-scrt cannot be greater than last-serf HEIP-SCREEN ARGV-OK 2DUP SWAP IF ABORT THEN ; PARSE-COMMAND -LINE DEPTH ARGC < IF HELP-SCREEN ABORT ELSE ARGV-OK THEN ; { - ) October 22, 1988 19:19:34 RDL102088 |0 II 12 \ RDL1 02088 13 \ duplicate the input parameters I 4 15 16 17 18 19 \ RDL1 02088 110 \ minimum acgumant count? \ wrong number of arguments \ start over \ verify argument validity 111 112 113 114 115 Screen 48 \ SHOW SECONDARY DECIMAL 46 47 THRO : HARD ( first-scrt laat-scrt - ) 2 TO ARGC PARSE-COMAND-LINE >PRINTER 14 SWAP DO CR ." Screen * I 30 SPACES CR" DATE T" SHELL I LIST CR CR LOOP >SCREEN ; October 22, 1988 19:19: RDL102288 \ RDL102288 16 37 10 II 12 13 14 |S 17 18 19 110 111 112 113 114 115 FOR THOSE WH 91 X 1.1 I) TO KXOW 68 MICRO JOURNAL TM 68 Mcro Journal April -80 16 Moil Of you will rononber Bob from hii tciic* of tellers onXBASIC If you lite il or wtni more, tel Bob or us toow We winl U) give yiw - mhal y>n waul! The Mathematical Design of Digital Control Circuits By: R. Jones Microma Reseirch Corp. 33383 Lynn Ave.. AMwuford. B.C. Cuud* V2S IB2 Copyrighted © by R Jones & CPI No TEST last time, ergo no solutions either! Mile 20 - heading for Mile 2 1 . Last time, we got Introduced to Boolean matrices, and learned how to interpret a matrix to let us trace paths through Its corresponding circuit, and we also met Pivotal Condensation, a technique which allowed us to derive a transmission function from a Boolean matrix. Now we'll learn the second oflhe two techniques I promised, which has certain advantages and also some disadvantages compared to the first. LAPLACE 8 DEVELOPMENT Our second method, known as Laplace's Development. Involves the reduction of our matrix-a to a final 3-by- 3. or several 3-by-3s. which, for the benefit of those of you who are familiar with matrix and determinant theory in advanced algebra, will then be evaluated In the same way as a third-order determinant. Except that all "- " signs will be replaced by "♦" signs, as there Is no such thing as a minus sign In logic theoiy. at least not at the level at which we're using It. For those of you who've never even HEARD of a determinant. I'll explain everything that it's necessary for you to know in order to work with a third-order Boolean determinant, which is nothing more terrifying than an end-result 3-by-3 matrix to all Intents and puiposes. For our first example, we'll begin with our matrix-a of Diagram 102. from which, as you'll see. we'll derive the function f l2 In no time at all. For 4-by-4 matrices. Laplace's Development gives a very quick solution indeed, but tends to become Increasingly cumbersome as we proceed to matrices of higher order, for which I'd recommend using Pivotal Condensation. DERIVING A DETERMINANT FROM A FOURTH-ORDER MATRIX You'll recall that In our first method we eliminated all nodes, commencing with the highest numbered node, until we ended up with only those of interest, namely nodes 1 and 2. Strange as it may seem, our new technique seems to dojust the opposite. That is. as were Interested In the function f |r WE BEGIN BY ELIMINATING ROW- 1 AND COLUMN-2 ENTIRELY from matilx-a. Actually we'll just loop them In red. but. for all practical puiposes we've scratched them out. Bang! They don't exist any more! How about that for a quick vanishing act? Then we write down all that remains, as In Diagram 104. and we have our third-order determinant. Note that there's no row- 1 or column-2! Because of this fact, our determinant Is strictly for f 12 . EVALUATING A THIRD-ORDER DETERMINANT I 5 fc t 2 51 £ b a 'A: Diagram 104 20 Apnl 89 68 Micro Journal The best way for a newcomer to "evaluate" such a determinant Is to Imagine a duplicate copy of 104 written immediately to Us right (why don*t you do just that?) , and then to follow my explanation step-by -step. The dotted lines are merely to show how a more experienced "reader" would achieve the same result without the need of the duplicate copy alongside. What we have to do Is to commence In the top-left corner, and slant down diagonally to the right until we reach the bottom row, multiplying together (that Is. ANDlng) the three elements encountered along the way. Our first effort produces 0.1.1 (which I haven't dotted, for clarity's sake). Then we move along the top row to the next element, and again we'll slant down to give us "bdc*. which we'll add (that Is. we'll OR) to what we've already got, so now we've got "0. 1. 1 + bdc". And once more, along to the top-right element, and slant down to give another block of three literals, viz. "eda". Newcomers will now slide along lhe top row and Into the duplicate matrix, doing exactly the same thing hut slanting downwards TO THE LEFT this time, to add yet another three terms to our expression. Once again. I've omitted the dots through the main diagonal, otherwise we wouldn't be able to see anything for dots! To give a complete picture. Ill mention that In normal algebraic operations, the second set would be SUBTRACTED, not ADDED, but everything Is ADDED In Boolean algebra. So now we have 0. 1. 1 + bdc + eda + O.d.d + b.a. 1 + e. 1 .c = + bdc + eda +0 + ab + ce = ab + ce + aed + bed and we're not at all su rprlsed to find that this Is the same as that found by path-tracing. In actual fact, of course, as we write down the original terms from the determinant, we'd omit completely any which contained a "0". and we'd also omit the " l"s from any term which contained one. This would immediately give us the second of the three expressions above, but without the O-terms! That wasn't so bad after all. was It? So now you know about Boolean determinants! FIFTH-ORDER BOOLEAN MATRICES Wow! Just look at that heading! Enough to set you running for cover, isn't It? So let's Just sneak up on it. and see if we can't lick this one too! I mentioned that this method gets rougher as the size of the original matrlx-a grows, but 1 think we're brave enough to tackle a 5-by-5. If not. there's always the helicopter back along the trail! All right then, we'll try our hand at the 5 x 5 of Diagram 105a. 1 2 3 *♦ S 1 3 1* S 1 r l a c $ 2 b e c 2 1 b e c t 3 a 1 d J 3 a b 1 d d F «*i, c d 1 d' ft c e d 1 J *l b' a' d' 1 5 kj c <i 4 i. (a) (b) bee bee bee a d t d' + c 1 d a + b' 1 d a a' d* 1 a' d' I 6 1 J fc) Diagram 105 68 Mcro Journal April 69 21 DERIVING A FOURTH-ORDER DETERMINANT FROM A FIFTH-ORDER MATRIX First of all. we'll establish that It's f,, we're Interested in, by eliminating entirely from 105a both row-1 and column-2 to give us the fourth-order determinant of 105b. Observe the convention of Indicating a matrix with delimiters to the right and left, with a curved top and bottom, whereas a determinant is delimited with simple straight lines. So ... curved lines mean that the matrix, even a reduced one produced by pivotal condensation, precisely describes Its original network. Straight lines indicate that at least one column and row have been COMPLETELY eliminated, and not absorbed into a new matrix, as happens with our first technique. It's possible, with experience, to read fourth-order determinants quite accurately, but as we're essentially lazy at heart we'll reduce ours to the slmpler-to-handle third-order variety! NOW FOR OUR THIRD-ORDER DETERMINANTS We 11 do this scientifically, by selecting ANY SINGLE ROW OR COLUMN for elimination, preferably one with as many 0-entries as possible. The reason for this will become clearer as we proceed! Here we have a choice of either row-2 or column- 1 (there is no row-1). so let's select column- 1 and loop it in red. We're now going to work our way down the selected column, one element at a time, and omit the row in which this element occurs. This element will then be multiplied by (ANDed with) the 3-by-3 determinant which remains. In case this all sounds a little confusing, and I've no doubt that it does, just tag along with me as I actually do this. The first element in our selected column isO, which, as I've just mentioned, has to be multiplied by some determinant. However, as times anything is equal to 0. we've really got nothing to do here. See now why we love to have as many Os as possible? Anyway, moving down to the next element, "a", In row-3, we must mentally eliminate row-3 from our 4-by-4 determinant of 105b (keeping In mind that column- 1 is already gone. In theoiy anyway), and create a little 3- by-3. which we've written in 105c, with an "a" to the left, as this is the element Involved In the deleted column- 1. Then down to the next element In this column, namely "c\ eliminate row-4 in our minds, and write THIS remaining 3-by-3 with a "c" to its left, and a "+" sign between the two third-order determinants to indicate that they're going to be ADDED together (that is. ORed). And finally to element b' In column- 1 of 105b which gives us the end determinant with a b" to its left. Now re-read my original instructions, and you should be able to follow them this time! MINOR DETERMINANTS These final determinants, not being the full determinant of 105b, are Individually called "minors". And now comes the tedious task of evaluating these minors, as we did with the 3-by-3 of our earlier example, and multiplying each one by whatever is outside It. This is the time when I wish that the column had more Os tn it! It's got to be done, however, so let's get on with it, first of all setting out the expression In full. a(b + ed'a" + cd'd + bd'd' + ed + a'c) + c[bd + ea'a' + cd' + bd'a" + e + cda') + b'(bdd' + ea'd + c + ba' + ed' + cdd) Now comes the horrendous task of multiplying each line of this complex expression by the literal outside each set of parens. But first note that inside each set of parens a single literal occurs, and we can therefore eliminate any term which contains this literal. At the same time, any term which contains the complement of the literal OUTSIDE the parens may be eliminated, because when multiplied together we'll have the situation xx'y, and as we know from our basic Boolean algebra (we do remember, don't we?), xx' = 0. thereby cancelling out this term. We've also got several occurrences of this sort INSIDE the parens! So let's do this reduction first, to give us a(b + ed) + c(bd + cd' + a'bd' + e + a'cd) + b'ta'de + c + d'e + cd) which then multiplies out to ab + ade + bed + (cd' + a'bed') + ce + a'cd + a'b'de + (b'c + bed) + b'd'e 22 April '69 68 Micro Journal I've highlighted Iwo sets of leims Inside parens, where we see lhal as the smaller term Is contained In lis entirety within the larger term, the larger term can disappear. cd\oo en II IO CO Ol II 10 1 1 1 1 1 I 1 1 1 I cd Xoe PI II oo ox II to 1 1 1 1 1 1 1 1 1 I 1 1 1 1 Diagram 106 The simplest way to finish off the simplification Is lo draw up a 5-varlable K-map as in Diagram 106. and read oul the expression as f 12 = ab + c + b'e If this all seems like a lot of work from beginning to end. Just imagine what would have been Involved If we'd started off with a 6-by-6 matrix, this would have reduced to a 5-by-5 determinant, which would have reduced lo live 4-by-4 first minors. EACH OF WHICH would then have reduced to four 3-by-3 second minors, for a grand total of twenty 3-by-3 minors to evaluate and simplify. Can you Imagine this? Not lhal it would be EASY by the Pivotal Condensation method, but I think it would be far. far easier than this!! MULTIPLE-OUTPUT NETWORKS AND BOOLEAN MATRICES Now that we're rolling along nicely and smoothly, lets recall my earlier bit of info lo the effect lhal Boolean matrices can keep track of several outputs simultaneously. We'll even complicate things a little, by making our outputs share part of the same network! la 4- - fo) H II - s - — 1 2 3 h S 1 2 1 '{0 a c N I r i 2 1 a b 2 1 3 a 1 e 3 ce a U q b I d 4 o 4- cd b 5 c c ' »j ce a' I de + cd* 1 b de I fc> 2 3 I b(a 4- cd) eCc + ad)^ k(a + cd) I a' 4- bde eCc ••- ad) a + bde I (J) Diagram 107 68 Mcro Journal April '89 23 The circuit of Diagram 107a is a three-terminal network, with power input at node-1. two outputs (Zl and Z2] at nodes 2 and 3. and two intermediate nodes, 4 and 5. Its Boolean matrix is shown In 107b. so we'll begin a preliminary, partial analysis by eliminating nodes 4 and 5 from matrlx-a. leaving only the terminal nodes 1. 2 and 3. First, let's eliminate node-5 by Pivotal Condensation, to produce malrlx-b of 107c. and then node-4 to give matrix-c of 105d. Note that these are MATRICES, not DETERMINANT'S, as we haven't actually eliminated anything, but merely CONDENSED our original matrix! In our final third-order matrlx-c. element a l3 details the paths from node- 1 to node-2 (but NOT via node-3), a 13 the paths *from node- 1 to node-3 (but NOT via node- 2). and a u the paths from node-2 to node-3 (but NOT via node-1). EVALUATING INDIVIDUAL PATHS If we wish to lind the expression which describes the complete transmission-function f between nodes 1 and 2. we must eliminate node-3 by Pivotal Condensation, to produce f 12 = b(a + cd) + e(c + ad)(a" + bde) = ab + bed ♦ (ce + ade)(a' + bde) = ab + bed + a'ce + bede + abde = ab + bed + a'ce Note that thellnal term supplies power to Zl via node-3! In similar vein, in order to find f 13 we'd eliminate node- 2 from 107d Instead, and if. for some reason, we were interested in knowing under what conditions nodes 2 and 3 would be connected together, we'd eliminate node- 1 Instead. The finally cleaned up functions are given below, which you could check for yourselves by actually doing each evaluation. f 13 = ce + ade + a "bed f M = a* + bee + bde A systematic method of node-elimination such as this gives us the transmission-function between any pair of nodes desired, automatically taking into account ALL possible paths through the network. Our other alternative Is to actually trace out the paths in the circuit-diagram, always with the possibility that one or more paths would be overlooked. You'll have noticed that the Laws of Boolean Algebra really come into their own during this reduction process, as the work is made much simpler if the elements are simplified as we go along. For example, when reducing from 107c to 107d. the element c l3 starts out as ce + (a + cd)(de) and Instead of just inserting this complex little beast Into element c 13 , we multiply out and simplify first. = ce + ade + cde (cde vanishes because of ce) to give e(c + ad) CONNECTION MATRICES AND OUTPUT MATRICES A matrix such as that of 107b. which contains only one literal per element is known as a "primitive connection matrix", while matrices such as 107c or 107d. which contain more than one literal In any element, are called simply "connection matrices". There's a third type of matrix, known as an "output matrix", which charts, not the physical connections in the network, but the FUNCTIONS describing the transmissions between nodes. As an example of an output matrix, we could draw up a 3-by-3 for our present circuit, and insert in each element the three transmission-functions Jusl evaluated individually. Such a matrix would, in each element, give the complete transmission-function between the nodes forming Its co-ordinates IRRESPECTIVE OF WHEIHER SUCH A PATH GOES THROUGH ANOTHERNODE OR NOT. whereas a normal connection matrix describes the connections between these nodes BUT NOT VIA ANY OTHER NODE WHOSE NUMBER APPEARS IN THE MATRIX CO-ORDINATES. The output matrix for the network of 107a is constructed by Inserting the three transmission functions already worked out. as shown next. 24 April '89 M Uiao Journal 1 2 3 I b(a + cd) 4- ace e« +■ ade +■ e/bcd 2 3 b(a ♦■ cd) + cfce ce + ode +- abed > cr** bcfc 4 d) a 4. b*(c + d) Diagram 108 The element d 14 now gives, not the PATH between nodes I and 2. but the CONDITIONS [ALL the conditions) under which transmission will occur between these nodes. Take time out to think this over, so you're reasonably clear about the distinction between the matrices of 107d and 108. even though both of them belong to the SAME network. That's got the ANALYSIS part of this thing out of the way. so now for SIMPLIFICATION. This wll always be our line of attack, first analysis, then simplification, and then the big one - synthesis. NETWORK SIMPLIFICATION BY BOOLEAN MATRICES Are we all ready to move on in our study of Boolean matrices? If so. we'll Illustrate the versatility of our new tool by eliminating, by Pivotal Condensation, the non-terminal nodes 3. 4 and 5 from the network of Diagram 109. thus deriving the FUNCTION f l2 in Its simplest form. -n a 3 Hr J * j£. -IF e a 4* a* Diagram 109 In our previous examples, no simplification was possible, as was obvious from the final expressions derived. The matrix for circuit 109 and the subsequent step-by-step elimination of its non-terminal nodes are set out in Diagram 110. The finally simplified expression for f ]2 shows that the original network can be completely replaced by a single NO-contact on relay-D, as this is all that's necessary to specify the transmission-function between nodes 1 and 2. Without Boolean matrix techniques, we'd never be able to figure this out from the network, as it's quite a tough Job even to figure out Its Boolean expression directly! 1 1 2 3 a <* r ■> 1 i i 3 ( do a <r ac 2 1 e a+ d 2 ad 1 e + a -»-d 3 a 1 d 3 a I d k e d \ d t* a'e c+ a +-d d I 5 a' a + d (4 e* 1 fc> i 2 f 1 1 a'd 2 a'd » 3 a +de' d (< f,2 -a'd" + J(a + Jd) >) » a'd + ad" + de 3 a ♦ de' d ' J =» d(a + a + O - dO+c' Diagram 110 68 Mora Journal April '89 25 Suppose, in addition, lhat there were an additional output being taken off at node-4 to drive some other device, and we desired to know the transmission-function f M . as well as f 12 Now we're getting a little complicated, to say the least, but let's go along with this assumption. Node-5 has already been eliminated in 1 10b. so we'll now proceed lo eliminate nodes 2 and 3 (so only 1 and 4 will be left), commencing with node-3 to produce the matrix of Diagram 1 11a. from which we'll work out the function f ending up with **a'e' + d". 1 2 V 1 C 1 o'd o'e'+od 2 o'd I e *■ a + d 4 <aV -»- <aci e 4- a 4- d 1 } mJ 2 , ii IK If r *- If Diagram 1 1 1 r l«* f ltt , - aV ♦ ad 4- a'd(e+a+cO s» a'«' ■+- ad •+ a'de 4- a'd « a V -v a J 4- a'd = aV-4dCaW) = oV 4- d The COMBINED dual-output circuit of the two simplilled functions is shown in 111b. But let's not forget.however. that MATRIX CONTRACTION YIELDS THE SIMPLEST FUNCTION, AND NOT NECESSARILY THE SIMPLEST CIRCUIT. If you don't believe me. Just go back and look at the FUNCTION derived from matrix contraction in diagram 104, and the circuit diagram which we could reasonably expect loconslructfromit. The simplest CIRCUIT is. of course, the original one of 101 . but the simplest FUNCTION describing it is that which describes the equivalent circuit of Diagram 100b. All clear? MORE CHIT-CHAT While matrix contraction yields the simplest transmission-function, matrix EXPANSION, which we'll be studying shortly, gives us a simplified circuit-diagram, if one exists. Don't be confused by the fact lhat matrix contraction has Just given us a MUCH SIMPLER network, as it often happens that the simplest function is co- incidenlally the simplest circuit also. If this DOES occur, remember that it's Just an accidental side-benefit derived from the true purpose of contraction, and you'll never be really sure whether it's truly the simplest circuit. Unless, of course, you end up with Just a few simple contacts, and you can tell at a glance that 11 can't be reduced further. And now it's TEST-time again for all you lucky people, so let's have a go at the following, but think carefully before you answer the question posed in Problem 2! TEST FIFTEEN 1. Evaluate f l2 by means of Laplace's Development, and draw the final circuit. \ -& -th h£ -L*/ ■I*. tV -#4- — #£- 26 April '89 68 Micro Journal 2. Nodes 1. 2 and 3 are terminal nodes. Eliminate nodes 4 and 5 by Plvolal Condensation, and draw the final circuit diagram. Is this a simpler circuit? 3. Evaluate f l2 by means of Pivotal Condensation and draw the final circuit. CAMP TIME AGAIN Come on now. and admit that although this has been a VERY complicated few miles, this new stuff is nowhere near as terrifying as our encounter with the Laws-of- Boolean Swamp. And Just think what It can do for your design skills? Anyway, while you're working at these three problems. I'm debating whether to lead you into another of those word-logic problems, similar to that Involved in Uncle Fred's experiences with 1-asku of the M'bulyan tribe. I'll let you know when we're ready to move off once more. ... End of Mile 20. and all gathered round marker Mlle-21. FOR THOSE WH 68 MICRO JOURNAL TM 66 Micro Journal April '89 27 Mac-Watch Kqt Those 83 1 ■ i 68 MJ ^^JhHHLX Tne Macintosh* Section Reserved as A pllei«® flfarp wfannr? ft fa <a nn fgi Bn ft a And ours. This and That Reviews of MaxPage and QulckLetter By James E. Law 1806 Rock Bluff Rd. Hlxson TN 37343 This month I want to briefly review several new software packages starting with MaxPage which its distributor. Applied System and Technologies, de- scribes as 'the affoidable page makeup program." A Review of MuF\ge A logical part of us knows that you don't get something for nothing but In spite of this, we search for that rare deal where we get a treasure for pennies. Occasionally it works out but most of the time you get what you pay for. When I received a piece of software which advertised page layout capabilities for a suggested retail of $89. my curi- osity was peaked. Could it really deliver? Let's find out. When you first open MaxPage. a blank page is presented with a set of menu bars. The page may be in landscape or portrait orientation. You may activate a grid (maximum size of 17" by 17") for ease in locating text or other items on the page. Oddly, the default settings are for the grid to print out when you print your page, but it may be ordered not to print. It is specially marked to facilitate the layout of bifolds and trifolds. There is only one view of the page - full size - and no page preview is provided. This is an unforgivable onunlssion for a page layout program! Working With Text To work with text, you first create a text block by clicking and dragging the cursor. Each text block has a drag handle for use in ie sizing the block. You can move text blocks around and position them anywhere you like on a page. There can be up to 75 text blocks per page. Upon sizing and positioning a text block, you can enter and edit text in the usual fashion. Multiple fonts and sizes within a text block are supported. Text can be typed into text blocks or imported horn other applications. Text can be aligned left, right, or center but may not be flush Justified. No tabs are available. MaxPage allows considerable control over line spacing. \bu can adjust the leading above or below the selected line or have leading automatically adjusted based on a number of options. If you enter or Import more text than a text block can hold, it stacks below the visible area of the text block. The hidden portion of the text can be seen by re sizing the block. This excess portion of the text can also be made to wrap over to a different text block on the same or a different page. This is a manual process that might work okay for a simple one or two page document but would not be practical for a more elaborate pioject. Blocks aie also used to accept and position PICT (draw-type) or PNTG (bit mapped) images. The image is automatically scaled to fill the active picture block. If you re size the block, the image is ie sized. MaxPage takes a highly unusual ap- proach to handling (mages in that the image is referenced to its location on your drive and is not leally Imported into MaxPage. Every time the page 28 April '6 68 Mcro Journal is re drawn, the image is re-imported. This means that the image file must stay on the same disks as the MaxPage file. For a complex image, a notice- able delay occurs in redrawing the page, lb avoid this delay, you may elect an option in which images are neither displayed or printed. Background* MaxPage allows you to import a full page PICT image into your document as a "background." Unlike the image handling covered above, back- grounds are actually imported into and become part of your MaxPage document. This feature is intended to contain master elements (e.g., bor- ders, letter heads, or logos) or to serve as a design templet. Conclusion After my look at MaxPage, I conclude that you get what you pay for. . . maybe. What MaxPage has going for it is its cheap price. Disadvantages include the lack of page preview, no text tabs or Justification, and nonstandard ways of handling graphics. If all you need is to lay out simple, occasion, single page, perhaps MaxPage will serve your needs. All in all, however, I cannot recom- mend this product.. A Review of QuickLetter Over recent years there have been a number of mini woid processors in desk accessoiy form for the Macintosh. The quality and number of fea- tures in those programs varied widely but with a few being quite powerful. I have just received QuickLetter from Working Software, Inc., and it is a jewel. What it Does QuickLetter is primarily designed for one thing - wilting letters. Tb that end it provides for setting up standard statlonaiy (with letter heads and logos), maintaining an address book. Importing addresses into your correspondence, and printing letters and envelopes. When you open QuickLetter, you may use the default "statlonaiy" (to be discussed later), open other McWrlte or QuickLetter documents or create a new document. Your document then appears as does a ruler (which may be hidden) with various tools and a set of menu bars. Your document may be either landscape or portrait oilentation in any standard size. QuickLetter 's text tools are fairly standaid so I won't detail them. A featuie that I like is the ability to use text size up to 200. Also, in QuickLetter, you have more control over setting margins and line spacing than in MacWrite. QuickLetter allows you to set up standard sta- tlonaiy to speed up the letter writing process. Such stationary already contains standaid graphics features (e.g., letter heads borders and /or logos) margin and tab settings, and text specifications. One such stationary design may be designated as the default stationary and will automatically ap- pear when you start QuickLetter. You may import graphics for use as a logo or letterhead. Whatever you import is automatically placed in the upper left hand comer of the page and cannot be moved. This lack of flexibility is a significant weakness in QuickLetter. Consider- able trial and eiror may be necessary to design the graphic so that when pasted into QuickLetter, it is corrected positioned. QuickLetter contains a handy address book for in putting addresses to your letter or envelope. This feature also can be used to print labels. You may search the address book for key words such as a part of a name. The print preview feature for QuickLetter is especially nice. Not only can you easily preview your letter and envelope as they will piint but you can move the text around on the page independent of the letterhead to visually position it as you like. The Bottom Line In summary, this is an unexpected nice prod- uct. I keep finding useful features that I didn't expect to find. If the imported graphics could be more easily located, it would be perfect. But that weakness can be worked around and so, on the balance, QuickLetter's strengths outweigh its weaknesses and is recommended. EOF fq® mam wm^ 68 MICRO JOURNAL 1 ' I 68 Micro Journal April '89 29 Telephone: ($ 1 5} 842-4600 SOU tft TtClS t 9^6 dlU OS-9, UrdJOEX, TL'LX, S'K'^OS Telex: $106006630 Jtue (615) 842-7990 ASSEMBLERS ASTRUK09 from S.E. Media •• A "Structured Assembler for the 6809" which requires the TSC Macro Assembler. FLEX. SK DOS. CCF • $99.95 Macro Assembler for TSC - Ihe FLF.X. SK DOS STANDARD Assembler. Special -- CCF $3500; FLEX. SKDOS $50.00 OSM Extended 6809 Macro Assembler from Uoyd I/O. -- Provides local labels, Motorola S-records. and Intel ilex records: XREF Generate OS-9 Memoiy modules under FLEX, SK DOS. FLEX. SK.DOS, CCF. OS-9 $99.00 Relocating Assembler/Linking Loader from TSC. -- Use with many of the C and Pascal Compilers. FLEX . SKDOS, CCF SI 50.00 MACE, by Graham Trott from Windrush Micro Systems ■■ Co-Resident Editor and Assembler; Cast interactive A.I- Programming for small to medium-sized Programs. FLEX. SKDOS, CCF . $7500 XMACE MACE w/Cross Assembler for 6800/1/2/3/8 FLEX. SK DOS. CCF ■ $98.00 DISASSEMBLERS SUPER SLEI..TH from Computer Systems Consultants Interactive Disassembler: exiremely POWERFUL! Disk File Binary/ASCII Examine/Change, Absolute or FULL Disassembly. XRliF Generator, label "Name Changer", and Files of "Standard Label Names" for different Operating Systems. Color Computer SS-50 Bus (all wt AL Source) CCO (32 K Req'd) Object Only $49 00 FLEX. SK DOS $99.00 - CCF Object Only $50.00 UntFLEX $100 00 CCF. »ilh Source $99.00 OS-9. $10100 ■ CCO. Object Only $50.00 68010 SUPER SLEUTH . Similar to 8- Bit Version except written in "C". 68010 Disassembler $100.00 FLEX. UntFLEX. UNIX, XENIX, MS DOS, SK DOS. OS-9 OS-9/68K Object Only $100.00 or mth Source $20000 DYNAMITE* -• Excellent standard "Hatch Mode" Disassembler. Includes XREF Generator and "Standard Label" Files. Special OS-9 options with OS 9 Version. CCF. Object Only $100.00 ■ CCO, Object OnlyS 59.95 FLEX, SKDOS . Object Only $100.00 - OS-9. Object Only$l50O0 UniFLEX Object Only $300.00 CROSS ASSEMBLERS CROSS ASSEMBLERS from Computer System Consultants -- Supports 1 802/5. Z-80, 6800/1/2/3/8/1 1/HCI 1 , 6804, 6805/HC05/ 146805. 6809/ 00/01. 6502 family, 8080/5. 802O/1/2/35/C35/39/ 40/48/C48/49/C49/50/ 8748/49. 8031/51/875132000 and 6BO0O/680I0 Systems. Assembler and Listing formats same as target CPU's format. Produces machine independent Motorola S Text. Includes Macro Preprocessor. Written in C'- 68000 or 6809 'Macintosh' Atari. FLEX. CCF. UntFLEX. OS-9, XENIX, UNIX. MS DOS. SK DOS any object $50 or any 3Jor $100 any source is an additional $50 or any 3 for $100 Set of ALL object $200.00 - wtlh source $50000 XASM Cross Assemblers for FLEX, SKDOS from S.E. MEDIA - liiis sei of 6800/1/2/3/5/8. 6301, 6502, 8080/5. and 280 Cross Assemblers uses the familiar TSC Macro Assembler Command Line and Source Code fonnal. Assembler options, etc., in providing code for target CPU's. Complete set, FLEX. SK.DOS only - $150.00 CRASM1I from LLOYD I/O - Supports Motorola's. Intel's. Zilogs, and other's CPU syntax for ihese 8-Bit microprocessors: 6800. 6801, 6303, 6804.6805,6809.6811 (all varieties): 6502, 1802/5, 8048 family, 8051 family. 8080/85. Z8, Z80.and TMS-7000 family, lias MACROS, Local Labels, Label XREF, Label Length to 30 Chars. Object code formats: Motorola S Records (text), Intel HEX Records (text), OS-9 (binary), and FLEX, SK-DOS (binary) Written in Assembler ... e.g. Vcrv Fist. CPU TYPE - Price each: For. MOTOROLA INTEL FLEX9 $150 SKDOS $150 OS 9/6809 $150 OS9/68K CRASMII 1632 from IXOYD I/O $150 $150 $150 OTHER COMPLETE SET SI 50 S399 SI 50 $399 $150 $399 $432 Supports Motorola's 68000, and has same features as the 8 bit version. OS9/68K Object code Fonnat allows this cross assembler to be used in developing your programs for OS-9/68K on y«ur OS-9/6809 computer. FLEX. SKDOS. CCF. OS 916809 $24900 COM MU NIC A TIONS CMOUEM Telecommunications Program from Computer Systems Consultants, Inc. -- Maiu-Dr ven: supports Dumb-Tcnninal Mode, Upload and Download in non-protocol mode, and the CP/M "Modem?" Chrisiensen protocol mode to enable communication capabilities for almost any requirement Wntlcn in "C" HEX. SK DOS. CCF. OS 9. UniFLEX. UNIX. XENIX. MS-DOS. with Source $100 00 - without Source $5000 X-TALK from S.ii. Media - X-TALK consists of two disks and a special cable, the hookup enables a 6809 SWTPC computer to dump UniFLEX files directly to the UniFLEX MUSTANG 020. this is the ONLY currently available method to transfer SWTPC 6809 UniFLEX files to a 68000 UniFLEX system. Gimix 6809 useis may dump a 6809 UniFLEX file to a 6809 UniFLEX live inch disk and it is readable by the MUSTANG-030. The cable is specially prepared with internal connections to match the non-standard SWITC SO/9 J/O Db25 connectors. A special SWTPC S+ cable set is also available. Users should specify which SWTPC system he/she wishes to communicate with the MUSTANG 020. The X-TALK software is furnished on two disks. One eight inch disk contains S.E. Media modem program C- MODEM (6809) and the oilier disk is a MUSTANG 020 five inch disk with C-MODEM (68020). Text and binary files may be directly iransfened between the two systems. The C-MODEM programs are unaltered and perform as excellent modem programs also. X-TALK can be purchased with or without the special cables, but this special price is available to registered MUSTANG 020 users only. X-TALK Complete (cable, 2 disks) $99.95 X-TALK Software (2 disks only) $69.95 X TALK with CMODFM Source $14995 XD ATA from SE Media A COMMUNICATION Package for ihc UniFLEX Operating System. Use with CP/M, Main Frames, other UniFLEX Systems, etc. Verifies Transmission using checksum or CRC; Re-Transmits bad blocks, etc, UntFLEX- $29999 A»»ltaWllly Ltiendi O.OS-J, S.SK«DOS r.FLEX.u.unirux CCO ■ Color Compute OS-» CO . Colo. Conpwltr KIXX South TCast 'Media S900 Cassandra Smith 1(J. - "HvQon., Ttt, 3 7543 ** Shipping •• AM 1* ILS.A. (mln.SlM) Fortlpi Su/Tve A44 S» J'ortlpi Almll Add !•*» OrC.O.D.ShlPBlatOnl; 'OS.»lllTridrni.r»orMkri»CTM»i»i)»>^»«»<»-'»- , LtX«^dU«IFUO(irtTr«a^lurlll<ltTKlinlc»l SjUfmiCnniultmU-'SK' DOS bmTmtmnrH oTSUt-K S«rft»»r» SjIM—Cwt. jq April '89 68 Micro Journal 'fefcfihoiic: (615) 842-4600 South "East Media OS -9, ilni'J L'LX, 'JL'LX.S'K'-DOS 'fa*: (615) $42-7990 PROGRAMMING LANGUAGES PL/9 from Windrush Micro Systems - By Gnhim Tru4i. A combuiauon Editor Compiler Debugger. Direct source -lo-objcct compilation delivering fail, compact, re-entrant. ROM-able. PIC. 8 & 16-bil (niegers & 6-digii Real numbers for all real-world problems. Direct control over ALL System resources, including interrupts. Comprehensive library support: simple Machine Code interface: step- by-step tracer for instant debugging. 500+ page Manual with tutorial guide. FLEX. SKDOS. CCF - J/94.00 PASC from S.E. Media - A FLEX9. SKDOS Compiler with a definite Pascal "flavor". Anyone with a bit of Pascal experience should be able to begin using PASC to good effect in short order. The PASC package comes complete with three sample progiams: ED (a syntax or structure editor), EDITOR (a simple, public domain, screen editor) and CHESS (a simple chess program). The PASC package comes complete with sourer (written in PASC) and documentation. FLEX. SKDOS 195 00 WHIMSICAL from S.E. MEDIA Now supports Real Numbers. "Structured Progtamming" WITHOUT losing the Speed and Control of Assembly Language! Single-pass Compiler features unified, user-defined I/O; produces ROMable Code; PtvccAi res and Modules (including pre- compiled Modules); many "Types" up to 32 bit Integers, 6- digit Real Numbers, unlimited sized Arrays (vectors only); Interrupt handling; long Var able Names; Variable Initialization; Include directive; Condiiiorui I compiling: direct Code insertion; aonuol of the Stack Pointer; etc Run-Time subroutines inserted as called during compilation. Normally products 10% less code than PU9 FLEX. SK DOS and CCF J/95 00 KANSAS CITY BASIC from S.E- Media - Basic for Color Computer OS9 with many new commands and sub. functions added. A full implementation of Ihe It''-T!iliN-iiLSI: logic is included, allowing nesting to 2SS levels Sitings are supported and a subset of the usual suing funcuons such as LEFTS, RIG 1 1 TV MIIK. STRINGS, etc. arc included. Variables are dynamical) y allocated. Also included arc additional features such as Peek and Poke A must for any Color Computer user running OS-9. CoCo OS-9 139 95 C Compiler from Wlndrush Micro Systems by James McCosh. Full C for FLEX. SKDOS except bil-f elds, including an Assembler. Requires Ike TSC Relocating Assembler if user desires to implement his owl Libraries. FLEX. SKDOS. CCF - J295.00 C Compiler from Introl - Full C except Doublet and Bit Fields, streamlined for the 6809. Reliable Compiler: FAST, efficient Code. More UNIX Compatible than most. FLEX. SKDOS. CCF. OS-9 (Level II ONLY). UniFLEX - 1575.00 PASCAL Compiler from Lucldata -• ISO Based P Code Compiler. Designed especially for Microcomputer Systems. Allows linkage to Assembler Code for maximum flexibility. FLEX. SKDOS and CCF -1190.00 OmegaSoR PASCAL from Certified Software - Extended Pascal for systems and real-time programming. Native 680QtV58<Efl Compiler, $575 for base package, options available. For OS-9/68000 and PDOS host system. 6809 Cross Compiler (OS-9/68000 host) $700 for complete package. KBASIC - from S.E. MEDIA - A "Native Code' BASIC Compiler which is now Fully TSC XBASIC compatible. Ihe compiler compiles to Assembly Language Source Code. A NEW, streamlined. Assembler is now included allowing the assembly of LARGE Compiled K-BASIC Programs. Conditional assembly reduces Run-lime package. FLEX. SK DOS. CCF. OS-9 Compiler /Assembler $99 00 CRUNCH COBOL from S.E. MEDIA - Supports large subset of ANS11 Level 1 COBOL with many of the useful level 2 features. Full FLEX. SK DOS File Structures, including Random Files and the ability to process Keyed Filet. Segment and link large programs at runtime, or implemented as a set of overlays. The System icquires S6K and CAN be run with a single Disk System. A very popular product. FLEX. SKDOS, CCF . 199 9S FORTH from Stearns Electronics -- A CoCo FORTH Programming Language. Tailored to the CoCot Supplied on Tape, transferable to disk. Written in FAST ML. Many CoCo functions (Graphics, Sound. etc.). Includes an Editor, Trace, etc. Provides CPU Cany Flag accessibility. Fast Task Multiplexing, Clean Interrupt Handling, etc. for the "Pro". Excellent "Learning" tcoll Color Computer ONLY ■ 158 95 FORTHBUILDER it a stand-alone target compiler (crosscompiler) for producing custom Foith systems and application progiams. All of the 83-standard defining words and control structures are recognized by FORTHBUILDER. FORTHBUILDER is designed to behave as much as possible like a resident Foith interpieter/compiler. to thai most of the established techniques for wiiting Foith «odc can be used without change. Like compilers far other languages. FORTHBUILDER can operate in "batch mode". Ihe compiler recognizes and emulates target names defined by CONSTANT or VARIABLE and it readily extended with "compile- umc" definitions to emulate specific target words. FORTMBUILDKR is supplied as an executable command file configured for a specific host system and target processor. Object code produced from the accompanying model source code u royally- free to Ikcntcd users. FLEX. CCF. SKDOS - J99 95 EDITORS & WORD PROCESSING JUST from S.E. Media — Text Formatter developed by Ron Anderson; for Dot Manix Printers, provides many unique features. Output "Formatted" Text to the Display. Use the FPRINT.CMD supplied for producing multiple copies of the "Formatted" Text on the Printer INCLUDING IMBEDDED PRINTER COMMANDS (very useful at other times also, and woilh Ihe price of the program by itsell). "User Configurable" for adapting to other Piintcrs (comes set up for Epson MX-80 with Graflrax); up to ten (10) imbedded "Printer Control Commands". Compensates for a "Double Width" piinted line. Includes the normal line width, margin, indent, paragraph, space, vertical skip lines, page length, page numbciing. centering, fill, justification, etc. Use with PAT or any oilier editor. * Now supplied as a two disk set Disk*! JUS12CMD object file. WS17.1TTPL9 source FIEX. SK DOS . CCF Disk 02: JUSTSC object and source in C. FLEX, SKDOS, OS 9. CCF The JTSCand regular JUST C source arc two separate piognms. JTSC compiles to a version that expects TSC Word Processor type commands, (pp sp ce etc ) Great for your older text files. TheC 0.04-*,S.SK*00£ F.n-EX. U. ttairLEX CO • Cote. Cunantir OS.* CCF . Color CooiooUr FUI South 'East 'Media 5900 Cassandra Smith %d. - Xi*f<m, Tn. 37343 rtv>.iVrr,ircl •• Shipping •• A<M 1* USA. (aria. 115a) tw^aa WTMAMn FVvtIcp Alm.ll AM 1*» Or C-O.D. S birr- Ina Ool, •OS.»U» Tr.««~r> 1 if Mkro^rtMo Mntorcla.'FLEX ■!>< UalPl-g»ar«Tr«4«iii«f>jof T«cb»tc»l Sj««i«C<mniln»l».*Slf POS tufroomumof 81«rHSWr-»rt Sfiu—C^y. 68 Micro Journal Apr. 89 31 Tefepfiorte: (615) 842-4600 SOU tfl 'EOS t 94e did OS-9, Vni'/L'EJC, JL'LX, $%'<D0$ Telex: HlOiiOlbnM) 7&KI (615) 842*7990 source compiles to * standard syntax JUST.CMD object Tile. Using JUST tynuu (.p ,u ,y etc.) Wjih all JUST funcuons plus several additional printer formatting (unctions. Reference the JUSTSC C source. For Ihose wanting an excellent BUDGET PRICED word processor, with features none of (he olhers have. This is ilt Disk (I) ■ PL9 FLEX only- FLEX. SK DOS & CCF - S49 95 Disk Set (2) . FLEX. SK DOS A CCF &. OS-9 (C version) - 169 95 OS-9 68K0O0 complete with Source - V9 95 PAT from S.E. Madia - A full feature screen oriented TEXT EDITOR with all the. best of "PIE™". For those who swore by and loved only PIE. this is for youl All PIE features and much morel Too many features to list. And if you don't like these, change or add your own. PL-9 source furnished. "C" source available soon. Easily configured to your CRT, with special config section. Regular FLEX, SK DOS $ 12950 • SPECIAL INIRODUCTION OFFER • S79.95 SPECIAL PATIJUST COMBO (with source) FLEX. SK DOS S99.95 OS-9 68K Version S229J0O SPECIAL PAT/JUST COMBO 68K S249J0O Note: JUST in "C" source available for OS-9 CEDRIC from S.E. Media ■ A screen oriented TEXT EDITOR with availability of 'MENU' aid. Macro definitions, configurable 'permanent definable MACROS' - all standard features and the fastest global' functions in the west A simple, automatic terminal config program makes this a real 'no hasscl* product. Only 6K in size, leaving the average system over 165 sectors for text buffer - appx. 14,000 plus of free memory I Extra fine for programming as well as text. FLEX, SKDOS S69 95 BAS-EDIT from S.E. Media - A TSC BASIC or XBASIC screen editor. Appended to BASIC or XBASIC. BAS-EDIT is transparent to normal BASIC/XBASIC operation. Allows editing while in BAS1C/XBASIC. Supports the following functions: OVERLAY. INSERT and DUP LINE. Make editing BASIC/XBASIC programs SIMPLE1 A GREAT lime and effort saver. Programmers love it! NO more retyping enure lines, etc Complete with over 25 different CRT terminal configuration overlays. FLEX.CCF.SK DOS 139.95 SCREDTTOR III from Wmdrush Micro Systems •• Poweiful Scieen- Oriented Editor/Word Processor. Almost SO different commands; over 300 pages of Documentation wilh Tutorial. Features Multi-Column display and editing, "dednul align" columns (AND add them up automatically), multiple keystroke macros, even/odd page headen and footer), imbedded piinter aontiol codes, all justifications, "help" support, iiore common command series on disk, etc Use supplied "set- ups", or remap the keyboard to your needs. Except for proportional pliming. this package will DO IT ALLI 6800 or 6809 FLEX, SKDOS or SSB-DOS. OS 9 . S175JOO SPELLB "Computer Dictionary" from S.E. Media -- OVER 150,000 words! Look up a word from within your Editor or Word Processor (with the SPHCMD Utility which operates in the FLEX. SKDOS UCS) Or check and update the Text after enliy; ADD WORDS to the Dictionaiy, "Hag" questionable words in the Text, "View a word in context" before changing or ignoring, etc. SPEUJ) lint cheeks a "Common Wad Dictionary", ihen the normal Dictionary, then a "Personal Woid List", and finally, any "Special Woid List" you may have specified. SPELLB also allows the use of Small Disk Storage systems. FLEX. SKDOS and CCF ■ SI29.95 STYLO-GRAPH from Great Plains Computer Co. •- A fulll-screen oriented WORD PROCESSOR - (uses the 51 x 24 Display Screen* on CoCo FLFJC/SK-DOS. or PBJ Wordpak). Full screen display and editing; supports the Daisy Wheel proportional printers. NEW PRICES 6809 CCF and CCO ■ S99 95, FLEX. SKDOS or OS 9 SI79 95. UniFLEX. S299 95 STYLO-SPELL Iran Great Plains Computer Co. - Fast Computer Dictionary. Complements Stylograph. NEW PRICES 6809 CCF and CCO - 169 95. FLEX. SKDOS or OS-9 - S99.95. UniFLEX- S149 95 STYLO. MERGE from Great Plains Computer Co. - Meige Mailing list to "Form" Letters, Print multiple Files, etc., through Stylo. NEW PRICES 6809 CCF and CCO . S59 95. FLEX. SKDOS or OS-9 . 179 95, UniFLEX. SI29 95 STYLO PA K -Graph + Spell ♦ Merge Package Deal III FLEX, SK DOS or OS-9 . S329.95. UniFLEX - S549 95 OS-9 68000 S695 00 DATABASE ACCOUNTING XDMS from Westchester Applied Business Systems FOR 6809 FLEX or SKDOS (5/8") Up to 32 groups/fields per record! Up to 12 character file names) Up to 1024 byte records! User defined screen and print control! Process files! Form filesl Conditional execution) Process chaining! Upward/Down waid file Unkingl File Joining! Random Tile virtual paging) Built in utilities! Built in text line editor! Fully session ortcnladl Enhanced forms! Boldface, Double width, Italics and Underline supported I Written in compact structured assembler! Integrated for FAST execution 1 XDMS-IV Data Management System XDMS-I V is a brand new approach to data management It not only permits users to describe, enter and retrieve data, but also to process entire Tiles producing customized reports, screen displays and file output. Processing can consist of any of a set of standard high level functions including record and field selection, sorting and aggregauon, lookups in other files, special processing of record subsets, custom report formatting, totaling and subtouling, and pnscntaUon of up to three related Tiles as a "database" on user defined output reports. POWERFUL COMMANDS! XDMS-IV combines ihe functionality of many popular DBMS software systems with a new easy to use command set into a single integrated package. We've includad many new features and commands including a set of general file utilities. The processing commands are Input-Proccss- Outpuf (IPO) which allows almost instant implementation of a process design. SESSION ORIENTED! XDMS-IV is session oriented. Enter "XDMS" and you are in instant command of all the feaiuies. No more waiting for a command to load m from disk! Many commands are immediate, such as CREATE (file definition). UPDATE (file editor), PURGE and DELETE (utilities). Olhers are process commands which are used to create a user process which is executed with a RUN command. Either may be entered into a "process" file which is executed by an EXECUTE statement Processes may execute other processes, or ihem selves, cither conditionally or unconditionally. Menus and screen prompts arc easily coded, and entire user applications can be run without ever leaving XDMS. IV O • CHS-*, s . SK'DOS r,nrt,u. uatnu nra , a*» t as s tft s a v os.« ccf . m. i, w ii> rucx South 'Lost Media 5900 Cassandra. Smith KjC ■ 9fc(fon. Tn. 37343 " Shipping •• am i* UiA. (jala. U.M) Pamaa Jaro» AM i% tmntmt Alraa IIU1 1«» Or COJJ. SAVrta, Oar> *l)S-*lii \ rtfamtth r>r Mkr.r*trt tnd M„t,f..u 'Kl IA ind l.nlM.EX iff I rtetmtr hiot Tnhnkal SynnniCwnuluml*.*SK* DOS !■ m Trademark oT sue .K Salt^irt SymnM tjra. 32 April '09 68 Mcro Journal Tefcpfione: (6 1 5) $424600 SOU tfl 'EtlS t (Media OS-9, ihu'.fL'LX, fC'EX, S'K'DOS Ti-tcx: SlObOObt&O fa*: fa 15) 842*1 ITS EASY TO USE! XDMSIV keeps data management simple! Rather than design i complex DBMS whidi hidei the Uue nuure of the data, we kepi XDMSIV Hie oriented. The user view of dais relationship! is presented in reports and screen output, while ihe actual data resides in easy 10 maintain dies. This aspect pcmuis customitcd presentation and reports without complex redefinition of the database files and structure. XDMS-IV may be used for a wide range of applications from simple record management systems (addresses, inventory ...) to iniegiaied database systems (order entiy, accounting...) The possibilities ire unlimited... FOR 6809 FLEX or SK-DOS(5"/8" Disk) (249.95 UTILITIES Bajlc09 XRef from S.E. Media •• This Basic09 Cross Reference Utility is a Basic09 Program which will produce a "pretty primed" listing with each line numbered, followed by a complete cross referenced listing of all variables, externa] procedures, and line numbers called. Also includes a Program List Utility which outputs a fast "pretty printed" lining with line numbers. Requites 8asic09 or RunB. OS-9 A CCO objtel only - S39.9S; with Source - $79.95 BTrtt Routines • Complete set of routines to allow simple implementation of keyed files -for your programs - running under Basie09. A real time saver and should be a part of every serious programmers tool-box. OS-9 A CCO objtel only ■ SS9 9S Lucldata PASCAL UTILITIES (Requires Pascal ver 3) XREF ■• produce a Cross Reference Listing of any text: oiiented to Pascal Source INCLUDE -- Include other Files in a Source Text, including Binaiy - unlimited nesting. PROFILER - provides an Indented, Numbered, "Structogram" of a Pascal Source Teal File; view the overall structure of large piogiams, program inlcgniy. etc Supplied in Pascal Source Code: iccjuirei compil atton. FLEX. SKOOS. CCF — EACH 5~ ■ UOW, «" - S50M DUB from S.E. Media - A UniFLEX BASIC decompiler Re-Creale a Source Listing from UniFLEX Compiled basic Prog/vns. Woiks with ALL Versions of 6809 UniFLEX basic. UniFLEX ■ J2I9.95 LOW COST PROGRAM KITS from Southeast Media The following kits are available for FLEX. SKDOS on ether J' or 8" Disk. 1. BASIC TOOL-CHEST $29.95 BUSTER. CM D: pretty priner LINEXREF.BAS: line crots-refercnccr REMPAC.BAS, SPCPAC.BAS. COMPACBAS: remove supcriluous code STRIP. BAS: superfluous line-numbers mipper 2. FLEX, SKDOS UTILITIES KIT $39.99 CATS. CMD: alphabetical! y -sorted directory listing CATD-CMD: date-sorusd directory listing C0PYSORT.CMD: file copy, alphabetically COPYDATE.CMD: file copy, by dale-order FILEDATE CMD: change file creation date INFO.CMD (4 fNFOGMX.CMD): tells disk attributes A. contents RELINK CMD {& RELINK82): reorders fragmented free chain RESQ.CMD: undelete* (recovers) a deleted file SECIORS.CMD: show sector order in free chain XLCMD: super text lister 3. ASSEMBLERS/DISASSEMBLERS UTILITIES S39.9S LINEFEED.CMD: modularise' disassembler output MATH CMD: decimal, hex, binary, octal conversions A tablet SKIP.CMD: column stripper 4. WORD - PROCESSOR SUPPORT UTILITIES $49.95 FULLSTOP.CMD-. chucks for capitalization BSTYdTBAS (.B AC- Stylo to dot matrix printer NECPRINT.CMD: Stylo to dot-matrix primer filler code 5. UTILITIES FOR INDEXING $49.95 MENU.BAS: selects nquintd program from list below INDEX BAC: word index PHRASES BAC; phrase index CONTENT.B AC table of contents INDXSORT.BAC: fast alphabetic son routine FORMATER.BAC: produces a 2-coiumn formatted index APPEND.BAC: append any number of files CHAR BIN: line reader BASIC09 TOOLS consist of 21 subroutines for Basic09. 6 were written in C language and the remainder in assembly. All the main cs arc compiled down to native machine code which makes them fast and compact. 1. CF1LL - fills a string with characters 2. DPEEK - Double peek 3. DPOKE - Double poke 4. FTOS - Current file position J, FS1ZE - File sue 6. FTRIM -- removes leading spaces from a string 7. GETPR -- returns the current process ID 8 GETOPT - gels 32 byte option section 9. GETUSR - gets the user ID la GTIME - gets the tune 11. INSERT - insert a suing into another 12. LOWER -- converts a string into lowercase 13. READY -- Checks for available input 14. SETPRIOR - changes a process priority 15. SETUSR - changes the user ID 16. SETOPr •- set 32 byte option packet 17. STIME - sets the time 18. SPACE -- adds spaces to a string 19. SWAP — swaps any two variables 20. SYSCALL - system call 21. UPPER - converts a string to uppercase For OS-9 - $44.95 • Includes Source Code SOFTOOLS The following programs are included in object foun for trrunedi ate application. PL/9 source code available for custom zalion . READ-ME Complete instructions for initial set-up and operation. Can even be printed out with the included text processor. CONFIG one time system configuration. CHANGE changes words, characters, etc. globally to any text type file. CLEANTXT converts text filet to standard FLEX, SK-DOS Tiles. COMMON compare two text files and reports differences. COMPARE another check file that reports mis-matched lines. CONCAT similar to FLEX. SKOOS append but can also list filet to screen. DOCUMENT for PL/9 source files. Very useful in examining parameter passing aspects of procedures. A.I ■ OS-t, I ■ SK-DOS r.FLtX.U.UalFLM cca . Goto* ca»p«itr o&-f CTF - Cofc* ConpuUT TXXX South "East 'Media S9Q0 Cassandi-aSmitk Hf. ■ 9&rftm, Tn. J7i4i L^ •• Sblppln* •• A44 1% u&A. (ai& tut) Fantaa JvrCn KU I* fW<Sa> Air— tH A** If* OrC.O-D.SlM»f*>aO«rj *p«-f t»« TT»*mMrt »f Mttrenww ma Matjrala.'ST.rtK mn* Unlin.EX in Tr»4fm»rtn «f Technical Sjujim Gnunlmiu-'SK*D08 hi i ■fc af Star-K IcrtTt 1 jsstsas Cfp. 66 Mrcro Journal April '89 33 'fctcpinmc: (6 I 5> 842-4600 SOU tft T,OS t fMcdid OS-9, Uni'JCLX, 'JLLX, S'K'^OS •I\'tf.\: 5106006630 tFwc (615) 842-7990 ECHO echo* lo other scran or file. FIND an unproved find command with 'pattern" matching and wildcaids. Veiy useful. HEX dumps files in both hex and ASCII, INCLU'OEi file copy program thai will accept "includes" of other disk files. K WIC allows rousing each word, on each line to ilie beginning. Very useful in a sort program, etc. LISTDIR a directory listing program. Not super, but better than CAT. MEMSORT a high-speed text file softer. Up lo 10 fields may be sorted. Veiy fast. Very useful. MU LTICOL width of page, number of columns may be specified. A MUST! PAGE similar lo UST but allows for a page header, page width and depth. Adjust for CRT screen or printer as set up by CONFIG. A very smart piinl driver. Allows printer control commands. REMOVE a fast file deleter. Careful, no prompts issued. Zap, and its gone! SCREEN a screen listing utility. Word wraps text lo fil screen. Screen depth may be altered al run time. SORT a super version of MEMSORT. Ascending/descending order, up to 10 keys, case over-ride, son on n" word and sort on characters if file is small enough, sons in RAM. If large file, son is constrained to size of your largest disk capacity. TPROC a small Nil ruvc tat formatter. Ilus i» a complete formatter and has functions not found in other formatters. TRANSLIT sons a file by x keyfiields. Check! for duplications. Up to 10 key files may be used. UNROTATE used with KWIC this program reads an input file and unfolds il a line al a lime. If ihe Tile has been soiled each word will be presented in tequoMC WC a word count utility. Can count words, characters or lines. NOTE: this set of utilities consists of 6 5-1/4" dbfcs or 2 8" disks, with source (PL9). 3 S- 1/4" disks or I 8" disk without source. Complete set SPECIAL INTRO PRICE: 5-1/4" with source FLEX or SK-DOS - S129 95 without source • $79.95 8" with source • $79.95 • without source $49.95 FULLSCREEN FORMS DISPLAY from Computer Systems Consultants - TSC Extended BASIC program suppons any Serial Terminal with Cursor Control or Memory-Mapped Video Displays; substantially extends Ihe capabilities of the Program Designer by providing a table- driven method of describing and using Full Screen Displays. FLEX. SK DOS and CCF. UniFLEX . S2HD0. with Source - $5000 SOLVE from S.E. Media - OS-9 Levels I and II only. A Symbolic Object/ Logic Verification St. Examine debugger. Including inline debugging, disassemble and assemble. SOLVE IS THE MOST COMPLETE DEBUGGER we have toot for the 6809 OS-9 series! SOLVE does it all! With a rich selection of monitor, assembler, disassembler, environmental, execution and other miscellaneous commands, SOLVE is ihe MOST POWERFUL tool kit item you can own) Yet. SOLVE is simple U> use! With complete documentation, a snap! Everyone who has ordeied this package has raved 1 Sec review -68 Micro Journal - December 1985 No 'blind' debugging here, full screen displays, rich and complete in information presented. Since review in 68 Micro Journal, this is our fastest mover! levels I A II only ■ OS-9 $69.95 DISK UTILITIES OS-9 VDisk from S.E. Media - For Level I only. Use ihe Extended Memory capability of your SWTPC or Giinix CPU card (or similar format DAT) for FAST Program Compiles, CMD execution, high spead inter-process communications (without pipe buffers), etc. - SAVE lhai System Memoiy. Virtual Disk size is variable in 4K increments up lo 960K. Some Assembly Required. Level I OS-9 object $79 95. with Source $149 95 O-F from S.E. Media ■• Written in BASIC09 (with Source), includes: REFORMAT, a BASIC09 Program that reformats a chosen amount of an OS-9 disk to FLEX, SK-DOS Formal so it can be used normally by FLEX. SK.DOS: and FLEX, a BASIC09 Program that does the actual read or write function to the special O-F Transfer Disk: user-friendly menu driven. Read Ihe FLEX. SKDOS Dirccloiy. Delete FLEX, SK.DOS Files, Copy both directions, etc. FLEX. SKDOS users use the special disk just like any other FLEX. SKDOS disk OS-9 - 6909/68000 $79.95 [.SORT from S.E. Media - A SORT/MERGE package for OS-9 (Uvel I A U only). Sons record* with fixed lengths or vaiiable lengths. Allows for either ascending or descending son. Soiling can be done in either ASCII sequence or alternate collating sequence. Right, lefl or no justification of data fields available. LSORT includes a full set of comments and errors messages. OS-9 $95 DO HIER from S.E. Media - HIER is a modern herarcnal storage system for users under FLEX. SK-DOS It answers the needs of those who have hard disk capabilities on their systems, or many files on one disk - any size. Using HIER a regular (any) FLEX, SK.DOS disk (8 . 5 . hard disk) can have sub directories By this method the problems of assigning unique names to Files is less buidentomc. Different files with ihe exact same name may be on the same disk, as long as ihey are in different directories. For the Winchester uaerlhis became* a must. Sub directories are Ihe modem day solution thai all current taige systems use. Each directory looks to FLEX, SK-DOS like a regular file, except they have the extension '.DIR'. A full set of directory handling programs are included, making the operation of HIER simple and straightforward. A special install package is included lo install HIER to your particular version of FLEX, SK-DOS. Some assembly required. Install indicates each byte or reference change needed. Typically • 6 byte changes in source (furnished) and one assembly of HIER is all that is required. No programming requited! FLEX ■ SK DOS $79.95 COPYMULT from S.E. Media -- Copy LARGE Disks to several smaller disks. FLEX, SK-DOS utilities allow the backup of ANY size disk lo any SMALLER size diskettes (Hard Disk to floppies, 8" lo 5", etc.) by simply inserting diskettes a* requested by COPYMULT. No fooling with dirccloiy deletions, etc COPYMULT.CMD understands normal "copy" syntax and keeps up with files copied by maintaining directories for both host and icceivuig disk syitem. Also includes BACKUP.CMD to download any size "random" type file; RESTORE.CMD lo restructure copied "landom" files for copying, or recopying back to ihe noil system; and FREELDNK.CMD *s a "bonus" utility that "relinks" Ihe free chain of floppy or hard disk, eliminating fragmentation. Completely documented Assembly Language Source files included. ALL 4 Program (FLEX. SKDOS. S" or 5') $99.50 OiOS.I.S .SKTX)S r.n.£x,u>ifeiri.EX CO • Csk» Conautar 0»-« CCf - Color CompuUr FLRX South 'East Media 5900 Cassandra. Smith %£. - tfixjon. In. 37343 fli.i<tft i rrrj< , 4 ■• Shipping •• Add J* VSJk. (nirt. tl W) Pari!** Sumw Add S% tertian AlniBll Add IS* Or C.O.D. Sklrwlm Oolj *OS.»lai TradiMrlior Micro— rrmd Mcur*™!* .'Kl.KX arid IrnlKLEX ir< Trsd«marhjqfTghnkBlSjiltmi l^mmlUnu.'^k'IMls Ii I ra nt Star. X Saftw r% Sjmtnw f mt ■. 34 April '89 66 Micro Journal Tekphom: (615) 842-4600 SOU tfl 'EdS t ^Mcdla OS -9, 'Um'JfLX, <J L'EX, S'K''DOS Tc(e\: 5W600bbM) faTQ {€15) 842-7990 COPYCAT from Lucid.ta - Pascal NOT required. Allows leading TSC Mini-R.EX. SK-DOS. SSB-DOS68, and Digital Research CPfM Disks whileoperaling under SK DOS . FLEXI.O. FLEX 2.O. or FLEX 9.0 wiih 6800 Of 6809 Systems. COPYCAT will not perform miracles, but. between the program and the manual, you stand a good chance of accomplishing a transfer. Also includes some Utilities to help out. Programs supplied in Modular Source Code (Assembly Language) to help solve unusual problems. FLEX.SKDOS and CCF 5"$50.00 FLEX. SK DOS 8"-$65.00 VIRTUAL TERMINAL from S.E. Media ■ Allows one terminal to do the work of several. The user may stan as many as eight tasks on one terminal, under V1RWALTERMINAI. and switch back and foith between tasks at will. No need to exit each one: Just jump back and fouh. Complete with configuration program. *I"he hest way to keep up with those background programs 6809 OS-9 & CCO . object only - $49.95 KLEX, SK DOS DISK UTILITIES from Computer Systems Consultants Eight (8) different Assembly Language (with Source Code) FLEX. SK-DOS Utilities for every FLEX, SK-DOS Users Tootbox: Copy a File with CRC Eirors; Test Disk for errors; Compare two Disks; a fast Disk Backup Program; Edit Disk Sectors: Linear \w Free-Chain on the Disk; pint Disk Idcntilicaiion; and Son and Replace the Disk Dircctoiy (in sorted order). — PLUS — Ten XBAS1C Programs including: A BASIC Resequenccr with EXTRAS over "RENUM" like check for missing label definitions, processes Disk to Disk instead of in Memory, etc. Other programs Compare, Merge, or Generate Updates between two BASIC Programs, check BASIC Sequence Numbers, compare two unsbquenccd files, and S Programs for establishing a Master Dircctoiy of several Disks, and sorting, selecting, updating, and printing paginated listings of these files. A BASIC Cross-Reference Program, wiiuen in Assembly Language, which provides an X-Ref Listing of the Variables and Reserved Words in TSC BASIC. XBAS1C, and PRECOMPILER BASIC Programs. ALL Uliliiitt include Source (either BASIC or A.L. Source Cod*). FLEX. SK-DOS and CCF - 15000 BASIC Utilities ONLY for UniFLEX - $30 00 MS-DOS lo FLEX Transfer Utilities to OS-9 For68XXX and CCOS-9 Systems Now READ - WRITE - D1R - DUMP - EXPLORE Fl-EX &. MS-DOS Disk. These Utilities come with a rich set of options allowing the transfer of text type files from/to FLEX A MS-DOS disks. >CoCo systems require the DP. Johnson SDISK utilities and OS-9 and two dnves of which one must be a "host" floppy. 'CoCo Version $69 95 68XXX Version $99 95 MISCELLANEOUS TABULA RASA SPREADSHEET from Computer Systems Consultants TABULA RASA is similar lo DESKTOP/PLAN; provides use of tabular compulation schemes used for analysis of business, sales, and economic conditions. Menu-diiven; extensive report-generation capabilities. Requires TSCs Extended BASIC. FLEX. SK-DOS and CCF . UniFl£X- $50.00. with Source . $100.00 DYNACALC - Electronic Spread Sheet for the 6809 and 68000. UniFLEX- $395.00. FLEX, SK-DOS. OS-9 and SPECIAL CCF - $250.00 OS-9 68K -$299.00 FULL SCREEN INVENTORY/MRP from Computer Systems Consultants Use the Full Screen Inventory System/Materials Requirement Planning for maintaining inventories. Keeps item field file in alphabetical order for easier inquiiy. Locale and/or piint records matching partial or complete item, description, vendor, or auribtiies: find backorder or below slack levels. Pnnl-ouls in item or vendor order. MRP capability for the maintenance and analysis of Hierarchical assemblies of items in the inventory file. Requires TSCs Extended" BASIC. FLEX.SKDOS and CCF. UniFLEX - $50 00. *ilh Source . $100.00 FULL SCREEN MAILING LIST from Computer Systems Consultants - The Full Screen Mailing Lisi System provides a means of maintaining simple mailing lists. Locale all records matching on panial or complete name, city, state, zip, or attributes for Listings or Labels, etc. Requires TSCs Extended BASIC. FLEX. SK DOS and CCF. UniFLEX- $50.00. w,th Source ■ $10000 DIET-TRAC Forecaster from S.E. Media - An XBASIC program that plans a diet in terms of cither calorics and percentage of carbohydrates, protcms and fats (C V 0%) or grams of Carbohydrate. Protem and Fat food exchanges of each of the six basic food groups (vegetable, bread, meal, skim milk, fruii and fat) for a specific individual. Sex, Age. Height, Present Weight, Frame Size, Activity Level and Basal Metabolic Rale for normal individual are taken into account. Ideal weight and sustaining calories for any weight of the above individual are calculated. Provides number of days and daily calendar after weight goal and calorie plan is determined. FLEX.SKDOS $59.95. UniFt£X ■ $89.95 GAMES RAPIER - 6809 Chess Program from S.E. Media - Requires FLEX. SK-DOS and Displays on Any Type Terminal. Features: Four levels of play. Swap side. Point scoring system. Two display boards. Change skill level. Solve Checkmate problems in 1-2-3-4 moves. Make move and swap sides. Play while or black. This Is one of the strongest CHESS programs running on any microcomputer, estimated USCF Rating 1600 + (belter than most club' players at higher levels) FLEX. SKDOS and CCF ■ $79.95 NEW MS.DOSIFLEX Transfer Utilities For 68XXX and CoCo* OS-9 Systems. Now Read, Write. DIR, Dump and Explore FLEX & MS-DOS Disks. Suppled with a rich set of options lo explore and transfer text type Isles from/to FLEX and MS DOS disks 'CoCo OS-9 requires SDISK utilities & iwo floppy drives. CCO $69 95 68XXX OS-9 $99 95 MS-DOS and Macintosh Software at Discounted Prices "Call for prices, il"ll be worth the savings." (615) 842-4600 FAX (615)842-7990 AT>il<Mlli;Ua)ciub O.OS-J.S.SK'DOS t > FLEX, U • UniFLEX QC* • Color CompuUr OS.« CCF- Color Compear FLEX B South 'East Media S900 Cassandra Smith J{£. ■ Xvpon, Tn. i 7i4i motive arc 1 •• Shipping •• Add 2* US.A. (mlo. SLS*) rorrtan Surfoc. Add t% Foritan Airmail Add 19% Or CO.D. SolpploiOnljr ■05i-«IJ»Tridtm»rl<<?rMI<romr<irni Mnloml.-'H.HX «nd l.'nlH.HX »rt 1 r»drnurki ot Tmrinkjl Srnjnii 1 jmiulunlj-'S If IKBi li ■ I nJfmirl .Ulii.H Wl»»f f S}ll.mi tjrp. 68 Mcro Journal April 89 35 FLEX 6809 DIAGNOSTICS Disk Drive Test ROM Test, RAM Test INTRODUCTION The following ulilites are available on 68 Micro Journal Reader Service Disk #34. However, they are being published here as well because of the sharing spirit of the author: Emery Korpi 40 Beatrice Ave. Syossei, NY, 11791. I, and thousands of loyal readers thank you Emery for your willingness to shaie your efforts with all of us. It is that type of spirit lhat has allowad us to survive. As a contributor supported magazine we urge all our readers to do likewise. If so, we will be around a long time to come. Again, thanks - Emery! DMW DISC DRIVE TEST (DDT) NOTES: INTRODUCTION - THIS PROGRAM HAS BEEN WRITTENTO AID IN THE ALIGN- MENTOF 5 INCH* DISK DRIVES UNDER THE FLEX OP-SYSTEM. IT ALLOWS FORTESTINGOF DRIVES O THROUGH 3 WITH THE WD179X SERIES OF DISK CONTROLLERS IN ANY OF THE SYS- TEM I/O SLOTS. THE PROGRAM MAY BE USED WITH EITHER 1MHZ OR 2MHZ CLOCKED SYSTEMS AND IS COMPATIBLE WITH DC-I TO DC-4 OR FD-2 CONTROLLERS THIS PROGRAM HAS THE FOLLOWING TEST CAPABILITIES . 1. DRIVE READY SIGNAL RESPONSE TIME 2. HEAD IjOAD DELAY TIME 3. BUSY SIGNAL RESPONSE TIME 4. DRIVE ROTATIONAL SPEED TEST 5. INDEX PULSE WIDTH TEST 6. TRACKO SWITCH ADJUSTMENT 7. DRIVE STEP TIME 8. TEST OF ALL DISK SECTORS 9. RADIAL ALIGNMENT lO.READ/WRiTING/MODIFYING DATA ON ANY SECTOR. THIS PROGRAM ALSO CAN OPERATE WITH THE DYSAN DIGITAL DIAGNOSTIC DISKS IN ORDER TO FIND; 1. RADIAL TRACK ALIGNMENT ERRORS (IN MILLI-1NCHES) 2. STEPPER MOrOR ACCURACY AND REPEATABILITY PROB- LEMS 3. REDUCED AMPLITUDE SIGNAL PROBLEMS 4. AZIMUTHAL HEAD ALIGNMENT ERRORS A. STARTING THE PROGRAM - THE PROGRAM MAY BESTARTED IN EITHER OFTHREE WAYS. THESE ARE; 1. DDT - THE OPERATOR WILL BE PROMPTED FOR THE DRIVE NUMBER (0-3). THE SYSTEM CLOCK SPEED (1 OR 2 MHZ) AND THE I/O QLOT THAT THE CONTROLLER IS INSTALLED IN. 2. DDT,# - THE PROGRAM WILL TEST DRIVE NO '#' AND AS- SUMES A 1 MHZ CLOCK RATE AND I/O SLOT ONE. THESE DEFAULT VALUES MAY BE CHANGED IN THE SOURCE PRO- GRAM WHICH IS SUPPLIED. 3. DDT,>iile THEPROCRAMWILLBE PROMPTED THE SAME AS IN DDT AND A COPYOF ALL TERMINAL DATA WILL BE WRI1TEN INTo TEXT FILE file. B. TEST OPTIONS - THE OPERATOR OF THIS PROGRAM HAS THE OPTION OF HALTING AFTER EACH TEST. IF THIS OPTION IS NOT TAKEN, THE PROGRAM WILL PROCEED AUTOMATICALLY UNTIL A FAILUREOR KEYBOARD ENTRY IS ENCOUNTERED; ATWHICH POINT IT WILL HALT. THE PROGRAM WILL THEN GIVE THE OPERATOR THE FOLLOWING OPTIONS: 1. RETRY - ENTERING T WILL CAUSETHE PROGRAM TO RE- EXICUTE THE CURRENT TEST 2. CONTINUE - ENTERING 'T WILLCAUSETHE PROGRAM TO CONTINUE TO THE NEXT TEST 3. EXIT - ENTERING '3»' WILL CAUSE A RETURN TO THE OP- SYSTEM C. DISK DRIVE TESTING PRINTOUT FORMAT I. DRIVE MOTOR STATUS(E0X8) A.READY RESPONSE TIME = XXXX.XX MSEC B.BUSY RESPONSE TIME = XX USEC 36 April 69 68 Man Journal C.HEAD LOAD DELAY TIME = XXXX XX MSEC D.WRrrE PROTECT = Y/N E INDEX PULSECOUNT = XX S/B GT 2 THESE TESTS ARE PERFORMED BY READING THE DISK CON- TROLLER STATUS WHICH IS FOUND AT MEMORY LOCATION E0X8. - READY RESPONSE TIME IS THE TIME THAT THE READY SIGNAL TAKES TO GO HI. IT IS THE TIME REQUIRED FOR THE DRIVE MOTOR TO REACH rrs DESIGNATED SPEED. - BUSY RESPONSE TIME IS THE TIME NECESSARY FOR THE CONTROLLERTO DIGEST AN INSTRUCTION AND INDICATE IT IS BUSY. - HEAD LOAD DELAY TIME IS THE TIME ALLOTTED FOR THE HEAD TO LOADBEFORE READING OR WRITING CAN PRO- CEED. - WRITE PROTECr INDICATES IF THE DISK IS WRITE PRO- TECTED. - INDEX PULSE COUNT COUNTS THE NUMBER OF INDEX PULSES WHICH OCCUR IN THE ALLOTrED TIME SPAN. FAIL- URE INDICATES NO DISK IN DRIVE. MOTOR FAILURE OR CLUTCH FAILURE. II. DRIVE SPEED TEST PERIOD = XXXX.XX MSEC S/B 200 +/- MSEC OBTAINED WITH A PHYSICAL SWITCHPOINT MIDWAY BE- TWEEN TRACKS AND 1. V. STEP RATE TEST TRACK STEP TIME = XXXX MSEC S/B LT 40 MSEC THIS TEST FINDS THE DRIVE STEP TIME. THIS IS THE SMALLEST TIME NECESSARY FOR THE DRIVE TO STEP FROM ONE TRACK TO THE NEXT. THIS TEST IS PERFORMEDB Y STEPPINGBETWEENTRACKS 32AND0ATDIFFERENTRATES. THERATEISCONSIDEREDTO BESUCCESSFULIFTHETRACKOSWITCHHASBEENSEr. THE STEP RATE IS FOUND BY A NINE1"H ORDER BINARY SEARCH AS FOLLOWS: TIME:=DELTA:=20.48MSEC FORI = I THRU 9 DELTA := DELTA/2 IF SUCCESSFUL THEN TIME :=TIME- DELTA ELSE TIME := TIME + DELTA END IF END FOR THIS TECHNIQUE YIELDS A MEASUREMENT RANGEOF0TO4 1 MSEC WITH AN ACCURACY OF 0.08 MSEC. THIS TEST MEASURES THE ROTATIONAL SPEED OF THE DISKBYMEASURINGTHETIMEPERIODOFTHE INDEX TULSE. MISADJUSTMENT IS A COMMON CAUSE OF SOFr DISK ER- RORS. THE SPEED MAY BE ADJUSTED AND CHECKED WITH THERErRYOPriON. UI. INDEX PULSE WIDTH WIDTH = XXXX.XX MSEC S/B 2.5-5.5 MSEC THIS IS THE PULSE WIDrH OF THE SIGNAL GENERATED WHEN THE INDEX HOLE PASSES IN FRONT OF THE INDEX DETECTOR CIRCUITRY OF THE DRIVE. IV.TRACK SWITCHTEST A. STEPTIMEFROM TRACK TO SWITCH = XXXX.XX B. STEP TIME FROM TRACK 1 TO SWITCH = XXXX.XX C. SWITCHPOINT(A/A+B) = XX% S/B 5% TO 95% VI DOUBLE DENSITY STATUS(EOX4) A. RESTORE USING INTRQ(D6) G0/NG B. READ SECrOR USING DRQ(D7) GO/NG C. NO. BYTES READ = XXXX S/B 256 BYTES SOME SYSTEMS USE THE STATUS WORD AT MEMORY LOCAIION E0X4 FOR DISK READ AND WRITE OPERATIONS. THIS TEST CHECKS THE FUNCTIONALITY OF THE TWO SIG- NALS ASSOCIATED WITH THIS STATUS WORD. VII. RADIAL ALIGNMENT TEST 1. CHECK ALL TRACKS 2. ALIGN 3. EXIT 4. CHECK SINGLE TRACK 5. NEXT TEST 6. PRINT SYMBOL TABLE THIS TEST VERIFIESTHE POSITION OF THE TRACK MICRO- SWITCH IS CORRECTLY ADJUSTED. THE TRACK 0' SIGNAL STOPS THE CONTROLLER FROM STEPPING WHEN A RESTORE TO TRACK COMMAND IS ISSUED. THIS TEST DETERMINES THE TRACK O SWITCHPOINT POSI- TION BY TIMING MEASUREMENTS. TIME MEASUREMENT AREMADESTARTING FROM THE STEP COMMAND UNTILTHE SWITCH TOGGLES WHEN: A. STEPPING FROM TRACK TO I AND B. STEPPING FROM TR ACK 1 TO 0. THE SWITCHPOINT IS THE RATIO OFONE MEASUREMENT TO THE SUM OF THE MEASUREMENTS (IE A/A+B). FOR DRIVES USING PHOTO DIODE POSITION DETECTOR SWITCHES. THIS TIMING TEST METHOD ACCURATELY RE- FLECTS THE PHYSICAL SWITCHPOINT POSITION. MEASURE- MENI"S OF 40-60% ARE TYPICAL FOR THIS TYPE OF DRIVE. FOR DRIVES WITH A MECHANICAL SWITCH. THE MEAS- UREMENI - IS INFLUENCED BY SWITCH SPRING TENSION. TYPICAL MEASUREMENTS OF 5 20% MAY BE -CHECK ALLTRACKSOR A SINGLE TRACK PRINrS. FOR EACH SECTOR. AN •■ FOR 256 BYTE SECTORS OR V FOR 128 BYTE SECTORS. THE ERROR CODE IS PRINTED FOR A SECTOR WITH AN ERROR. THE ERROR CODES MAY BE PRINTED BY ENTER- ING OPHON '6\ THETRACKS ARETESTEDOUTOFORDER.TO TEST FOR STEPPER MOTOR SKIP ANDDIRECHONALOFFSErS. - SECTOR SYMBOL TABLE: • 256 BYTE SECTOR OK . 128BYTESECTOROK U 0-127 BYTES READ INTO BUFFER. STATUS OK B 129255 B YTES READ INTO BUFFER. STATUS OK O OVER 256 B YTES READ INTO BUFFER. STATUS OK T TRACK NO ERROR. ON WRONG TRACK N NO DATA ON TRACK OR UNFORMATrED TRACK S MISSING SECTOR C ID FIELD CRC ERROR A ADDRESS MARK ERROR IN DATA FIELD D DATA FIELD CRC ERROR 63 Mcro Journal April '69 37 - ALIGNMENT MAY BE PERFORMED WITH EITHER: AN OP- SYSTEM DISK COR DISK FROM ALIGNED DRIVE) OR WITH A DYSAN DIGITAL DIAGNOSTIC DISK. - USING THE OP-SYSTEM DISK: I ADJUSTTHE STEPPER MOTOR CLOCK WISE UNTIL ABOUT HALFOFTHESECrORS HAVE ERRORS AND MARKTHIS POSI- TION. 2. ADJUST THE STEPPER MOTOR COUNTER CLOCKWISE UNHLHALFOFTHE SECTORS HAVEERRORSANDMARKTHIS POSITION. 3. CENTER THE STEPPER MOTOR BETWEEN THE TWO MARKED POSmONS. - USING THE DIGITAL DIAGNOSTIC DISKETIE(DDD) A. REDUCTION OF SIGN AL STRENGTH TEST 1. 5/12 AMPLITUDE - ALTERNATE SECTORS OFF SET BY 7 MILS 2. 4/12 AMPLITUDE - ALTERNATE SECTORS OFF SETBY8MILS 3. 3/12 AMPLITUDE • ALTERNATE SECTORS OFF SETBY9MILS THE HEAD PICKS UP A SIGNAL IN A 12 MIL WIDETRACK ON THE DISK. ALTERNATESECTORS ARE DISPLACED BY7. 8 AND 9 MILS ON TRACKS 21.24 AND 27 RESPECTIVELY TO PRODUCE AN EFFECTIVE REDUCTION IN SIGNAL STRENGTH. THIS TEST CHECKS FOR WEAK DRIVE ELECTRONICS, HUB CENTERING, STEPPER PLAY, NEED FOR HEAD CLEANING AND MISAD- JUSTED HEAD LOAD PADS. VIII. READ/WRITE SECTOR TEST 1. PATTERN 1 2. PATTERN 2 3. MODIFY BUFFER(M) 4. EXIMINEBUFFER(E) 5. READ SECTOR(R) 6. WRITE SECTOR(W) 7. NEXT SECTOR(N) 8. RETURN TO FLEX(F) 9. HELP(H) - PATTERN 1 INITIALIZES THE BUFFER TO HEXADECIMAL BDBDBD... - PATIERN 2 INITIALIZES THE BUFFER TO HEXIDECIMAL 242424... - MODIFY BUFFER ALLOWS EDITING OF THE SECTOR BUFFER USING: N = NEXTBYTE L = LAST BYTE RTN = EXIT MODIFY - EXIMINE BUFFER PRINTS THE 256 BYTE SECTOR BUFFER IN A HEXIDECIMAL AND ASCII FORMAT - READ SECTOR READS THE SELECTED SECTOR INTO THE SECTOR BUFFER AND DISPLAYS THE SECTOR BUFFER - WRITE SECTOR WRITES THE CURRENT CONTENTS OF THE SECTOR BUFFER INTO THE SELECTED SECTOR - NEXT SECTOR READS THE NEXT SECTOR IN THE FILECHAIN • RETURN TO FLEX ENDS THE PROGRAM - HELP DISPLAYS ALL TEST VIII TEST OPTIONS B. TRACK OFFSET IN MILLI-INCHES TRY STEP DIRN=IN STEP DIRN=OUT LIMITS GO/NG 1. +XX.XMILS 2. +XX.X MILS 3. +XX.XMILS 4. +XX.XMILS 5. +XX.X MILS +XX.X MIIS +XXXMILS +XX X MILS +XX.X MILS +XX.X MILS +/-1.5 GO +/-1.5 GO +/-1.5 GO +/-1.5 GO +/-1.5 GO •NOTE-SOURCE PROGR AM MODIFICATIONS ARE NECESSARY TO TEST 8 INCH DISK DRIVES. •DIGITAL DIAGNOSTIC DISKETTE' IS A TRADEMARK AND •DYSAN' IS A REGISTERED TRADEMARK OF DYSAN CORP. 1244 REAMWOOD AVE. SUNNYVALE, CA, 94086 : 408-734-1634; 800551-9000 THERE ARE SIX TRACKS AVAILABLE TO TEST RADIAL ALIGNMENT. THESE ARE TRACKS 0, 5, 16, 19, 30 AND 39. THIS TEST CHECKS THE SELECTED TRACK WITH PROGRESSIVE OFFSETS AND CALCULATES THE DISTANCE THE HEAD IS OFF OF CENTER. THIS IS DONE FROM TWO DIRECTIONS IN ORDER TO CHECK FOR H YS1ERISIS OR STEP DIRECTION DEPENDENT PLAY.THE MEASUREMENT RANGE IS -6 TO +6 MILS WITH AN ACCURACY OF +/0.5 MILS. C. AZIM UTHAL HEAD ANGLE TEST HEAD ANGLE = XX.X MIN S/B +/- 12 MIN THE DISK HEAD READS AND WRTTES ON THE RADIAL CENTERLTNE OF THE TRACK. THIS TEST MEASURES THE SKEW ANGLE OF THE HEAD RELATIVE TO THE TRACK CEN- TERLINE BY TESTING SECTORS WITH PROGRESSIVE ANGLE OFFSETS ON TRACK NO. 34. MEASUREMENT RANGE IS -18 TO +18 MINUTES WITH AN ACCURACY OF +/-1.5 MINUTES. DYSAN DIGITAL DIAGNOSTIC DISKS PART NUMBERS: 5.25" 48TP1 SSSD(*508-100) SS()D(#5<»-200) DSSD(#50g-30Q) DSDIW508- 4(10) 5.25" 96TPI SSSD(#506.I(X>) SSI>D<»506-200) DSSD(#506-30Q) DSDD(»506- 400) 8.00" 48TPI SSSD(#808-IOO) SSDD(#8O8-200) DSSD(*808-300) DSDD<«808- 400) FLEX 1 IS A TRADEMARK OFTECHNICAL SYSTEMS CONSULTANTS EOF 38 Apnl'89 68 Micro Journal DDT DISK DRIVE AND CONTROLLER TEST PROGRAM (5.25 DRIVES, DC-1 TO DC-4 CONTROLLERS) THIS PROGRAM TESTS SINGLE OR DOUBLE DENS DISKS OPERATE ON THE SPECIFIED DRIVE. THIS PROGRAM ALLOWS FOR TRK ADJUSTMENT. ROTATIONAL SPEED ADJUSTMENT AND RADIAL ALIGNMENT USING A FORMATTED DISK FROM AN ALIGNED DRIVE OR A DIGITAL DIAGNOSTIC DSK EXAMPLE: ODT.D DDT DDT > FILE WRITTEN BY: (TESTS DRIVE 0) (ALLOWS CHOICE OF CONTROLLER I/O SLOT) (BUILDS PRINT FILE) EMERY KORP1 40 BEATRICE AVE SYOSSET.NY 11791 VERSION-2 ; DATE-8/87 REFRENCES €502 ASSEMBLY LANGUAGE SUBROU- TINES BY LANCE A. LEVENTHAL AND WINTHROP SAVILLE PUB. OSBORNE/MCGRAW-HILL 1982 THE FLEX DISK OPERATING SYSTEM BY TECHNICAL SYSTEMS CONSULTANTS, INC OR. COBB'S JOURNAL PEOPLE'S COMPUTER COMPANY VOLUME 8, ISSUE 12, DECEMBER 1983 68' MICRO JOURNAL COMPUTER PUBLISHING INC. 2/84, 6/84, 6/84 ISSUES NOTE: TO CUSTOMIZE THE DEFAULT MODE FOR YOUR SYSTEM, CHANGE THE SYSTEM DEPENDENT VARIABLES. * * • WARMS • FLEX LOCATIONS * EQU SCD03 SBUG EQU SF814 FMS EQU SD406 FMSCLS EQU SD403 FCB EOU SC840 RPTERR EQU SCD3F GETFIL EQU SCD2D SETEXT EOU SCD33 TTYEOL EQU SCC03 NXTCH EQU SCD27 LNPNTR EQU SCC14 MB MEND EQU SCC2B PSTR EQU SCD1E GETCHR EQU SCD15 DRVTST ORG $1000 BRA START0 VK FCB 2 VERSION START0 LBRA START • • • TRMSTA * SYS DEPEND VARS. * EQU SE004 TERMINAL STATUS TRMDAT EQU $B005 TERMINAL DATA DRVADD FDB SE010 DISK CONTROL A0DR CLKFLG FCB 0-lMHZ, 1-2MHZ T7SECT FCC SOA ,$14, $12, $24 * • * SSSD.DSSD.SSDD.DSDD SECTORS * PROGRAM VARIABLES * PADDR FDB $3000 PRTFLG FCB SAVLNP RMB 1 DRVNO FCB 1 DRVCMD FCB 1 CNTR FCB 1 DIGNO RMB 3 COUNT RMB 3 TEMP FCB 1 TEMP2 FCB 1 TEMP3 FCB 1 TIME RMB 2 BLSB RMB 2 BINARY LSB DMSB RMB 2 DECIMAL MSB BNCTIM FCB 1 BOUNCE TIME ERRFLG FCB 1 ERROR FLAG RETRY RMB 2 RETEST ADDR DVR RMB 3 DIVISOR DVD RMB 3 DIVIDEND T6FLG RMB 1 TEST 6 PASSED DDFLG RMB 1 TEST 6 GO FLG MAXSEC RMB 1 1 SECS/TRK FORMAT RMB 1 DISK FORMAT T70PTN RMB 1 T7 OPTION DNSFLG RMB 1 1-DD TRK RMB 1 TRACK NO SEC RMB 1 SECTOR NO ADDR RMB 2 MEMORY ADDRESS LIMT RMB 2 LO LIMIT STPSIZ RMB 2 CONTROLLER STEP SIZE STPREF RMB 2 TIME TO TRKD SWITCH STPCMP RMB 2 T5 MEASUREMENT RSLTN RMB 1 OFFSET RESOLUTION T7CNT RMB 1 LOOP COUNTER PCNT RMB 1 POSITIVE OFFSET NCNT RMB 1 NEGATIVE OFFSET LENGTH RMB 1 BYTES TO DIV LCOUNT RMB 2 HIDE1 RMB 3 HI0E2 RMB 3 RMOPTR RMB 2 REMAINDER ADDR TMPPTR RMB 2 TEMP ADDR DVRPTR RMB 2 DVR POINTER DVDPTR RMB 2 DVR POINTER • • • • PROGRAM MESSAGES * TITLE FCC "DISK DRIVE TEST PROGRAM IREV-8/87] ' , 4 USGERR FCC 'Usage: DDT (, drive) ( > print file) ', 4 SLTMSG FCC 'DISK CONTROLLER IN SLOT (0-7, 1-STD) 7 ',4 SLTERR FCC 'INVALID DISK ADDRESS, RETRY(YZN)? ',4 DRVMSG FCC 'DRIVE NO. (0-3)? ',4 CLKMSG FCC 'SYSTEM CLOCK RATE (1-1MHZ. 2-2MHZ) 7 ',4 HLTMSG FCC 'HALT AFTER EACH TEST (Y/N) 7 ',4, DSKMS1 FCC 'INSERT DISK IN DRIVE ' DSKMSH FCC ' , HIT RTN TO CONTINUE' , 4 68 Micro Journal April '88 39 T1MSG FCC T1AMSG FCC T1AMSH FCC T1BMSG FCC T1BMSH FCC T1CMSG FCC T1CMSH FCC T1DMSG FCC T1DMSH FCC T1EMSG FCC T1EMSH FCC T2MSG FCC 72AMSG FCC T2AMSH FCC OPTMSG FCC \4 T3MSG FCC T3AMSG FCC T3AMSH FCC NUL GTHN T4MSG T4AERR FCC TIMES', 4 T4AM1 T4AH3 T4BM1 T4BM3 T4CM1 T4CH2 RESMSG FCC T5MSG FCC T5MSH T5MSI T5EHSG FCC RATBL FCC T6MSG TfiMSl T6MS2 T6MS3 T SMS 3 A FCC INTDAT FCC NGO OK T7MSG T7MSG1 FCC 3. EXIT ' T7MSH1 FCC TEST ' T7MSH2 FCC T7FORM FCC T7SYTB FCC SECTOR' T7SYTC FCC T7SYTD FCC T7SYTE FCC T7SYTF FCC T7SYTG FCC T7SYTH FCC T7SYTI FCC T7SYTJ FCC T7SYTK FCC T7SYTL FCC T7TRK FCC T7TNO FCC T7CHSG FCC UNTIL' T7CM5H FCC T7CMS1 FCC CLOCKWISE' T7CKS2 FCC T7CMS3 FCC T7CMS4 FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC FCC SOD.SOA, 'I. DRIVE MOTOR STATUS (E0XB» ' , 4 ' A. READY RESPONSE TIME - ' > 0000.00 MSEC S/B L? 620 MSEC , 4 ' B. BUSY RESPONSE TIME - ' 0000', 0,' 00 USEC S/B LT 34 USEC , 4 ' C. HEAD LOAD DELAY TIME - ' ' 0000.00 MSEC S/B L? 200 MSEC , 4 D. WRITE PROTECT - ' ' \4 E. INDEX PULSE COUNT - • • S/B GT 2', 4 $0D,S0A, 'II. DRIVE SPEED TEST' , 4 • PERIOD • ' '0000.00 S/B 200 + /-1 MSEC, 4 $0D,S0A, ' OPTIONS: 1. RETRY 2. CONTINUE T7CMS5 FCC 'SWITCH POINT S/B * T7CMS6 FCC '0000' , $00, ' 00%' , 4 T7MS2 FCC SOD, S0A, 'ENTER DISK TYPE' , SOD, S0A, ' SYSTEM DISK' T7MS2A FCC SOD, S0A, ' T7MS3 FCC 'WHICH SIDE T7MS4 FCC S0D.S0A, 'A. TESTS' ,$0D,$0A 2. DIGITAL DIAGNOSTIC DISK (0 OR 1)7 ',4 REDUCTION OF SIGNAL STRENGTH OP- ',4 1. 5/12 AMPLITUDE - ALTERNATE SECTORS MILS', 4 4/12 AMPLITUDE - ALTERNATE SECTORS SOD.SOA, 'III. INDEX PULSE WIDTH' , 4 ' WIDTH - ' '0000.00 MSEC S/B 2.5-5.5 MSEC , 4 0,0,0 'GT • $0D,$0A, 'IV. TRACK SWITCH TEST' , 4 'TRK SWITCH FAULT; SHOULD TOGGLE GT 5 3/12 AMPLITUDE - ALTERNATE SECTORS LIMITS A. TIME FROM TRACK TO SWITCH - > ' 0000.00 MSEC ,$0D,$0A B. TIME FROM TRACK 1 TO SWITCH - ' ' 0000.00 MSEC, SOD, S0A C. SWITCH POINTU/A+B) - ' '0000', $00, '00% S/B 5-95%', 4 BUSY STUCK HI AFTER RESTORE' , 4 SOD.SOA, 'V. TRACK STEP TIME TEST' , 4 STEP TIME - ' '0000.00 MSEC S/B LT 40 MSEC , 4 ' SEEK TO TRACK TIMEOUT' , 4 $0F, SAO, $07, $D0, $03, $E8, $01, $F4, $00, $FA $0D,$0A, 'VI. DOUBLE DENSITY STATUS (£0X4) ' A. RESTORE USING INTRQID6) - ',4 B. READ SECTOR USING DRO(D7) - >, 4 ' C. NO. BYTES READ - ' '0000' ,$00, '00 S/B 256 BYTES', 4 '101' 'NG' 'OK' $0D,$0A, 'VII. RADIAL ALIGNMENT TEST' 'ENTER-1. CHECK ALL TRACKS 2. ALIGN T7MS5 FCC OFFSET' T7MS5A FCC ' BY 7 T7MS6 FCC ' 2. OFFSET BY 8 ' T7MS6A FCC 'MILS' , 3.EXITI7MS7 FCC ' 3. OFFSET BY 9 ' 77MS7A FCC 'MILS', 4 T7MSB FCC $0D,$0A, 'B. TRACK OFFSET IN MILLI- INCHES(MILS)' ,4 T7MS9 FCC 'TRACK NO. (00,05,16, 19,30,39) 7 ',4 T7MS10 FCC 'TRY STEP DIRN-IN STEP DIRN-OUT GO/NG' T7MS1Z FCC $0D,$0A T7MS1A FCC T7MS11 FCC T7MS12 FCC ' *XX.X ',4 T7MS13 FCC 'MILS ',4 T7MS14 FCC T7MS15 FCC T7MS16 FCC T7MS17 FCC T7MS18 FCC T8PROM FCC T8MSG T8SYM2 — ',4 ' X. ' *XX.X 'MILS GO' , $0D,$0A, T8SYM 4 T80PS C. AZIMUTHAL HEAD ANGLE TEST' , 4 HEAD ANGLE -' , 4 MIN S/B ♦/- 12 MIN',4 -1.5 TO +1.5* ,4 • ',4 FCC SOD, $0A, 'VIII. READ/WRITE SECTOR TEST' , 4 FCC '0123456789ABCDEFNL' , SOD, 4 FCC ' 12MERWNFHB' ,4 FCC 'ENTER OPTION: 1-PATTERN 1 2-PATTERN 2' M-MODIFY PATTERN E-EXAMINE TBOPS1 FCC PATTERN' , SOD, $0A, ' R-READ' ' SECTOR W-WRITE SECTOR N-NEXT SECTOR ' H-HELP B-BACK TO TEST VII', 4 'BYTE NO. (00-FF) ? ', 4 'CHANGE ',4 'OPTIONS: N-NEXT BYTE L-LAST BYTE RTN- $0D,$0A,' 4. CHECK SINGLE TRACK 5. NEXT '6. PRINT SYMBOLS ',4 'ENTER FORMAT-l.SSSD 2.DSSD 3.SSDD 4.DSDD 'SYMBOL TABLE! ' , SOD, $0A, * $0D,$0A,' , - VALID 128 BYTE SECTOR' $0D,$0A, ' T - TRACK NUMBER ERROR' $0D,$0A, ' C - ID CRC ERROR' S0D,$0A, ' N - NO DATA ON TRACK' $0D,$0A, ' S - MISSING SECTOR' $0D,$0A, ' A - NO DATA ADDR MARK' $0D,$0A, ' D - DATA CRC ERROR' $0D,$0A, ' O - OVER 256 BYTES' $0D,$0A, ' B - BETWEEN 129-255 BYTES' $0D,$0A, ' U - UNDER 128 BYTES' , 4 TRACK (00-39) 7 ',4 'TRACK 00 ',4 $0D,$0A, 'A. ADJUST STEPPER MOTOR CLOCKWISE ' HALF OF THE SECTORS HAVE ERRORS' , 4 SOD.SOA, 'B. ADJUST STEPPER MOTOR COUNTER- ' UNTIL HALF OF THE SECTORS HAVE ERRORS' , 4 $0D,$0A, 'C. ADJUST STEPPER MOTOR SO IT IS' ' CENTERED BETWEEN POINTS A. AND B.',$0D,$0A T80PS2 FCC FLEX' TBOPS3 FCC TBMSG1 FCC TBMSG2 FCC $0D,$0A TBMSG3 FCC EXIT', 4 T8MSG4 FCC TBMSG5 FCC TBMSHS FCC T8MSG6 FCC «0 00 00 00' - VALID 256 BYTZ8MSH6 FCC ' AAAAAAAAAAAAAAJU ' , 4 T8MSG7 FCC SOD, S0A, • SECTOR (01-36) ? TBMSG8 FCC 'SECTOR READ ERROR' , 4 •BYTE 00 •BYTE: ' IB »C •X0-XF 00 ',4 10 II 12 13 14 15 16 17 18 19 »A' 10 «E IF ASCII', 4 00 00 00 00 00 00 00 00 00 00 00 00 T8MSG9 FCC T8MS10 FCC TBMSll FCC TBMS12 FCC T8MS13 FCC T8MS14 FCC 'WRITE PROTECT ON' , 4 'WRITE ERROR', 4 •SECTOR VERIFIED' , 4 'TRACK 00 SECTOR 00' 'AK£ YOU SURE1Y/N)? ' ■SECTOR OUT OF RANGE' BFULL FCC $0D,$0A, 'BUFFER FULL-SAVING' , 4 Aprt'M UMcroJttaTMl ■ « • PROGRAM SUBROUTINES • •INCHNE-INPUT CHARACTER, NO ECHO INCHNE LOA TRMSTA TERMINAL STATUS BITA • $01 DATA7 BEQ INCHNE NO LDA TRMDAT GET CHARACTER ANDA IS7F RTS INCH LDB TRMSTA BITB • 501 BEQ INCH LDA TRMDAT ANDA • $7F JSR OUTCH RTS OUTCH PSHS B,Y LDB PRTFLG BEQ OUTCH1 LDY PADDR CMPA ISO A BEQ OUTCH1 CMPY MEMEND BUFF FULL? BNE OUTCHO PULS B, Y,X LDX IBFULL JSR PSTR JMP EXTO OUTCHO STA ,* + STY PADDR OUTCH 1 LDB TRMSTA BITB IS02 BEQ OUTCH1 STA TRMOAT PULS B,Y,PC •FILCHK-PUTS FILE NAME INTS FCB FILCHK LDX LNPNTR SAV FIL NAM STX SAVLNP LDX • FCB JSR GETFIL BCS FILERR LDA 11 FCB<.TXT JSR SETEXT INC PRTFLG RTS * FILERR PULS X JSR RPTERR JSR FMSCLS JMP WARMS •CHKIN -CHECK CHARICTER IS IN RANGE, ECHO CHKIN STA HI + 1 1 JPPER LIMIT STB LO+1 LOWER LIMIT CHK1 JSR INCHNE GET CHARACTER LO CMPA • SFO COMPARE UPPER BLT CHKER HI CMPA • SOD COMPARE LOWER BGT CHKER BRA CHKOK CHKER LDA • 507 BELL JSR OUTCH BRA CHK1 CHKOK JSR RTS OUTCH OUTPUT CHARACTER •INCMP -INPUT CHAR AND COMPARE TO CHARS INCMP CLR TEMP CHR POINTER JSR INCHNE GET CHAR INCMP1 CMPA ,x CHAR? BEQ INCMP3 INC TEMP LDB , X* EOF? CMPB • $04 BNE INCMP1 LOOP INCMP2 LDA • 50? BELL INCMP 3 LDB TEMP JSR OUTCH PRINT RTS •HEXCON-CONVERTS *A" REG TO ASCII •MSB IN M(U) , INC U: LSB IN M(U), INC HEXCON PSHS A STORE BYTE ASRA GET MSN ASRA ASRA ASRA BSR HEXPUT CONV TO ASCII STA ,u+ STORE MSB CHAR PULS A READ BYTE BSR HEXPUT CONV TO ASCII STA ,u+ STORE LSB CHAR RTS HEXPUT ANDA l$DF ADDA • $30 CMPA • $39 BLS OP I ADDA • $07 CONV TO LETTER OP1 RTS •BDCON-BINARY TO DECIMAL CONV. •X-BINARY MSB,U>DECIMAL LSB •B-NO. OF BINARY BYTES ON ENTRY •B-NO. OF DECIMAL BYTES ON RETURN BDCON BRA BDS1 BBYTES RMB 1 NO OF BINARY BYTES DLSB RM8 2 LOCATION OF BCD LSB BBCNTR RMB 1 L2 COUNTER DBYTES RMB 1 NO OF DECIMAL BYTES CCREG RMB 1 TEMP CONDITION CODE L3CNTR RMB 1 LOOP COUNTER * BDS1 LOA • $01 STA DBYTES SET NO DEC BYTES STB BBYTES SET NO BIN BYTES STU DLSB DECIMAL LSB POINTER CLR , U LI LDA l$08 NO BITS PER BYTE STA BBCNTR SET COUNTER L2 LDA LSLA ,x LOAD BIN BYTE GET MS BIT STA ,x SAVE BIN BYTE LDU DLSB SET U TO DLSB LOB l$00 CLEAR L3 CNTH STB L3CNTR L3 LDA .u GET DEC BYTE ADCA ,U C+2 (DEC BYTE) TFR CC,B STORE CARRY STB CCREG DAA DEC ADJUST STA .U STORE RESULT TFR CC,B ORB CCREG ANY CARRIES? LEAU -1,U POINT TO NEXT DBYTE INC L3CNTR INC COUNTER LDA L3CNTR CMPA DBYTES DECIMAL MS BYTE? BNE OUT3 ANDB • $01 CARRY? BEQ L2END NO INC DBYTES NEED MORE SPACE CLR ,U CLEAR SPACE STARTING AT IX) OUT3 TFR B,CC RESTORE CARRY BRA L3 L2END DEC BNE BBCNTR L2 LOOP COUNTER L3END DEC BEQ LEAX BBYTES FIN 1,X GET NEXT BIN BYTE 66 Mere Journal Aprt -as 41 BRA LI FIN LDB DBYTES RTS •DBCON-DECIMAL TO BINARY CONV . •X-DECIMAL MSB, U-BIN LSB •B-NO OF DEC BYTE-ON ENTRY •B-NO DF BIN BYTES ON RETURN DBCON LDA IS01 INIT BIN BYTES STA BBYTES STB DBYTES NO DEC BYTES STU BLSB STX DMSB DBL1 LDA IS08 HO OF BITS STA BBCNTR DBL2 ANDCC »S00 CLR CARRY LDB DBYTES DBL3 JSR DBSHR RT SHFT LDA ,U GET BIN NO RORA STA , U LDX DMSB DEC BBCNTR BNE DBL2 NOP DBL4 LEAX 1,X INC X STX DMSB DEC DBYTES NEXT BYTE? BEQ DBOUT FINISH LDA ,X GET BYTE BED DBL4 ZERO? LEAU -1,0 NEXT BIN BYTE INC BBYTES BRA DBL1 SHIFT DEC DBOUT LDB BBYTES RTS •DEC AD J- IF BCD CHR GE 6, ADD 3 DECADJ PSHS A ANDA ISOF LS ND CMPA »$08 BLO DCAS1 SUBA »$03 DCAS1 STA TEMP PULS A ANDA »SFD MS NO CMPA #$80 BLO DCAS2 SUBA »S30 DCAS2 ADDA TEMP RTS •DBSHR-ROTATE LS BIT INTO CARRY DBSHR LDA , X GET DEC BYTE RORA PSHS CC CARRY JSR DECADJ STA ,X+ RESTORE DEC PULS CC DECB BIT CNTR BNE DBSHR RTS •GETDEC-TAKBS TMO KEYBD INPUTS, CONV TO BIN * IN A GETDEC LDD #$3930 GET MSB JSR CHKIN ASLA PUT IN MSB ASLA ASLA ASLA STA TEMP LDD #$3930 GET LSB JSR CHXIN ANDA #$0F ZERO MSB ORA TEMP STA TEMP LDU ITEMP2 BIN I LDX |TEMP DEC I LDB #$01 CONV 1 BYTE JSR DBCON LDA TEMP 2 RTS •XFER-TRANSFER DATA; A-l BYTES, •X-START OF SOURCE DATA, U-START OF DESTINATION XFER LDB ,X+ GET CHAR STB , U + STORE CHAR DECA COUNTER BNE XFER RTS •TIMER-20USEC TIMING LOOP TO COUNT UNTIL •BIT IN 'A' IS LO IB-27) OR HI IB-26) •RESULTS STORED AT COUNT+1 TIMER STA BIT + 1 TIMED BIT LOCN STB BRHILO TIMEOUT ON 1/0 LDX ($0001 INIT COUNTER TCMD LDA l$0F RESTORE STA 8,Y CMD REG TLOOP LDB 8,Y STATUS BIT BITB l$00 CHK BIT BRHILO BRA TIMOUT COMPARE NOP DELAY LEAX 1,X INC CNTR BNE TLOOP LOOP MAXTIM LEAX -1,X REMOVE OFFSET STX COUNT+1 STORE RESULT LDX IGTHN GET UPDATE BRA TCI CONTINUE TIMOUT LEAX -1,X REMOVE OFFSET STX COUNT+1 STORE RESULT LDX INUL GET UPDATE TCI LDA #$03 NO OF BYTES JSR XFER UPDATE MSG LDA #$O0 INTERRUPT CMD STA 8,Y CMD REG JSR DELTS 1/10 SEC DEL RTS •UPDMSG-UPDATE MESSAGE; TAKES DATA IN •COUNT, CONVERTS TO DECIMAL DATA IN MSG UPDMSG PSHS U TEMP STORE U LDA #$00 CLR SCRATCH STA COUNT BIN MS BYTE STA DIGNO DECIMAL I STA DIGNO+1 ANDCC #$00 CLR CARRY LDX (COUNT BINI SETUP LDU #DIGNO+2 DEC( SETUP LDB #$03 NO OF BYTES JSR BDCON CONVERT PULS U RESTORE MSG PNTR LDA DIGNO GET MSB JSR HEXCON CON+UPDATE LDA DIGNO+1 GET NEXT BYTE JSR HEXCON CON+UPDATE LEAU 1,U SKIP DEC PT LDA DIGNO+2 GET LAST BYTE JSR HEXCON CON+UPDATE RTS •CLKCMP-COMPENSATION FOR CLOCK RATE CLKCMP LDA #$00 CLR SCRATCH STA COUNT CLR MS BYTE LDA CLKFLG BNE CLKC1 SKIP IF 2MHZ ANDCC #$00 CLR CARRY ROL COUNT+2 CNT:-2<CNT) ROL COUNT+1 ROL COUNT CLKC1 RTS •PASFAL-CHECK TEST LIMITS, SET ERRFIC IF NG PASFAL CMPD COUNT+1 HI LIM OK? BHS PSF1 OK Aprf'88 68 Mcro Journal INC ERRFLG ERR STA OPTDIR+1 JUMP TABLE PSF1 LDD LIMT GET LO LIMIT OPTDIR BRA OPTDIR JUMP CMPD COUNT+1 LO LIM OK? BRA OPT1 BLS PSF2 OK BRA OPT2 INC ERRFLG ERR BRA OPT3 PSF2 LOA COUNT MSB-OT OPT1 PULS X BEQ PSF3 OK LDA ISOD INC ERRFLG ERP JSR OUTCH PSF3 RTS LDB B, Y •TESTOK-CHECK IF TEST GO, OPTIONS IF NOGO JSR DELHS TESTOK JSR PSTRNG PiUNT LDX RETRY LOA ERRFLG ERROR? JMP ,x BEQ OKOUT OPT2 CLR ERRFLG JSR OPTION RTS OKOUT RTS OPT3 PULS X CLR RTS ADDR •DELAY -56USEC DELAY JSR PCRLF DELAY LBSR DEL2B *. .PROGRAM EXIT DEL2B LBSR DEL14 EXIT JMP EXTO DELI 4 LBSR DEL3 EXTO TST PRTFLG DEL3 RTS BEQ EXT2 •DELHS -HALF SECOND DELAY LDX IFCB DELHS PSHS D LDY SAVLNP FILE NAME JSR DELTS STY LNPNTR JSR DELTS JSR GETFIL JSR DELTS BCS OPTERR LDA ISO? LDA 11 DELL1 LOB B,Y READY? JSR SETEXT BITS IS80 LDA 12 OPN TO MR BEQ DELOT STA ,X JSR DELTS JSR FMS DECA BEQ EXT1 BNE DELL1 LDA 1,X IF FIL EXISTS! DELOT PULS RTS D CMPA BNE 13 OPTERR •DELTS -TENTH SECOND DELAY LDA 112 DELETE FILE DELTS PSHS X STORE X STA O.X tst CLKFLG » JSR FMS BNE DlT BNE OPTERR LOX IS3333 INIT CNTR LDY SAVLNP RESTOR FIL NAM BRA D2T STY LNPNTR D1T LDX IS6666 JSR GETFIL D2T LEAX -l.X DEC CNTR BCS OPTERR BNE D2T LOOP LDA 11 PULS X RESTORE X JSR SETEXT FCB<.TXT RTS LDA 12 OPN TO WRITE) •PCRLF -CARRAGE RTN.LINE FEED STA 0,X PCRLF LDA ISOA JSR FMS JSR OITCH BNE OPTERR LDA ISOD EXT1 LDA ISOO , JSR OITCH STA ,x RTS LDY IS3000 •PSTRNG-PRINT AND SET ERRFLG IF KEYBOARD ENTRY EXTLP LDA . Y + PSTRNG LDA TRMSTA TERMINAL STAT JSR FMS BITA fSOl DATA? BNE OPTERR BEQ PSTO NO CMPY PADDR INC ERRFLG BLO EXTLP PSTO JSR PCRLF LDA ISOD PDATA LDA ,X+ GET CHAR STA ,Y CMPA IS04 EOF? CMPY PADDR BEQ PST2 BLS EXTLP CMPA ISOO NOP? LOA 14 CLOSE BEQ PDATA STA ,X JSR OUTCH PRINT JSR FMS BRA PDATA BNE OPTERR PST2 RTS EXT2 JMP WARMS •OPTION-TEST OPTIONS ; U-RETRY ADDRESS OPTERR JSR RPTERR OPTION LDX tOPTMSG OPTIONS JSR FMSCLS JSR PDATA PRINT LDX ISLTERR+22 LDD »S333l SET LIMITS JSR PSTR JSR CHKIN GET CHAR JSR GETCHR ANDA ISOF CONV TO DEC CMPA 1* Y SUBA IS01 SUBT 1 LBEQ EXIT ASLA MULT BY 2 BRA EXT2 6S Moo Joumri April 'M 43 *HUL3X1-C0UNT:-CNTR X DVD MUL3X1 LDU ICOUNT+3 LDX fDVDO CLR TEMP CLR TEMP + 1 JSR MUL1 JSR MUL1 STA COUNT RTS HUL1 LDA , -X LDB CNTR MUL ADDD TEMP STB , -U STA TEMP+1 RTS •MBDIV-MULTIBYTE DIVIDE DVD :- DVD/DVR •DATA FORMAT - MSB AT LOWEST ADDR •CARRY-1 FOR ERROR, -0 FOR VALID RESULT •SET 'LENGTH' TO NUMBER OF DATA BYTES •RMDPTR POINTS TO LSB OF REMAINDER. MBDIV PSHS X,Y,U LDA LENGTH BNE DINIT LBRA EREXIT •LCOUNT :• 8 (LENGTH) +1 DINIT LDB IS08 MUL ADDD IS0001 STD LCOUNT •CLEAR ARRAYS i SET LSB POINTERS LDA LENGTH LDY IHIDE1 LDU IHIDE2 ZAIL CLR ,Y + CLR ,U+ DECA BNE ZAIL STY RMDPTR STU TMPPTR •SET LSB POINTERS DVRPTR, DVDPTR LDD IDVD AODB LENGTH ADCA 1500 STD DVDPTR LDD IDVR ADDB LENGTH ADCA 1500 STD DVRPTR •DIVISOR-0? LDA LENGTH LDX DVRPTR CHKL1 LDB , -X BNE DIV1 DECA BNE CHKL1 LBRA EREXIT •DIVIDE USING TRIAL SUBTRACTION DIV1 ANDCC ISFE CLR C DLOOP LDA LENGTH LDX DVDPTR SLLP1 LDB , -X ROLB STB , X DECA BNE SLLP1 •DECR BIT CNTR AND EXIT IF DONE •NO CHANGE TO CARRY L0X LCOUNT LEAX -1,X BEQ OKEXIT STX LCOUNT •SHIFT THE CARRY INTO THE LS BIT OF THE •UPPER DIVIDEND LDA LENGTH LDX RMDPTR SLLP2 LDB ,-X ROLB STB , X DECA BNE SLLP2 •SUBTRACT DVR FROM REMAINDER PLACING THE •DIFFERENCE INTO TEMP STORAGE LDA LENGTH ANDCC ISFE CLR C LDX DVRPTR SUBLP LDB ,-Y SBCB , -X STB . -U DECA BNE SUBLP TFR CC,A COMA TFR A,CC •IF CARR* -0, SUBTRACT NOT SUCCESSFULL, •NEXT BIT OF QUOTIENT IS 0. •IF CARRY=1, SUBTRACTION SUCCESSFULL, •NEXT BIT OF QUOTIENT IS 1, •SET HDEPTR TO NEW REMAINDER. LDY RMDPTR LDU TMPPTR BCC DLOOP EXG Y,U STY RMDPTR STU TMPPTR LBRA DLOOP OKEXIT ANDCC ISFE CLR C BRA DEXIT EREXIT ORCC IS01 SET C DEXIT PULS X,Y,U,PC •RESTOR-RETURN TO TRACK RESTOR LDA 1503 SLOW RESTORE LDX ISOOOO 5 SEC TIMOUT STA 6.Y CMD REG RESLP JSR DELAY 28 USEC LDB 8,Y STATUS BITB 1501 BUSY? BEQ RESOUT NOT BUSY LEAX 1,X COUNT BNE RESLP BUSY LP RESERR LDX RESMSG TIMEOUT JSR PSTRNG PRINT JSR OPTION OPER OPS RESOUT RTS •RDDATA-READ ONE SECTOR OF DATA RDDATA LDB TRK TRACK-0? BEQ RDDATO SINGLE DEN LDB DNSFLG BEQ RDDATO ORA IS02 DOUBLE DEN RDDATO LDB T6FLG IF T6 USE E0X4 BEQ RDDATl BRA RDDAT2 RDDATl STA 8,Y CMDREG JSR DELAY BRA RDD4 RDD3 LDA 11, Y DATA REG STA ,X+ STORE DATA ROD* LDB 8,Y STATUS BITB 1502 DRQ? BNE RDD3 BITB 1501 BUSY? BNE RDD4 RTS •RDDAT2-READ DATA USING E0X4 STATUS RDDAT2 STA 8.Y CMDREG April '8B 68 Wero Journal JSR DELAY BRA RDD40 RDD30 LDA 11. Y DATA REG STA ,X+ RDD40 LDB 4.Y STATUS BKI RDD30 DRQ? BEO RDD40 LDB 8.Y STATUS RTS •HRDATA-WRITE DATA TO SECTOR OF DISK WRDATA LDB TRK TRK-D? BEO HRDATO SING DEN LDB DNSFLG BEO HRDATO ORA IS02 DOUBL DEN HRDATO LDB T6FLG IP T6.1 USE BEO HRDAT1 E0X8 STATUS BRA HRDAT2 •NRDAT1-WRITE DATA USING E0X9 STATUS WRDAT1 STA 8.Y CMDREG JSR DELAY BRA HDD4 NDD3 STA 11. Y DATA REG NDD4 LDA .X* NDD5 LDB 8.Y STATUS BITB IS02 DRQ? BNE HDD3 BITB IS01 BUSY? BNE MDDS RTS •WRDAT2-WRITE DATA USING E0X4 STATUS WRDAT2 STA 8.Y HRITE CHD JSR DELAY BRA HDD 40 NDD30 STA 11. Y DATA REG W0D40 LDA .X* W0D50 LOB 4. Y STATUS BMI HDD 30 BEO HDD 50 LDB 8, Y RTS •SEEK- UPDATES TRACK. SECTOR ( SIDE < SEEK STB 10. Y SEC REG JSR DELAY CKPA 9.Y TRK REG BEO SKSEC SKTRK STA 11. Y DATA REG JSR DELAY LDA tsia SEEK TRK STA 8.Y CMD REG JSR DELAY SKT1 LDA 8.Y STATUS BITA IS01 BUSY? BNE SKT1 LOOP SKSEC LDA DRVCMD TST 9.Y TRK BEO SKSD TST DNSPLG DOUB DENS? BNE SKDD SKSD CKPB • 50A SEC GT 10? BGT SKSDl SIDE 1 BRA SKSDC SKDO CKPB IS1Z SEC GT 18? BGT SKSDl SKSDC ANDA IS40 SIDE 0? BEO NOUP NO UPDATE LDA DRVCMD SET TO ANDA »$BP BRA SKCH SKSDl ANDA IS40 SIDE 1? BNE HOUP LDA DRVCMD SET TO 1 ORA IS40 SKCH STA DRVCMD STA 4. Y CMD DRV JSR DELTS .1 SEC DEL NOUP RTS •SCRAMBLE-GENERATES TRK FROM CNTR SCRAMB LDA CNTR GET TRK CNT LSRA MOV D2 TO DO LSRA ANDA • SOI STA TRK TEMP STORE LDA CNTR LSLA MOV DO TO D2 LSLA ANDA IS04 ORA TRK COMBINE STA TRK TEKP STORE LDA CNTR ANDA ISFA CLR D0,D2 ORA TRK REVERSE DO. D2 STA TRK IN TRK RTS •CONTIN-CONTINUE AFTER MANUAL ACTION CONTIN JSR PSTRNG PRINT LDX IDSKMSH 1+1 HIT RTN JSR PDATA JSR INCKNE WAIT LDB 8. Y STATUS JSR DELHS SPEEDUP RTS •ASCII -EVALUATES IF REG A IS ASCII, ASCII CMPA • 520 LO ASCII BLO ASET CKPA • S7F HI ASCII BHI ASET BRA AOUT ASET LDA »' . PERIOD AOUT RTS •RDSEC. -READ TRACK 'TRK' SECTOR »SEC 256 BYTES •'0 IF OVER 256 BYTES, " VALID 256 128 BYTE SEC. •'D IF STATUS ERROR RDSEC LDX tsoooo DATA POINTER LDA TRK SET TRACK LDB SEC SET SECTOR JSR SEEK STEP TO TRK LDA • S8C READ SEC CMD JSR RDDATA READ SEC RDSEC 1 BITB • SOS CRC ERR? BEO RDSEC2 LDA I'D BRA RDSEC9 PRINT RDSEC2 BITB tsio REC NOT FND BEO RDSECS LDA «'S BRA RDSEC 9 RDSEC3 CMPX tsoioo 256 BYTES? BEO RDSEC5 OK • BLO RDSEC4 LOH LDA • '0 OVER BRA RDSEC9 RDSEC4 CMPX •S0080 128 BYTES? BEO RDSEC6 OK . BLO RDSC4A UNDR LDA *'B BTHN BRA RDSEC9 RDSC4A LDA »'U UNDER BRA RDSEC9 RDSECS LDA • ' * BRA RDSEC9 RDSEC6 LDA • ' . RDSEC9 RTS •RDTRK -OUTPUT TRACK t AND SECTOR STA IF NO SET TO I' *A' - 'U IF UNDER VALID UMcroJoumd Aprt'tt RDTRK LDA TRK GET TRK 1 STA TEMP 2 TEMP STORE LDX ITBMP2 ADDR LDU ITEMP DEC ADDR LDB • $01 CONV 1 BYTE JSR BDCON BIN TO DEC LDA TEMP LDU IT7TNO+6 MSG ADDR JSR HEXCON UPDATE MSG LDX #T7TNO MSG JSR PSTRNG PRINT LDB • $01 FIRST SEC STB SEC RDTR1 JSR RDSEC READ SECTOR CKPA I'S REC NOT FND7 BEO RDADD FIND ERR TYPE CMPA • ' * OK 7 BEQ RDDUTO CMPA • ' . OK 7 BEQ RDOUT0 BRA RDOUT RDADD CLR TEMP LOOP CNT RDADD 1 I.DX •$0000 LDA ISC4 READ ADDR JSR RDDATA RDADD2 BITB • $10 REC NOT FOUND BEQ RDADD3 NO DATA LDA I'N ON TRK BRA RDOUT RDADD3 LDX •$0002 SEC NO LDA ,x CMPA SEC BEQ RDADD5 ID CRC INC TEMP LDA TEMP READ ALL SEC7 CMPA MAXSEC BNE RDADD1 LODP RDADD4 LDA »'S NO SEC | BRA RDOUT RDADD5 BITB • $08 CRC ERR7 BNE RDAD06 TRK ERR LDA I'C BRA RDDUT RDA0D6 LDA .X TRK 1 CMPA TRK BEQ RDADD7 DATA ADDR MARK LDA • 'T TRK ERR BRA RDOUT RDADD? LDA I'A DATA AM RDOUT INC ERRFLG ERROR RDOUTD LDB TRMSTA KEYSTROKE? BITB • $01 BEO RDOUT1 INC ERRFLG RDOUT1 JSR OUTCH PRINT INC SEC LDA TRK TRX 07 BNE RDOUT2 NO LDB FORMAT DISK FFMT ANDB • $01 SET TO SD LDX IT7SBCT GET MAX SEC LDA B,X BRA RDOUT3 RDOUT2 LDA MAXSEC RDOUT3 CMPA SEC LOOP? LBHS RDTR1 RTS •STPRTE-STBPS IN TO TRK0 FOR TIMB-20USEC(X) « AND THEN CL8ARS CONTROLLER STPRTE LDX STPSIZ SET DELAY LDA IS0F RESTORE CMD STA b,y CMD REG DBL20U LBRN DBL20U DELAY LBRN DEL20U NOP LEAX -1,X COUNT BNE DEL20U LDA l$D0 INTRPT STA 8, Y RTS *OFFSET-FIND THE OFFSET OF THE HEAD WITH CAL DISK *CENTERLINE. RSLTN CONTAINS THE STEPSIZI QUANTITY •IN MILS OR MINUTES. OFFSET LDA • $00 INIT VARS STA SEC STA CNTR OFFSET CNTR STA PCNT +OFFSET STA NCNT -OFFSET OFFHI INC SEC JSR RDSEC READ SEC CMPA • ' . CORRECT? BEO OFFHI1 CMPA • '* CORRECT? BNE OFFLO OFFHI1 LDA CNTR IF CORRECT STA PCNT OFFLO INC SEC JSR RDSEC READ SECTOR CMPA • ' . CORRECT? BEO OFFLOl CMPA • ** CORRECT? BNE OFFEX OFFLOl LDA CNTR IF CORRECT STA NCNT OFFEX INC CNTR ALL SECTORS? LDA CNTR CMPA ($08 BNE OFFHI LDB NCNT NCNT:— NCNT NEGB STB NCNT LDA PCNT READ ALL +OFFSETS? CMPA • $07 BNE OFF1 ADDB IS0C PCNT:-NCNT+12 STB PCNT BRA OFF2 OFF1 CMPB l$F9 READ ALL -OFFSETS? BNE DFF2 SUBA • S0C NCNT;-PCNT-12 STA NCNT OFF2 LDB PCNT B:-ABS(PCNT-NCNT) ADDB NCNT BMI OFF3 NEG OFFSET? LDA #' + BRA OFF 4 OFF3 LDA NEGB • *- OFF4 STB TEMP 3 STA T7MS12+1 STORE SIGN LDA RSLTN COUNT :-B*RSLTN MUL STD COUNT SET LS DIGIT LSR COUNT OF RESULT ROR COUNT+1 BCC OFF5 LDA • *5 BRA OFF 6 OFF 5 LDA • '0 OFF6 STA T7MS12+5 LDX •COUNT SET MS 2 DIGITS LDU IDIGNO+1 OF RESULT LDB • $02 JSR BDCON ApriTM 68 Mero Journal LOU IT7MS12+2 LDA DIGNO+1 CONV DIGIT J5R HEXCON LDX »T7MSli" J5R PDATA PRINT * • RT5 • MAIN PROGRAM • START J5R NXTCH DRIVE NOT 5TA TEMP LDX (TITLE J5R PSTRNG LDA TEMP DRIVE NO. 7 CMP A ISOD IN CMD LINE? BED PROMPT NO DATA CMP A #S0A BEQ PROMPT CMP A i$30 VALID ENTRY BLT STRTO CMPA •S33 BGT STRTO BRA DEFLT STRTO CMPA r> BEQ STRTl USGER LDX IUSCERR J5R PSTRNG JMP HARMS STRT1 J5R FILCHK BRA PROMPT DEFLT J5R NXTCH CMPA IS20 BEQ DEPLT CMPA • •> BNE DEF1 J5R FILCHK OEF1 LDY DRVADDR INIT SLDT LDA TEMP 5TA DSKMSH ANDA • 50F STA DRVCMD LDX •STRT3 5?X RETRY * LDA ISOO 0-1M, 1-2MH2 * STA CLKFLG CLR ERRFLG LBRA Tl PROMPT LDX ISLTMSG I/O SLOT? J5R PSTRNG PRINT LDD •$3730 SET HI/LD J5R CHKIN GET CHAR ANDA • SOP ASCII TO BIN AS LA ASLA A5LA ASLA STA DRVADD+1 SET CONT . ADDR LDY DRVADD I/O ADDR LDA I5AA TEST DATA STA 11, Y DATA REG JSR DELAY LDA 11. X CONTROLLER? CMPA • $AA BEQ PROMP 2 LDX ISLTERR JSR PSTRNG JSR INCH CMPA • ' Y BEQ PROMPT LBRA EXIT PROMP2 LDX • DRVMSG DRIVE NO? STRT2 P ROMP 3 STRT3 NOHLT NOHLT1 Tl T1A TIB T1B1 T1B2 JSR LDD JSR STA ANDA STA LDX JSR LDD JSR ANDA LSRA STA CLR LDX JSR JSR CMPA BNE INC LDX STX BRA LDX STX LDX JSR JSR LDX STX TEST LDX JSR LDA STA LDU LDD JSR JSR LDD STD LDD JSR JSR LDX JSR LDU LDA STA LDB STB LDA STA BRA NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP NOP GET CHAR STORE DRIVENO PSTRNG ■$3330 CHKIN DSKMSH ISDF DRVCMD •CLKMSG CLOCK RATE? PSTRNG PRINT IS3231 LIMITS CHKIN ISOF ASCII->BIN A:-A-l CLKFLG ERRFLG CLEAR ERRORS IHLTMSG HALT AFTER TST PSTRNG INCH • ' Y NOHLT ERRFLG SET ERROR IHLTMSG OPT2+1 NOHLT 1 (ERRFLG OPT2+1 tDSKMSl INSERT DISK PSTRNG PRINT INCHNE WAIT •PROMP3 RETEST AODR RETRY STORE 1 - DRIVE MOTOR STATUS (E0X8) •T1MSG DRV MOT STATUS PSTRNG PRINT DRVCMD GET CMD 4,Y ORVREG (T1AMSH READY DEL TIME IS8027 TIME D7-LO MEAS TIME CLK COMPENS LO-0SEC TIMER CLKCMP 150000 LIMT ISF230 PASFAL UPDMSG HI-.62SEC LIM OK? UPDATE MSG •T1AMSG GET MSG TESTOK PRINT IT1BMSH BUSY RESP TIME IS01 INIT TIMER CNTR STORE IS18 INIT BRANCH T1B2+1 STORE RESTORE NO VER CMDREG 3(J DELAY 2U DELAY ISOO 8,Y T1B3 To Be Continued Next Month WWereJoumtl April '89 47 By. All of us 'Condm ,%% • tyui Vlpfag ', DMW '86 MOTOROLA INC. Microprocessor Products Group 6501 William Cannon Drive West Austin, Texas 78715-8598 CONTACTS Shcmaj DlVCT runmnchamCommuwaOan, Inc. (40^^62 0400 Ow Mosfcy ecu 28 M Microprocessor Pfndutit Group (Slimi-M" HCL AMERICA ANNOUNCES 6S030-BASED MINICOMPUTER System Iteornoraics up to Su 68030 Processor* and 68HK2 Math Coprocessors AUSTIN, Teaas, Feb 6, 1989— HCL America (Sunnyvale, Cahi). a subsidiary of Indus Largest actrrpuier manufacturer, loday announced a mulupnxts^w muucrsnsjsvieT based, on Motorola's 6SO30 (030) proensv and 68882 1882) imA crjpxocea**. TV system, called M3000 series, uses up K> six 25 MHz 030s each coupled wish a 25 Mhx 882 and is targeted ai office auicarwbon and on-line iranucoori processing. HQ/s announcement r^s to a growing base of manufattiBeji offering 030 based products including Apple Conjurer, Apollo. Hew Lett -Packud. NEC and Sony Mktmysknn The 030 n ihe topW- the- line micruonmur in Motorola'i 68000 firmly that includes ihe 68000. 68008. 68010 and 68020- The 68000 family has invented (he largest software base (S3 billion} and most ejiablished hardware base (S100 bsllton), The 030's enrnpaj ibOsty with picceding membtn of the 68000 family protects manufacturers' o*iginal auftware imxumcnis by allowinl for easy migration of software ftorn one 68000 dumxw*- to another. "'IV030'ivert*n]iiy«ndpncc/porfarTTunceadrtnii|ema)teii a rialuial choke for a wide range of markets. " said Murray A. Goldman, senior vice preside m and general manage* of Motorola i Microprocessor Products Group (Austin, Texas). 'X)ur scon -to- be- a ntwunted 680^0 will be a revolutionary produci thai suppnna our large 68000 installed base " HCL America is a wholly ownad jutMitary of HC1. Ltd., Indus largen computer company andkading supplier of Uni* based maefwoes KO, AmcT%*. rfeadQuanxicd in Sunnyvale, Calif, markets IJrua.ba^multtyrocusj* computer syslems and software, as well as cuAoraiard %at rware services to OEM cuspmcjY Moaoioti's 512 billion Semuamrfuctoi Products Sector [Phoeni* , Aha.), which includes ihe Mkn^nu-etwc PioducaOffoup (Austin. Texas), ts a part of Motorola Inc his the largest and broadest supplier of ctTRurocrurcsft in Noc^ Atttj*^. wiih a balanced portfolio of over 50,000 device*. SOFTWARE DEVELOPMENT SYSTEMS, INC 4246 BEUJE AIRE LAKE • DOWNERS GROVE. ILLINOIS 606 IS . USA PHONE 1-312 971^I70« FAX: 14124714613 Contact: James E. Ouilienger February 21, 1989 Mr Larry Williams Exocutpns Editor 68 Micro Journal Computer Punishing. Inc. S^Ca*ar*fca Smith Road Mooon, Tennessee 173*3 Dear Sir. Thanfci for letting ui know about your journal and please canoe rac for not gotuag bock So you. Since your magazine b ucdcaied to ihe orwerafe of Motorola rnkrcfanravMon, | would expec) thai your readers will be interested in ihe ertloacd press release rogardtrrg our CrwasCodc C produci, which bjvM now being mtmsurxxJ under the VAX VMS operating sysirtrt- CrtXSCodcCb an optimizing CcmspiieT that rum on ihe VAX and generates ROM*bleeodeforalJmcTT*ct»of MoUKOla'aC^QCO Eamiryofn CroasCode C is uacd to develop a wide variety of fi£0uT>baaed applaaiiorw. including autoototrvc control iIcvwq. aerospace and defense product*, amuaernrnt devices, cttrnriuu? and cnrnmunkaiiorti equipment, oraumo produda. User scanning equipment, and medical uulnimenla. CroasCode C alfc/ws our cusaumar* to dcveact> code for these ibin using the VAX which provide* a complete and friendly software OLVLkr^uou envinsnmcOL Here's taw iiworktc VAX VMS Ikl " taw) hirdwue The programme i&cs all the powerful and Umfltar tools on the VAX (it terminals, primer*, file syslan. text editor, eic.) to produce C source code for a program that will evcniuilly iut> on the 48000 based hardware- CroaaCode C iben on^ c* lib aowrr code iruo an object fib which feprescnis the actus 1 irwruciian codes to he exccuieO by ihe 68000. Tno objeel file b then convened tmo a formal that b suiiaNe rot tranvruMjtn down a data link lo ibe 68000 hardware, where ihe object code a omfled in a nutshell lhat's how OiKmor C b used. h miTv tvy&at^ realty tprcvd about this ^mmcr Omducau*? Ye* there b- Crautbdc C b flow one of the very few 66000 CcomrderafrUl prwviAa identical capabOisiea under ihieer diSer^tor^atingryitcmerrvtrortnienta: VMS. UNIX, and MS-DOS. Not only thai, bui it allows users lo move their source, object code, and object llhrvka lo different types of m^iuKavtfhriii rwwuj to rocompile « rdink thenL That's because CroasCode Ci object f ilea are inderendem of the machine on which they were g»icr»tcd- Fnr ezampi e, a fronted* C user can take object file* or Ubrarie* that were generated en a VAX. move them loa PC or a SUN and iheo use ihe PC or SUN as aorrcks^rocn) machine Poupie who have nctwvrts con uuning macKines that *W*i two or rhrer different operating systems (VMS, UNIX or MS-DOS) can share their object fOea across the entire network wjthoul having lo worry about what type of machine they happen to be using at the moment Any mac hineort the network, can he used aa a iScvelonrrwi i «**uon to generate code foe a 68000 based uppJicaiJon. Here's an eiatnpie. 48 ApriJ'89 68 Micro Journal twining VMS Ibftfli/fg MS .MS UNIX machine I hope iha Iclfcc/ has giwn >tw enough mfcwnultoo lo decide wr»cthcT our release will bitcrol your rcadc/a. K you have any tnJcstiona about OnwsVode C or ilk significance' to VMS lUcra, ptease give men coll 141 I*312-*71.*I7K> onU III gladly g/ve you more informalion I'veenrUiscd (wo Mack -and white phMta. Sul if )nu nrefri a rotor pJiotO lit send yo\> one On* enckMod pliolo a to Undv^p* o«VM»ll. - and the o4h*r tt a portrait If you plan id prim ihe proJ release, ptcauc civuioW that your rcadcra may want lo see a pkturr of ihc pnxluct DianV you tor yo^r lime and »umt*Dn Stnwd$ DjivJ F Zitlrr. Vrr F'rcftnlenl ROMable 68000 C Compiler Now Available Under VAX VMS Eight months after uw/<*h**ng in CnWodc C aawyOer under MS-OOS and UNIX !ollwc Dcvck»pfi«jii Syucmi. Inc. <SDSI) has announced thai the compiler a now hcing mide tvailaMe under VAX VMS CrotvCod* C is in optima uig C anpOer chal generate* ROMabk code fa all 6TO00 Family fnnr*CT\ indjtfcng the hitfwnd 68020 and «VlV) nrDD"wa> ihe W88 1 and C««2 Doalingpofnl cravoLW<ara, and ihe 68*5 I MMU. Users may wrile source code in any awnhmafian of C and aiiernhry language compile (he ccoe into object file*, and then [ink the object files to form a complete Utad for *ny 6BO0C family SDSl fognd (hat many «f the enmpanica using VMS (or emheddtxJ syaom development were also using MS-DOS and UNIX hosed machaiea- Because of thtx. SDSl dnj^^ C*fK*Codc Clorun identically ind to share source and object files acrou all three operating ryitenu. Tor example, the CroaaCodeC linker on the VAX can link object (Bra that were genera icdhy ihe CnoaCode C compler under MS-D«S This feature tltow* users lo move the)/ fOurttr code, objocl files and libraries from machine lo machine without making any changes at all. Craiat'odt- <- come* with a MmwnU^ornruEMe macro ivvrmMer. ■ linker, a librarian, and a C Irhaiy ctinlaining the source to over 47 C function* Mao included tt a uncvenal dptanJaaarr that enmmunkatea with EF*KOM prt3gr*/nmaw emulators and target hardware. The ddwnloader convert* user*' miyded C code nut Motorola S-RccorJs ami a variety of ouW «ndua**/y oa/vlard fDe formats. II also emit* avmKaie information lo provide symbolic debugging capahQiuea on many popdar emulator*, including those made by Applied Mjcmtyucfni. Atron. liunltville MkxtSYslema, Mkx<OS£_ Norlhmeal Instrument!. Orion [reiriumruiv Softaitl Sophia, .iml /AX CpduCooVC iw on all member* of the VAX famOy, including the MiaoVAX 2000, McroVAX I], MkroVAX 3500/3600, VAX 730 and 380, VAX 6200. and VAX BOOT Scrie* Piicea aurt at tt J95 on (he MkroVAX 2000 SOFTWARE DEVELOPMENT SYSTEMS. INC w*» formed in 1 9M (o provide CTOU development ioWt to people who huikl ROMable annlsEa (ion*. Sbice Ukfl SOSI haa prw*W orasm tnspilai and ctcjb •^onbrro w many hundmb of cortpanuB and their araihanls. Software Oevelopntfn t Systems, tnc is dbdsaicd entirety to the design. (Tumdkctune. and support of Cfm* dcvci»vjmQTl software ULTRASCIENCE The PC-68K1 PC Co-Processor Prnmnat t twnputrni lli-rimw Computrr Sytlrnu Why? The standardized, widely accepted, and ubiquitous 1DM Personal Computer represents an enormous, yet largely untapped, resource or computing power. Unless con- nected to a costly LAN and fileserver, the single-user "PC" provides neither the level of interaction nor the con- venient sharing of data so necessoiy in business, science, or industry. However, the simple addition of a PC-6SKI expands and enhances a PC into a real-time, multi-user, multi-tasking, professional computer system, running OS-9 and MS-DOS concurrently. A PC-68K1/OS-9 equipped PC can host a wide variety ofhigh performance application software, previously reserved for much larger and far more expansive machines, without snenftang MS-DOS applications. The OS-9 operating system offers extensive networking capability and a range of popular languages, including C. BASIC, PASCAL, FORTRAN. etc. Powerful C language VAX/UNIX extensions and a special set of OS-9/UNIX library routines make convert- ing UNIX applications tu the more efficient OS-9 a simple and straightforward task. How? All that is required to get un-line is an operating PC, a PC-68K1 and OS-9. The PC-68K1 allows the user to ftip between active MS-DOS und OS-9 tasks, using the Per- sonal Computer monitor, keyboard, and existing disk drive*. MS-DOS files are available to OS-9 and vice versa. TheMEMlOX Memory/ IO Expansion Board Expand Your PC to 14 Ports and 2 Mega RAM Why? Anyone who has discovered the capability of real-time, multi-user, multitasking OS-9 and the power of the PC-66K1 co-processor knows that adding 10 more efficient serial ports and another megabyte of wait state RAM to a PC-66K1 will turn a "PC" into a substantial computer system. Such a system can run a business, control a plant, automate a laboraloiy, or support an entire class, and the price is right! How? The MEMIOX board plugs onto the PC-66K1 and it is ready to go. For standard installations, there are no jumpers or switches to set The convenient R.J45 port connectors are mounted right on the MEMIOX board set, so there is no clumsy hardware to attach and connecting peripherals could not be easier. Ultrascience offers a complete line of shielded R.J45 modular cable and adapters to make cabling a snap. FAX#l-3l2-2SMM7 PHONE 1 1312 2S6 0080 TB.EX I 910-997-0379 1624 WIlMEm AVE. ■ P.(X 80X556- WiUCTTE. I 60091 68 Micro Journal April '89 49 s FCC 'GOTO' FCB FDB OOGOTO-CTABIZ FCC 'IF* FCB FDB DOIF-CIABLE FCC 'NOTE* FCB FDB OONOTE-CTABIE FCC 'ON* FCB FDB DOON-CTABLE FCC 'PAUSE' FCB FDB DOPAU5-CTABI£ FCC 'REM" FCB FDB OOREM-CTABLE FCC 'SHIFT' FCB FDB DOSHFT-CTABLE FCB FCB FDB PARON FCC 'THEN' FCB FDB PARTHN FCB * MESSAGES FOR RESIDENT PART OF 'DO' 'FLEX comnartd error' 4 'BREAK entered' 4 'Batch file aborted' • PARAMETER TABIE e s PTABLE FCC 'CONTINUE FCB FDB PARCON FCC 'ELSE' FCB FDS PAREI5 FCC 'ENDIF' FCB FDB PARNDF FCC 'ERROR' FCB FDB PAREHR FCC 'EXIST* FCB FDB PARXT FCC •GOTO' FCB FDB PARCIO FCC 'IF* FCB FDB PAR IF FCC 'NOT' FCB FDB PARN07 FCC •OFF' FCB FDB PAROFF FCC 'ON' CMJE3H FCC FCB 9REAKM FCC FCB ABORTH FCC FCB FILERH FCC FCB PREFH FCC FCB CHDPEH FCC FCB CONTNM FCC FCB NOLABM FCC FCB INIFH FCC FCB NOIFH FCC FCB KEN DM FCC FCB w o ^ O < _l a r 1 * M « GO © 9 « 63 9 'Illegal file type' 4 'Illegal parameter reference In batch file' 'Batch file conma/id error' 'Preaa any key to continue ... ' 'Label not found' 'ELSE or ENDIF not found' 'ELSE or ENDIF without IF' s n » 9" 3 3 ? o n n (/> (/> o 'MEMEND changed by batch file; DO rranory not released' RES ID JSR PCRIF RESID1 LEAX HEADCH.PCR ROUTINE STX INCK+1 LBSR TSTABT LBSR READLN LDA LINBUF CHPA »CR BEQ RESIDl BSR TSTCMD BNE RESIDl JSR OOOWD STB I.STERR.PCR TST ERRFLG.PCR BEQ RESIDl SETUP INPUT VECTOR TO 'DO' CHAR INPUT TEST FOR ABORT FROM THE USER READ NEXT LINE FROM COMMAND FILE IGNORE EMPTY LINES TEST FOR A 'DO' INTERNAL COMMAND IF NOT, CALL DOS SAVE LAST DOS ERROR ONLY CHECK ERROR IF ERRFLG SET s FN0IB1 LBNE OS KERR LBSR READLN IDA LIMBUF CHPA • 'g BNE FNDIB1 JSR NXTCH LOX LABPTR, PCR LBSR COMPAR BEO FNDL81 IDA SAV9CH PCR STA ECHOPL, PCR OR GOTOFL PCR RTS * process *if' ocmwo DOIF CLR NOTFLG,PCR LBSR CHKPAR CMPB •PARNOT BHE DOIF1 COM NOTFLG,PCR LBSR CHKPAR DOIF1 CMPB • PARXST BNE DO IF? IDX toewn L£AX -F_LEN, X JSR GETFIL BCS OOERR LOA •MOPNR STA F_FUNC, X JSR FMS BEQ DOIF1A IDA F_ERR, X CHPA •E_FNF BBQ OOFLSE doerr * LBRA DSXERR 0OIF1A IDA IM_CLOS STA F_PUNC, X JSR FMS BNE DOERR * LBRA DOTRUE DOIF2 CHPB •PARERK BNE DOIF 3 TST LSTERR,PCR LBNE OOTRUE BRA DOFLSE DOIF3 LDY LINPTR PSHS y DOIF4 JSR NJflCH ERROR READ LINE FROM COMMAND FU£ IF NO LABEL THEN CONTINUE (UNTIL EOF) ELSE MOVE LINE POINTER TO START OF LABEL COMPARE LABELS CONTINUE IF NO MATCH RESTORE ECHO FLAG CLEAR GOTO FLAG EISE RETURN CLEAR 'NOT' PARAMETER FLAG CHECK PA/WE TER IF 'NOT' THEN SET 'NT' PARAMETER FLAG AND GET KEXI PARAMETER IF 'EXIST' SETUP TEHPORARY FCB REPORT DOS ERROR, INVALID FII£NAME TRY TO OPEN FII£ FOR READ IF ERROR m FILE NOT FOUND THEN FALSE ELSE DISK ERROR IF NO ERROR, CLOSE FII£ AND GOTO TRUE ELSE IF 'ERROR' THEN IF LAST ERROR FLAG <> THEN TRUE ELSE FALSE ELSE LOOK FOR '-' WA •CR LBEQ CWPER CHPA • *- BNE D0IF4 LDY LINPTR LOA •SPACE STA ,-Y LEAY 1,1 LOA 0,Y CHPA •SPACE BHI DOIF5 JSR NXTCH DOIF5 PUIS X LDY LINPTR LBSR COMPAR BNE DOTRUE STY LINPTR D0IF6 JSR NXTCH CMPA > SPACE BHI DOIF6 DOFLSE TST NOTFLG.PCR BNE DOTRUl DOFLS1 LBSR CHKPAR CHPB •PARTHN BBQ DOFLS2 CMPB IPARGTO LBNE CMDPER RTS DOFLS2 IDA eCfiOFL,PCR STA SAVBCH,PCR CLR 8CB0FL.PCR INC OOTOFL.PCR CLR IFLEVX,PCR DOFLS3 LBSR READLN LBSR CHKPAR CHPB fPARIF BNE 0OFLS4 INC IFLEVX.PCR BRA D0FLS3 DOFIS4 CHPB •PARNDF BNE DOFLSS DEC IFI£VX,PCR BPL DOFLS3 BRA D0FIS6 DOFI.SS TST IFLEVX.PCR BNE DOFIS3 CMPB •PARELS BNE DOFLS3 INC IFLEVL.PCR IF CR FOUND, HAVE PARAMETER ERROR CHANGS *-' TO SPACE IF NEXT CHAR - DELIMITER THEN GET NEXT CHAR RESTORE POINTER TO FIRST STRING SAVE LINPTR COMPARE STRINGS RESTORE LINPTR H3VE PAST SBCOND STRING TO CHAR <- SPACE FALSE; TEST 'NOT' FLAG DEFINITELY FAI.SE; CHECK PARAMETER IF NOT 'THEN' OR 'GOTO' HAVE ERROR IF 'THEN' SAVE CURRENT ECHO FLAG DISABI£ ECHO SET GOTO FLAG TO DISABI£ PARAKS CLEAR BYPASS 'IF' I£VEL READ NEXT COMMAND LINE IF 'IF' THEN INCREMENT BYPASS 'IF' LEVEL CONTINUE IF 'ENDIF' THEN DECREMENT BYPASS 'IF' I£VEL EXIT LOOP IF < IF BYPASS 'IF' I£VBL > THEN CONTINUE LOOP IF 'ELSE' THEN INCREMENT 'IF' NESTING LEVEL TSTB BEQ RESID1 LEAX ERRLAB,PCR IBSR FNDLAB BRA RESID1 IF DOS ERROR THEN GOTO LABEL IN 6RR1AB * INTERNAL COMMAND PROCESSING ROUTINES * TEST FOR INTERNAL 'DO' COMMAND, AND PROCESS IT * RETURN Z = FOR INTERNAL COMMAND , Z - 1 OTHERWISE 8 S 2 § TSTCK) IDA fLINPTR] CMPA t' < BEQ TSTCKL CMPA *'8 BNE TSTCM2 TSTCM1 LBSR DOR EM BRA TSTCM3 TSTCM2 LBSR CHKCMD BEQ TSTCM4 JSR D,X TSTOO 1DB »1 ROUTINE TSTChH RTS r n o 2. • PROCESS 'ECHO' COMMAND ft DOECHO LBSR CHKPAR CHPB IPARON BNE DOECH1 IDA i$FF STA ECHOFL, PCR RTS * DOECH1 CMPB IPAROFF I BNE CMDPER CLR ECHOFL, PCR RTS • PROCESS ELSE COWBND DOELSE TST IFLEVL, PCR X BEQ NIFERR s o LDA ECHOFL, PCR STA CLR SAVECH, PCR ECHOFL, PCR GET FIRST CHAR OF NEW COMMAND IF COMMENT CHAR THEN PROCESS COMMENT IF LABEL CHAR THEN PROCESS AS A COMMENT ELSE CHECK FOR AN INTERNAL 'DO' COM«ND IF IT IS, CALL ROUTINE TO PROCESS IT CLEAR ZERO FLAG TO INDICATE INTERNAL CHECK PARA»«TER IF 'ON' THEN SET ECHO FLAG ON ELSE IF NOT 'OFF' THEN HAVE ERROR ELSE SET ECHO FIAG OFF IF 'IF' NESTING LEVEL • THEN HAVE ERROR ELSE SAVE CURRENT ECHO FLAG INC GOTOFL,PCR CLR IFLEVX, PCR ooei-Sl LBSR READLN IBSR CHKPAR CMPB IPARIF BNE DOELS2 INC IFLEVX, PCR DOEI.S2 CMPB IPARNDF BNE DOE LSI DEC IFLEVX, PCR SPL D0ELS1 CLR GOTOFL,PCR IDA SAVBCH, PCR STA ECHOFL, PCR D0ELS3 DEC RTS IFLEVL, PCR PROCESS ENDIF COMMAND DOENDF TST IFLEVL, PCR BNE DOELS3 NIFERR LEAX NOIFM, PCR LHRA P ERROR PROCESS 'EXIT' COMMAND DOEXIT LBRA EXIT PROCESS 'GOTO' COMMAND DISABLE ECHO DOGOTO L£AX LABEL, PCR IBB *e DOGOTD JSR NXTCH STA ,X+ CMPA ISP ACE BLS DOGOT1 DECB BNE DOGOTD CLR ,x DOGOT1 1£AX LABEL, PCR FNDIAB STX LABPTR, PCR LDA ECHOFL, PCR STA SAVECH, PCR CLR ECHOFL, PCR INC GOTOFL, PCR LEAX DOTCB.PCR IDA IMRWHD STA F FUNC.X JSR FMS SET GOTO FIAG TO DISABLE PARAMS CLEAR BYPASS 'IF' IEVEL READ NF.XT COt*IAND LINE IF 'IF' THEN INCREMENT BYPASS 'IF' LEVEL CONTINUE UNTIL 'ENDIF' DECREMENT BYPASS 'IF' LEVEL CONTINUE IF >- CLEAR GOTO FLAG RESTORE ECHO STATUS DECREMENT 'IF' NESTING LEVEL IF 'IF' NESTING LEVEL <> THEN DECREMENT 'IF' NESTING IEVEL ELSE REPORT MISSING 'IF' MESSAGE MOVE LABEL TO LABEL BUFFER SAVE IABEL POINTER SAVE CURRENT ECHO FLAG TURN ECHO OFF SET GOTO FLAG REWIND TO START OF COMMAND FII£ CONTACTS Sob Anundson B8open Consortium. Ltd. <S03>6&2-57*)3 cWOPEN CONSORTIUM PROMOTES OPEN STANDARDS FOR MOTOROLA 8W0O RISC MICROPROCESSOR; DEMONSTRATES "NEW WAV OF DOING BUSINESS" Ihe 88opcn Consortium ii a groupof leading tariwoiT uid software vendor*, formally chartered to Apnl 1988 to establish and («w»it ihe industry- *ide adoption of open cuManb tor pioducts using the Mmorola 8 WOO RISC (Reduced tnstnaziart Sei Computing) microprocessor arcraaxturc The rncrnhcxs of SSopen are leading the way id a new way of doing business in Die computer iMuicy — ocmonsrjanng thai vendors who c ompe t e vn"orous|y against ooe anmhr. in ihe marketplace can benefit from dedicating reaaurccs to a coopuaove effon id achieve mutuaj success. Early Success: The 8*000 Binary Compatibility Standard Only three months after bdng formaDy chartered, the Contoroum published the Trial Uir Stanaord dorun^nt for the 88000 Binary Compatibility Standard (BCS) Vk first BCS puWiihed for any RISC rnjcroproExasr. this standard wtll allow software written id its specifications id run on any BCS-omipUani 88000-bascd system Ihe 88000 BCS is evmpatibJt with si>ch widely rsaocpted industry standards as POSIX. XJOpen • and ihe AT&T UNIX System tntetfaee Definition. It supports 90 percent of ihe spctifkarioAS developed for the propmed Mouaola 68000 BCS. simplifying ihe portability of applications from one ucht<Ecruic to the other. In addition* the 88000 BCS u UTipknxnicd ai the iystctn call tcweLuid is therefore applicable to any varum of ihe UNIX operating syjtcm or to a non-UNIX opera dng system. The 88open Mission The raiswai of tbe88open Consortium reflects in* .ocottkmi belief of us membership thai the Motorola B8000 tiucmpraEesaor RISC architecture offers a superior platform Tor the ocveioprneni of cnrnpuier tystents and software. The timely nnplemertiaaori of wandards for hardware and software interfaces to the 88000. prior to ihe development of a substantial body of systems and software, wilt achieve clear benefit;! for both vendors and users of 88000- based products. The existence of standards will rtrinee the complcjury and cosi of developing opcrtiing systems, compilers, applications software, and peripheral hardwire, Moracrver, having invcued once in designing for a standard environment, vendors can then invest mote of their devcloptncrii rcsourco in improving and differentiating their pioduci*. The elaborate parang activities typically needed to move software from one vendor's system to mother's will be virtually eliminated. The resulting cost savings will benefit vendors and users alike. B8opcn welcomes participation in its standanis setting activities from all interested parties, including ihe academic and user communities. All standards developed by (lie Consortium will be shared openly and equally. 8*open members will autornatkally receive copfes of iiAodard* related documents, and non-members may receive them for a nominal fee. Thenxmbasof the Cartwrtivm believe product development based on standard* wtll accelerate market acceptance of the 88000 through carker availability of a broader selection ofinoov&ovt systems and software at rca&MAbk pnecs Customers a* well as vendors will be more willing id invest in produces based on open standards if they arc confident of not being penalized for early aartrretmeni to a young technology The 88open Charter The Iftoprft Gsfliofli urn charter aims to bting about the market siioctis of the 88000 RISC archjfcxaure through a variety of programs, beginning wiih educational and marketing activities ioprw™« the 88000 product family as the noa complete and viable RISC processor platform available to the market As a pan of tit mission to develop open standards and encmirege their use, the Qnsortium wdlpamcipateuisiandarD^Kiting activities woriflwiifc, whenn^approonaic The Cunsoroum also will speak as a coUectivc voice to influence future development of the 88000 product fami ly Ihe Conowoum may negotiate wiih hardware and software vendors to produce 8 8000- rebued products for ihe benefit of &8open members In addiboa. ntonbasmay coocer»ie » develop pmdicts that meet mutual needs, through coUabnraut*) berween rnernbm or through march and development sponsored by the Consortium. It is expected that Consortium members will benefit from more favorable acerat to such products. The 88000 BCS is a subset of the proposed AT AT application biliary utierfaac ( Afil). a n^inr cnrnplete interface standard that will even uilly specify extension ftaniroiucb as ihamd librtiysupoor.netwcirlt interlace and the "look and feel" of the windowing systrrn. ATAT has indicated informally Us intention to use the 88000 BCS as the basis for an 88000 A8I. Ihe 88opcn Consortium will work wiih ATAT to complete the B8000 ABL and will continue to develop certification vols to verify thai application programs conform to the BCS standard. The 88000 BCS exemplifies ihe manner in which smicpc cooperation among enlightened compan ies can yield substantive benefits, not only to the direct participants, but also to vendors and customers iluoughcui the industry There is no dotibt that ihe establisfcrrmi of this standard so near the incoducuoo of the 88000 architecture will radUlate lis tnpid acceptance and muovaie a vonciy of developers, to exploit its powerful capabilities. Organization 1 he OMKoruurn ts ayjaao^ ast not -for-profit me m b er onjanuai>on It is governed by a Board of Directors representing member ce>mpan>cs. w uh three directors from hardware vendors, two from software vrrvtivs, one from the academic awamiioicy, one from ibe roonbtfsrup at large, and one frotn Motorola. Day-to-cUy operations am run by a full time executive director, who also serves on the Board of Directors, Bob Anundson, formerly of Tektronix. Inc.. is executive director He is assisted by a small staff. The work of toe Consortium is carried do through member volunteer c**nroanccs_ Die Technical Committee is charged with proposing and developi ng definitions, standards and extensions relating to the use of the 88000 architecture and tat use in hardware and software; this commiow pioduced the B8D00 BCS. Ihe Business Committee deals with business issues relabng to the Cceoontum. and has broad ■caponsibiliry for the promotion of the 88000 product family in the ma/kCTptaoc. The Wot Id Strategies Committee will focus on i he extension of the CctisssrtiunTs activities and irifl ik nee worldwide. 88open Consortium offices and siarTarc pla/iiicd f« Europe and the Pacific Rim Corpora te numbers have full voting r ghls, and thus participate in setting ihe goals and ecserreining ihe activities of the Coniortium. Non 'vocbg meoihenhjps ore otTeredto indivxtuals and non-profii orsjsnizations who are inmrsted at the goal j of the CtmsorouiiL for mere information, contact BBopcn Consoroum. Ltd.. 8560 SW Salish Lane. Suite 500, Wilsonvilte, Oregon 97070; telephone (503) 68Z-5703. 68 M cro Journal April 89 53 Juttae nmrwwy-Ampiiib Iw|um* Prwmawa, lac U0frt1»4M FOR OAMEDIATE R£LCA5£: Bob Aa&odrnj (xajsawros I -Data General riu^bu Language Processor*., lac. Announce* Compiler Product* far Metorele 83000 RISC Architecture C— »jH— Tea* Fore* L>ft**a Scrt»*r* t>n'ilnji— t , Fpbmarr II, I WW - Laiunue* Pnxeewm. lar |UPfJ today Upon Receipt Andrew Hettinger (408) 898-4060 Hfttut product* for Mm Motorola MQOO RISC aftbiuwlu.'* u4 tfeal it feat )«intd lb* gftopm CoaaortLaei Lid Sohwar* UlUaUV* Proff eat Tb* ta r n w ia ^ i n agi *«* mad* *i a am «0or*fftn bald by avopas *ad attepdvd by bom that) Ift ia^pwdni aoflwani vaadent eonmEti«J La tb< mm rlaodard, LPI bu baan « matabar of lb* iftepan CaeanAium dac« June L08* Slsce Jttlrtl* 19W.Swv« W Wrloia/i. Sank* VW Pmidnt o* Ratetrtb wd 0«««k>p(B€Bt At LPI. baa bald a poslUoe. da tb* Wop*» Bovd of Director*. Uacuift Prmmmt*. let- » pot tJM iu> Faaut* of Coaa4l*r* to Lb* HOW FLE5C *KbJt*rtur*. Tha LPI Faally of Co»ptlere lactase* BASIC. C, COBOL, FORTRAN, PjacaL, PL/], and RFC D pro*/ aa mi a* limpiNpa Amo iwledad » CadaWateb. u laiarertw, w(«-li«l dtbuffff , Tb*a» ptodtKu as* np*cud to b« retaaead u> tha anad «vert*r 9/ (BCT. Tba Softsar* (aftiaitv* ww forovd la Au|WI IflSf te pOKftOt* lb* dW*kfpm*Bt of aoftau* f«* ^>* **«» Klx' arcbmaamal Re" mpM ;?"-*< Ml ftfaVAj ;*.v aliMnM rm |M Bttt MHb1 10 • portiaa lab, HOOD *>**b>pB**t fcardman. wad ra*ra*tiai and UtbAkral tappatt frofe lb* bespat CoavTUsn. Depot SoAwas* laUUtl** aponaw* jotted* t>*U Gra«*ai. Dotpbtp Cm^w. a wbaUisry of Nefab D«U. UofafMt, NCR. Opaa SrvWen and Saoyo/koo. laLM-malioaai *W# hav* aaaa •'•riin| ajf Mill) *ith ia4*o*f>d*fii aoftwan hbiWi ta niuN th»l asalica-Uwi «UI b* iataudiaidy a*atlabM f*r MOOO-baaae ayatsa*.* Mid Bab A*ynd*oa, Eiacalin Dirariot of ib* M a f ia Goaanrtiua *Tb* >u<rraa* add moBMLun of lb* pracrast m du*tu> fmiavas to tb* ftoviaj aoaaatu* far tb* MoLoreU MOOC arcbitactvn-* LPl'i cacaptlan a*d mfttiM tool* in bales d —tlopad 10 ba aompliaat wlcb Oopa'a Blsur Comp*UoUi(y fiUoda^d (&C81- Tb* BGfi bat baaa tdopcad at put of ttc ATftT AppttcaUaa Riuir Icttrfact (AB41. tt b capwiad tbai BC8 aompllaat toftwait artll i&u«aav rrrtcuu for tefivara dwttoptrt aod prevldt tad umu witb ifavlnh-wrapptd aoftwan toIutioAt for aU 8800O. •LPI «UI ba tbfl Ant laA4ua(« vondofa 14 taaka arailablo an cotirv Utaiif of iatc(FBl«l toKw* dmlapaMBi lAolt \at lb* bMorolt MOOO RISC tDtuMiMt. 1 tutat Roy A, Fbutay, PraaidaBi of LPL "Tbtaa iaoit will flv* tbt and uatr taa ability to artabiadk fovras toftwart appJicaUoat ob lb* nOQO aKbiiKtar* 4 ■> •*!( aa 4*ntap aaw oaaa Tbia abowa LPI< coairailDwat to Mopaa aad o«r btab wpwlttwat of iba b a aitaai oppottaaiUat ttao<V t1 «d *ifc» MiwnU't WOW RiSC Lit, a j n im m af aaalHr toftwata drr t lopa n at toobv off*** a ML nil* of JJ-btt | Uaniaaa*. 1^1^*4 BASIC, C. COBOL. FORTRAN, PASCAL. PL/L Ri*G 0, and u bjtBBfilr** a*aw*.W*at dihuajar. GoaaWaKb Tbaat aoflvar* pradtttta an praamlly aff*n4 an ladaattr-Maadard tytvma aacb at Mataaala a MUtO, WE130M. wd tntalt MOM maaiac UNO*. X£NDC*, aad MS-D09* apatatia, trtitaa H^daaartand I* PrtaabgHtra. Maatacauaatta, LPI it a na/kat Uadar io qoably |4ojuaa> ronpiW* a»d toftaaM daYalopnaant Uwla. Siaet iU tauptLai is 1M0, LPl'tiMbteU^cd aa*aalat* bat auraclad m^ot tuHowm* i»c|gdinf «nnpui*r uuiufw^^f*, Fartun* VOO FompaDtaa, tb* fadaral jd T iM a H, kpplkattoa aoflwif* davalopara, ud a k»i hat of dcMnaatw atd uiiaroaiacxLal aoftvara diHrtbwl«wv DATA GKNSAAL UNVCXLS riRST 1NDVSTRY STANDARD FAMILY OP RISC-BASED SYSTEMS SAN FHAHC1SCO, Pab. 28, 19A9 -- Data General today unvallad tha flrat ajaaab*ra of ita n«i* faailiy of products baaad on tho Motorola 88000 architecture. The ennounceaenta include; a really of hlgh-perforaence, VKE-bssed syatMfl that function aa aervers or auiti-utar ayateaia for comarclei and technical appi lcationa; a new faally of desktop pereonet worketatlona; and 00/uX 4.1. tha lnduetry's only version or UNIX aupportlng e full RISC-beeed faally of eyeteaa. The BBOOO-baaad faally eetabllahee lnduetry-leedlng prlce/perfoxeanca atandarda for RISC-based syeteae. The eyeteaa perform at 20 or 40 MIPS or greater end tho worketatlona perform at 17 or 20 MIPS or greater, beeed on Dhryetone aeaaureaente. 'Today'a announceaent aetebllehae Date General aa the Induetry'e leader in coet-ef fectlve. dletrlbuted ooaputlng ■yitflan. Data General *e eaperlence In building large, wall-balanced, high-throughput ayateaia glvea our eaerglng faally of 88000- baead product* a aejor advantage over elaller aachinee from workstation and PC vendors. Thie experience hee enebled Deta C«cw»rel to deeign aiachlnea apeclflcally for the newly emerging atylaa of computing euch as the Cllent/aarver model.' eeld Herb Oeher. division director of product marketing. "Todey we ere Introducing the flrat an abara of a binary-compatible feally of products thet ere feeter, lata coat lv end more flexible than any other eyeteaa on the market. for the price of todey'e pereonel coaputere, the BBOOO-beeed worketetlon dellvare tha performance of a aid-range workstation. TTtm 8 8000 -baaed eyeten la not a heedleee worketetlon or e repeckeged alnlooaputer , it le designed with eophietlceted 1/0 and large cepeclty etorege to eupport a large number of ASCII terminate, networked X-Wlndow grephlce tarmlnela end worketetlone. " The powerful, mldrenge BBOOO-baaad eyetem le available with one or two proceeeore. By ewepplng the CPU board the user cen upgrede fraei s elngle to s duel procaaeor, neerly doubling CPU performance. The ayetem le evallable In deekelde end 14-lnch-hlgh. MEMA-Standard reckmount verelona. Both ayetaaa Include e ten-alot. induatry-etandard VMB-compatlble chssele, up to 208 MB of error-correcting code) ■aeorv end cartridge end reel-to-reel tepe options. Tha deskside verelon Includee up to 2 glgebytee of ■•■■ etorege end the reckmount verelon contelna up to 16 glgebytee of meee etorege through duel-ported B-lnch SMD dieka. induatry-etandard wide and local aree ccaaain lcetlona ere provided by eyncKronoua contrnllera (VSC/A) end tha vme bue Ethernet controller [VIC). Up to 512 RS-232 esynchronove ccavxectlona are handled by terminal eervicee hoet edepter boerde (VTJA/128) which connect up to 128 eaynchxtmoua devices vie clueter controller boxes. The worketetlone can support from 4 to 28 MB ef memory In a 2-1/Z-lnch-hlgh pedeetel encioeure. The worketetlone use e elngle multilayer board containing a Motorola 88100 CPU end FPU, two Motorola 88200 CMMUe (Cache and Memory Management Unite), and monochrome or color grephlce hardware co-proceeaors. Interfaces for e PC/AT keyboerd. e three-button optlcel eouee, e SCSI port, a parallel port, a aerlel port and 802.3 Ethernet are atandard. 54 April '89 66 Micro Journal The •ooochrowo workatationo uit tho NSC UP072120 Advanced Graphic* (Hep ley Controller. Tho eight-plane color unit, based on a custom graphics coprocaaaor can display 256 alnulteneoue colore fro* a peiette of 16.7 allllon colore. The 20-inch monochrome end 19-lnch color flicker-free ■ooltora operete et K)Kz end deliver 1280*1024 pine! raeolutlon. SCSI peripheral enpenelon for tho dlakleaa workatatlone la via compact deekeop housings. Up to three of theae housings can be aonnacted to a workstation. Porlpherel options lncludo e 160-H8 or 322-HD S-l/4-lnch full-height UincAester disks, a 15D-MB quarter-Inch cartridge (QIC) tspn drive end 5-1/4-inch or 3.5-lnch floppy drive. UG/UX 4.1 la an enhanced version of Data General's coraerclsl-grsds dc/UX 4.0 UNIX operetlng eyetea end le dealgned to take full edventege of the Kotorols BB000 RISC architecture. DO/UK 4.1 lncorporetea the major featuraa found In UNIX ayeteaa with today'a doainent UNIX atanderde. "DC/ux 4.1. Date Generel'e UNIX operating ayetea with advenced symmetric ami tlprocssslng, eneblee tho BflOOO-bssod system to deliver cost-effective computing end Modular growth while protecting Investments tn eppllcetlon software end peripherals, " ssld Oehor. "OG/UX 4.1 conforms to the BBopen Binary Compatibility Standard allowing cuetontere to run all BCS-coajpliant aoftwere developed on 08000 -based systems from any vendor. Application develop* re can uae Data General's oq/ux 4.1 on ell these aysteae to produce ehrlnk-wreppad software for the entire BB000 aarketplece. " JMICR&NICS ^^""^ RESEARCH CORP. Microcomputer! • Hard* arc and Software CtMIX* Sale*. ServKcandSuppon 11 Feb 1989 Dear Don, Just two typos to report in the February 'logically Speaking' - both mine. Page 19, 6th line up from bottom - should read "making* Page 22, para 6, line 3, should read "internal-01 doesn't consistently line up . ..", JOT "intemal-23" Now that I've got just over 2 years of programing in 68000 Assembler under my belt, I can see what a tetrible price we have to pay by writing in position- independent code. I feel that programs such aa my KBASIC would execute at last 2 to 3 tiroes as fast if there were only a special version of SK*D0S dedicated to the single end-user (no multi-tasking or multi-user stuff in the works) , eosething like the old 6809 FLEX. With RAHDISK at the top of cBnory, and a fixed Load- Address for applications programs, we could really blast along] Itie Load-Address could be set high enough (say $10000) to allow for future improvements to SK*nos. Ah, if only this could be so . . . I'd gladly take tbe time to re-write RBASIC for such a set-up, and still keep up with the present version to run under a future multi- whatsit SK*CQSU Bow about it, Pete? (Stark, that is) Maybe readers who program in 68000 Assembler would be interested in learning of the many snags which lie in wait for the unwary, and some of tbe procedures I've evolved to make a programmer's life a little easier! I«t me know! 1 Don Willisas, 68 Micro Journal, 5900 Cassandra asith Road, aixson, W 37343 Sincerely, u R. Jones President DG/UX 4.1 la e full PQS I X implementation, compatible with Syate* V Interface Definition laaua 2 <SVID2) and Berkeley Software Dletrlbution (BSD) 4.2. DG/UX offere the beneflte of eny etandard UNIX ayatea while providing unique, application- transparent capabilities auch aa fully symmetric multiprocessing eupcort; e robuat. flexible file eyetea.' an enhanced scheduler end extended error reporting. it also supports TCP/IP comnunlcetlona so f Were, NFS 4.0 end X-Wlndowe Verelon 11 Releeaa 3. DG/UX 4.1 fully implements the BSopan Binary Coepatiblllty Standerd. "*» a founding member of BBopen Date General haa actively participated In the BSopan Software Initiative program dealgned to eupply eoftware for BBOOO-beecd eyetaata, " addod Oeher. The BBopen le a non-profit consortiua of companies that aanufecture, develop, eell or uae products beeed on Motorola's BB000 RISC architecture, CROSS ASSEMBLERS The CSC Cross Assemblers arc actually a set of table- driven assemblers for several different microprocessors, currently including 1802/5. 6800/1/11/6303, 6804. 6805. 6809. 6502/3/65C02. 8080/5, 8048. 8051, Z-8. Z-80. with others being added occasionally. Separate 68010 and 32000 cross-assemblers are also available for the same price. A macro preprocessor is included with any purchase of cross- assemblers. In addition to the basic ability to recognize the target computer's assembly language and to process symbolic address expressions, the cross-assemblers have the ability to support structured assembly programming, the ability to define conditional assembly directive expressions, and the ability to assemble progiams stored in modular form. Since the Motorola 6801 instruction set forms a superset of the 6800, 6802, and 6808 instruction sets, and is the same as the 6803 instiuction set, the CSC 6801 Assembler is also capable of assembling programs written for all of those microprocessors. The CSC 6801 Assembler is also capable of assembling programs for the Hitachi 6303, as it recognizes the six extra instructions, and for the Motorola 68HC 1 1 , as it recognizes the extra instructions and addressing modes. The CSC 6809 assembler also recognizes 6800 and 6801 mnemonics, generating code for ihem to closely simulate the register and memory changes caused by the instructions on the 6800 and 6801. The CSC 6805 Assembler is also capable of assembling programs written for the Motorola 146805 and 68HC05, as it recognizes the extra instructions implemented by those proeesscffs. CSC also offers simulators to assist in the debugging of object progiams destined for 6800/1, 6805, 6809. 6502, and Z-80 systems, on selected host systems only. The cross-assemblers, with printed manuals, binary utilities, and macro pre-processor, are available for a tetail price of $50 each $100 for three, or all for $200. The C sources arc available for an additional $50 each, $100 for three, or $300 for all. The programs may beoideiad from the following address: Computer Systems Consultants, Inc. 1454 Laua Lane, Conyers, CA 30207 Telephone Number 404-483-4570/1717 68 Micro Journal Ap«i| '89 55 Jim Howell 5472 Playa Del Ray San Jose. Ca. 95123 Dear Don. I have been using Introl C. ver- sion 1.5. for about three years on my OS-9 (tm Mlcroware) Level 2 system (6809). and am quite pleased with it. It seems more bug-free than some of the MS- DOS C compilers. A few years ago. Dr. Dobb's Journal pointed out several bugs in various C compilers for MS-DOS. I tried all of these tests on the Introl C com- piler and found that it has none of these bugs. 1 have, however, found one bug and a "problem" in the Introl compiler. The bug occurs when a character (type "char") variable or expression is used as a subscript. In Listing l. "tabl" is an array of structures, and thecharactervariable "x" Is used as a subscript. For the last line of the listing, the code generated by the compiler multiplies the subscript "x" by the size of the structure (10 bytes) to get the offset of the desired element of the array. In this ex- ample, the result will be 260 (deci- mal) or 0104 (hex). The high -order byte of this offset is then cleared (apparently because It is a "charac- ter" value!), resulting in 0004 (hex), or simply 4. Ihe offset is then added to the address of the start of the array in order to point at the desired element of the array. The result is four bytes from the start of the array, and not 260 as It should be. Slmilarbehavior will occur if an "int" (Integer) array is subscripted with a character variable or expres- sion whose value is 128 or more. This bug can be gotten around by usingan "int" as the subscript or by using a cast, such as: val = tabl((int) x].a: The "problem" mentioned in the first paragraph can be illustrated by considering the program "sample.c" (listing 2). To compile and link this program, one uses the commands: ice sample ilink sample This results in an executable mod- ule called "sample", on the same disk as the source file, in my sys- tem, this is usually disk "/dl". 1 usually "load" a program first be- fore running it. as follows: load /d 1 /sample sample And now for the problem. When "sample" is run (the last line above), an access to the system disk is made before any output from "sample" occurs. After some inves- tigation. I discovered what Is hap- pening. When "sample", or any C program. Is run, It starts In a "start-up" rou- tine. This start-up routine does some initialization and set up. be- fore calling "main". One of the functions of this initialization is to put in initial values for external data, which is all variables declared outside of all C functions. External variables which were declared with initial values will have those values inserted at this time. All other external variables will be set to (zero). In order to set up the declared ini- tial values, the start-up code looks for a module whose name is the same as that of the executable module, with the characteis ".DAT" added at the end. For this example, it would look for a module named "sample.DAT". This module, if it exists, is assumed to contain a table of initial values and where in the data area to put them. First, the "link" system call is used, to see if the module is already in memory. If this call fails, the "load" system call is used, to look for the module in the commands directory. This is the disk access which 1 would like to avoid. If the "load" fails, too. then plugging initial val- ues Is skipped. Then "main" is called, which in this case prints a message. This scheme works fine for pro- grams that declare one or more initial values. In this case, the module sample.DAT (or whatever the name happens to be) is part of the executable file "sample". That is. if "sample.c" had contained one or more variables with an initial value, the file "sample" would con- tain an executable module called "sample" followed by the data module "sample.DAT". The "link" system call in the start-up code would succeed and no extraneous disk access occurs. In the example, however, no initial values are given and the module "sample.DAT is NOT part of the file "sample". In fact. "sample.DAT" is not created anywhere. An easy "fix" would be to declare an initial value for some (external) variable. However. I preferred a solution that did not require an extra module, so I came up with the solution In Listing 3. The added function "fixload" patches the startup code to skip the attempt to load the module from the com- mands directory. It does this by looking at the function "cprep". the function in the introl llbraiy that looks for the data module. The first two tests in "fixload" check to see if the module has already been patched. The rest of the tests verify that the code being patched is really the intended code. If all of the tests succeed, a branch instruction is inserted. Thlsbranches around the attempt to "load" the data module. With this fix. the program still looks for the module "sample.DAT" on disk the first time it is run. since the startup code is executed before "main" is called. After that, as long as the in-memory copy of the pro- gram is run. the extraneous disk access does not occur. To make the change more permanent. SAVE the 66 April '69 68 Micro Journal program (alter running It at least once) and use VERIFY to correct the CRC. Then use ATTR on the corrected module to make It ex- ecutable. Note that the values used In "flxload" apply to the partlcularverslon of the Introl C Compiler that I have, and may or may not be correct for other versions. The solution I have described is adequate for my use. but It requires ihe addition of code (the function "flxload") to most of the C pro- grams that I write for OS-9. A more elegant solution would be to write a separate program that patches the disk copy of the module directly (and updates the CRC). Listing 1 Shows bug in Version 1.5 of Introl C Compiler. struct int a; lnt b; int c; lnt d; lnt e; ) tabl [30]; main () lnt val; cha r x; x - 26; val - tabl [x] .a ) The compiler generates the following assembly output. The "elra" instruction should not be there. tabl comm loc 300 _main: leas ldb -3,s 126 stb Ida 2,3 110 mul clra leax tabl, y ldd d, x std 0.3 leas 3,3 rts end Listing 2 /* sample .c 06/09/88 V ma i n ( ) { printf ("This Is a sample program\n") ; ) TRANSLATE 6809 TO 68020 A service to provide a raw translation of 6809 assembler to 68000 or 68020 Cut 50% or more off hand translation time Converts to 68000 or 68020 assembler Old code may be included in comments Special conversion of psh andpul to movem All 6809 addressing modes supported Choice of register assignments and sizes Upper to lower case conversion included In-line warnings for likely problems Don't pitch It - translate It Call or fax fot example listings and farther information. Call (or informs tloo on OS-9 consultancy aervkes, dlak cacnlng, graphic*, word proceoon, languages and file """"\p"* 'OS-? Is a Trademark of Microware Inc. n n ; __| Windsor Systems L 340T Unw (On Lana, LeuMO*. KV 40223 U.&A » »( * c» w : (SCO) 420 8000 Fax: (Mq 42S 3844 Listing 3 /• sample. c 06/09/88 •/ /* with the function "loadskip" •/ main!) ( fixloadO; print f ["This is a sample pro<iram\n") ; ) loadskip.c 06/13/86 15:35 This routine patches "cprep" so that it will not look on the disk for <module>-DAT when the program starts . int cprep [) ; fix load!) ( char *p; p ■ (char *) scprep; if(pI58l « x6< it p[57J — Jx3( " p[53) — 0x10 It Pl54] — 0x26 " p(55) — 0x00 u P(56] — 0x16 it p|75) =- 0x10 it P(76] = • 0x27) p|S7) > 0x20, P(58] ) m 0x10, 60Mcro Journal April 69 57 Q Classifieds ^^^^ sue***,*™ j Surplus Unused Motorola VME Modules & Electronic Solutions Enclosures for Sale at Discount MVME133 CPU Module-68020, 1MB DRAM, 68881 FPP, $675 3 serial Pons, EPROM Sockeis, VMEbus Interface MVME225-I 1MB DRAM Module, A32/D32 VMEbus Inierface $380 MVME320A Winchester/ Fllopy Controller $490 MVME332 8 Channel intelligent Serial Communications Module $675 Series 7 Electronic Solutions 7 Slot Desktop Enclosure, P1/P2 $695 Backplane, 325W PS, Space for Winchester/Floppy/Tape Respond to: John Gannon, RPG. P.O. Box CI 2399, Ste 162, Scottsdale. Arizona 85267 Phone (602) 951- 3373 *** S+ Memory Cards, CPU Cards. Hard Disks w/Controller Cards, I/O Cards, Cabinets, Power Supplies. S/09 CPU Cards, Memory, I/O Cards, Controller Cards. Cabinets, Power Supplies 3-Dual 8" drive enclosure with power supply. New in box. $125 each. 5-Siemens 8" Disk Drive, $100 each. Tom (615) 842-4600 M-F 9AM to 5PM EST *** (JMX-20 with 68881 Unillex, utility pgmsand C; 16 MHZ. This month only $1900. Unused. Marc Tailsman MD, 30818 S Coast Hwy, S Laguna, CA 92677, MSG: 714 499 1877 UBiMSfaBfBjgMBiBjgraEEEMaMotgjgjgfgjiifi^ia^^ |T« f| 1 1 1 X /S K -IIOS />! S-IIOS Transfer Utilities For 68XXX and C0C0* OS-9™ Systems Now READ - WRITE - DIR - DUMP - EXPLORE FLEX, SK-DOS & MS-DOS Disk These Utilities come with a rich set of options allowing the transfer of text type files from/to FLEX & MS-DOS disks. *CoCo systems require the D.P. Johnson SDISK utilities and OS-9 and two drives of which one must be a "host" floppy. CoCo Version: $69.95 68XXX Version $99.95 S.E. Media PO Box 849. Hixson.TN 37343 (615)842-4601 Telex 5106006630 FAX (615)842.7990 BUEjaiErBGJgjgffljEiarBBBEjfflEjaa^ 58 April 89 68 Mao Joumal r ■\ SK*DOS768K Read the fine print to see what's in SK*DOS/68K: Full DOS documentation plus on-line helpO Multiple directories O User-installable device driven D Install up to 8 different I/O devices O Keyboard type-ahead D Print-screen D Virtual (RAM) disk O Disk cache D Up to 10 dtivts □ SW or 3H* floppy diives G Hard drives to 64 megabytes each O I/O redirection to drives or I/O O Time/date stamping Of files O File or disk write protect (even hard disk) Q Batch Tiles O Support for 68000, 68010. 68020 □ Monochrome or color vidao board support O Read and wiite MS- DOS disk files □ 6809 Emulator □ Powerful utilities such as copy-by- date, undelete, show differences between fiiles, prompted delete, tent fie browse, and more - all included O Simple Basic included O Fast assembler included O Line editor included O User support via newsletter and BBS O Available software; C compiler, full Basic, screen editors, disassemblers, cross-assemblers, spelling checker, text formatter, music editor, hard disk manager, ROM-based debugger, modem communications programs, etc. More compilers coming, O (Some features may not be implemented in all hardware manufac- turers' implementations.) Individual copies of SK*DOS/6SK are $140; less in quantity or when bundled with hardware. Send Jor our 6809 / 68K hardware and software catalog. Also available as part of hardware/software educational course. our f"_ ■ \J Software Systems Corp. VT A P . K p - ° Box 209J ^l"l\ IN Mt Kisco, NY 10549 (914)241-0287 BBS (914) 24 1-3)07 Fax (914) 241-8607 OS-9.m SOFTWARE FORTH09 A Foith-83 system lor OS-9. many exciting fea- tures, integrated wel into OS-9 environment Allows you to easily save executable program modules written in Forth (See t)» FORTH column in this magazine.) $150.00 Manual only with credit toward later software purchase.. $25.00 FORTH09 for OS-8768000 win be released soon! Inquire! MSF MS-DOS File Manager module used with SDISK3 driver to get complete transfer capabilities to/lrom MS-DOS disks (360K and 720K formats inducing 3.5*) For CoCo-3. REQUIRES SDISK3 MSF alone $45.00 with SDISK3 included . $65.00 SDISK3 replacement floppy "driver" lor tlte COCO-3 w/OS 9 L2. Supports all OS-9 formats, plus required by transfer programs to access non-OS-9 lormats. $29.95 L1 UTILITY PAK The most extensive and useful utility package for either Level 1 or 2 OS-9. 40 programs, usable with any OS-9 system. Includes MACGEN command language compiler. $49.95 L2 UTILITY PAK More utilities, mostly for Level 2. Includes COCO specific software Ram Disk driver and error driver, others are non-system dependent. $39.95 L1+L2 COMBINATION Both together for.. $75.00 Diskettes are 5.25* Color Computer OS-9 Format unless requested otherwise All orders must be prepaid or COD. VISA/MC accept- ed. Add shipping, $2 00 min (S3 00 for FORTH09) Order from! D. P. Johnson, 7655 SW Cedarcrost St., Portland, OR 97223 (503) 244-8152 OS-4 I* a trademark or Miooarara and Miami* Inc. MS-OOStoamdamarkalMtoTauA mc FORTMDS It a IrarfarraA <t OPMtmn SOFTWARK «8000 C CROSS-COMPILER SltC . SKOOSMSIX}5,l>NlX,XKHll IO»JI\CT OSI.1t Acftpv Kak C IM6MJC. ffl»erilc*6a00Ou»<>^ ode l*bda«g0l0cnarancn>bkr.kfejvfc> gn»*al l*r SKUOS.ba m»y he modified CROSS-ASSKMUI.KRS WITH MACRO CAPABILITIES km ii ti6.n.t;x,Oii.vnin.KX.usuos.i'Mi<.Ki>os.xMix mtaa m.uum Sprtilr IKM. »«n. MOI/I I, MW4.M05, MOT. M. JM. W «. (Ml. KOK5. (>80 10. 3?000 UtUtf noH uacaiMcn in C. with kodThriki** uuliuet Sounxt for «u moral SWearb.JIUOfor J, iy» tin ill CMOUKM TELECOMMUNICATIONS PROGRAM siM-nzooi,sKnoi.iwn*itm>si.xKNix.vmw:x otJKcr-Oxi.r kacii jjs Mnni«iicn •>■ nmiMl rnnk. I>k mtfei. MOW Ml. XcK Xl*> m SUPER SLEUTH DISASSEMBLERS e»c« tntir.x inmst iitt-vwrtm oairn-axtf k*cw t$o fir.xait.coco barms tflrfnEnttioarce on dM* **h1j|**v mfcidfiirrl. tajr> edVtMf Srm.it riaro u.i.m.wmm mn m ucvnin m™ COCDOO5^»V*tr»M00.UJ.5.1.ft«a! an. (uUM«l.ll~l) Uatt «ccu» tIMtLtX.OSMlNIFLEX.MSOOS.IINIX-SliDOS. XENIX DEBUGGING SIMULATORS FOR POPULAR 6. BIT CPUS nm vt-rtxt inmost uhwfui oijscr-osir hih u**coco run. coco <st tmrmutfl* nrriitiir pmrwrv .ithidn dnawhtt I.Tmidjnf , himry nlat SfxityUm rJ0iyi.<u»*».«»i.<*>9OS»o.!r. zjrjRix ml, ASSEMBLER CODE TRANSLATORS FOR 6S02, 6800/1. 6809 uai »<an tn-TutxwJXist».VNtruu ami m um 4 mm wf-imd. tit-tiEi m~as* o**i hhwfuj FULI..SCREEN XBAStC PROGRAMS »«> nv rerun* AVAILABLE FOR FLEX, UNIFLEX, AND MSDOS Urrrafaf Oram** I Oorfl SMI WWC Hi llOOWUne. DO 1100 •nwit. SUntau Tlfaeb Urn Spraditee HOOWtonc. 5M »Mro DISK AND XBASIC UTILITY PROGRAM LIBRARY SU-FIS1 litVNtriJLXWSBOS Ed4 drtt Melon. ion drrotlcrf.mavmfinwia* acting, da dnk sou. ns^mDtmncorall BASIC puv«l. uri BASIC |ne«l. Mr mlUX wrantoncludcwnindnsquDKZvcoV PROFESSIONAL SERVICES FOR T W ¥. COMPUTING COMMUNITY CVSTOMtZEO PROGRAMMING We will cuttttniM my <4 (far fwtyw** dc*oitn*4 i ■YBuliKd cu>u*n«Ei w Or to Ctr*a tarw *-n*-^-K*n ttpon \bt raarti-w^xiilfy «f the n**di/k ■.••m. I iha ■l^rrvafn-ra Hirobt bc-xKure ror the c harp for lisrJicxaarniutK-^ rfcycndi CONTRACT PROGRAMMING We wilicruic new p-rt-fiwdcaT modify ciuuof pn-*zra*au cm a tgnt-m taSw. >tctti» ** n*tc (Mbrtitd lora-fa twenty -pean; the t»w»T*L-w-ri en wtucb w* Hvt pt-rtt-nwil cnj-o-an fw*>raWfnt3j u-cJyrJc owop^auiu nvwfcto <rf n wi ff iPMney hk IxJaHa; ifiM, Qtwiaqp*, Um*-*ic. r4nacj-»«-J , nw-u popular r-tfmfcb of ra.nti»**«*w nxfualuti OCC. IBM. DC. HP, AT*T. tod otoa -x-fwl* bwd> trf mwrmmp***. .rx l*1,n f ftMCVl . 6*09. 2». tSOL MOtO. n Unfxjtpi ■*) onfflKHif it-V-mi. i» Lt-4rfni ra-afuiff m u wnjV ftrwrdf oDD-oAm: -be chwsc far ami puism hi » k»uOr bt i* to* or by ibe taii CONSULTING We aTTet ft -nde -i-^ 0/ rasTft-a*-*a« *a) «■. mfci t ftpd dfup* <-> hi* mp-c r-fb-rav-i tWtd ■**!****■ BUM. LTaT**fll ttU CS ( Tr**a*a*a*aV Computer Systems Consultants Inc. khiauk CW^x. Go|i 3CCDI IWHII^IO IWIII1I1I1 ii«l M fc I- hJ m_ n« a ■ ^*. VQA *4 UUTtS CMP l| I Uil,.«fa.«<4 M <...Ml« 68 Micro Journal AprH '89 59 Clearbrook Software Group (604) 853-911 8 CSG IMS is THE full featured relation- al database manager for OS9/OSK. The comprehensive structured ap- plication language and B +Tree Index structures make CSG IMS the Ideal tool for file-intensive applications. CSG IMS lor C0C02/3 OS9 LI/2 (single user) $169.95 CSG IMS lor 0S9 L2 or 68000 (multi user) $495.00 CSG IMS demo with manual $30 MSF - MSDos File Manager for CoCo 3/OS9 Level 2 allows you to use MSDos disks directly under OS9. Requires CoCo 3, OS9 L2. SDISK3 driver $45.00 SERINA - System Mode Debugger for OS9 L2 allows you to trace execution of any system' module, set break points, assemble and disassemble code and examine and change memoiy. Requires CoCo3 or Glmlx II, OS9 L2 & 80 col. terminal $139.00 ERINA - Symbolic User Mode Debugger for OS9 lets you find bugs by displaying the machine state and Instuctlons being executed. Set break points, change memory, assemble and disassemble code. Requires 80 column display, OS9 L1/2 $69.00 Shipping: N. America - $5, Overseas - $10 Clearbrook Software Group P.O. Box 6000-499, Sumas, WA 98295 OSS It * tradtmart of Mlcrovwa Syt lemi Corp.. MSDos It a Iradtmarfc of Microsoft Corp. SPECIAL ATARI™ & OS-9™ NOW! - If you have either the Atari 520 or 1040 - you can take advantage of the "bargain of a lifetime" OS-9 68K and BASIC all for the low, low price of: $150.00 Call or Write S.E. /Wedia 5900 Cassandra Smith Rd. Hixson, TNJ 37343 615 842-4601 FAX (615J842-7990 r ^\ ATARI & AMIGA CALL As most of you know, we are veiy sensitive to your wishes, as concerns the contents of these pages. One of the things that many of you have repeatedly written or called about is coverage for the Atari * Amiga"* series of 68000 computers. Actually we haven't been too keen on those systems due to a lack of serious software. They were mainly expensive "game-toy" systems. However, recently we are seeing more and more honest-to-goodness serious soft- ware forthe Atari & Amiga machines. That makes a differ- ence. I feel that we are ready tostait someserious looking into a section for the Atari & Amiga computers. Especially so since OS-9 is now lunning on the Atari deview copy on the way for evaluation andiepoit to you) and rumored for the Amiga. Many of you are doing all kinds of interesting things on these systems. By sharing we all benefit. This I must stress - Input fi-om you on t be Atari & Amiga. As most of you are aware, we are a "contributor supported" magazine. That means that YOU have to do your part. Which Is the way it has been for over 10 years. We need articles, tech- nical, reviews of hardware and software, pro- gramming (all languages) and the many other fac- ets of support that we have pursued for these many years. Also I will need several to volunteer to do regular columns on the Atari & Amiga systems. Without constant input we can't make it ftyl So, if you do your part, we certainly will do ours. How about it, drop me a line or give me a phone call and I will get additional information right back to you. We need your input and support if this Is to succeed! DMW 60 April -89 J 66 Mcro Journal- THE 6800-6809 BOOKS ..HEAR ye. HCAR OS-9" User Notes By: Peter Dibble The publishers of 68' Micro Journal are proud Co make available Che publication of Peter Dibbles OSS us« tores Information for the BCCOntCR Co the HO, regular or CoCo 0S9 Dalng 059 HELP. HINTS. PROBLEMS, REVIEWS. SUGGESTIONS, C0MP1AINTS, 0S9 STANDARDS, Generating a New Bootstrap, BulJdlnR a new System Disk, 0S9 Users Croup, etc. Program loterfsclng to 0S9 DEVICE DESCRIPTORS, DIRECTORIES. "FORKS*. PROTECTION "SUSPEND STATE", "I I PES", 'INPUT/OUTPl T SYSTEM", etc. Programing Languages Assembly Language Programs and Interfacing: Baslc09, C, Pascal, and Cobol reviews, programs, and uses; etc. Disks Include ■o typing all the Source Listings to. Source Code and, where applicable, saaembled or compiled Operating Programs. The Source and the Discussions in che Columns can be uaed "as la", or as a "Starting Point* for developing your OWN nore powerful Programs. Programs sonetlnes use aultlple Languages such as a ahort Assembly Language Routine for reading a Dlreccory, which Is then "piped* to a Bsalc09 Routine for output formatting, etc. BOOK $9.95 Typeset — w/ Source Listings (3-Hole Punched; 8 x 11) Deluxe Binder _._-_s5.50 All Source Listings on Disk 1-8" SS. SD Disk - 2-5" SS. DO Disks - - $14.95 - $24.95 FLEX™ USER NOTES By: Ronald Anderson The publishers of 68 MICRO JOURNAL are proud to ke available the publication of Ron Anderaon'a FLU USER MOTES, In book form. This popular monthly column has reen s regular feature In 68' HICKO JOURNAL SINCE 1979. It has earned the respect of thoussnds of 68 MICRO JOURNAL resders over the yeara. In fact, ott'a column has been described as the 'Bible' for 68XX users, by some of the world's leading microprocessor professionals. The most needed and popular 68XX book available. Over the years Ron's coluan hss been one of the most popular In 68 MICRO JOURNAL. And of course 68 MICRO JOURNAL is the most populsr 68XX magazine published. Listed below sre s few of the TEXT files Included In the book and on diskette. All TEXT files in Ihe book are on the disks LOGO C 1 File load program to offset memoiy — ASM PIC MEMOVECl Memoiy move program — ASM PIC DUMP Cl Pnnler dump program — uses LOGO — ASM PIC SUBTEST Cl Simulation ot 6800 code to 6809, show differences — ASM TERMEM C2 Modem input to disk (or olher port inpul to disk) — ASM M C2 Output a file 10 modem (of another pan) — ASM PRINT C3 Parallel (enhanced) pnnter driver — ASM MODEMC2 TTL output lo CRT and modem (or olher port) — ASM SCIPKG Cl Scientific math routines — PASCAL U C« Mini-monitor disk resident, many useful (unctions — ASM PRINT C4 Parallel pnnler driver, without PFl AG — ASM SETCS Set printer modes — ASM SETBAS1 C5 Set printer modes — A.BASIC NOTE: .C1..C2, etc. -Chapter 1. Chapter 2, etc. ••Over 30 TEXT files included Is ASM (assembler J-PASCAL- PIC (position Independent code) TSC BASIC-C, etc. Book only: $7.95 + $2.50 S/H With disk: 5" $20.90 + $2.50 S/H With disk: 8" $22.90 + $2.50 S/H Shipping A Handling $3.30 per Book. $2 30 per Disk xt Foreign Olden Addtt 50 Surface Mail O-S7.00 JUrMiU If paying by check - Please alkw t*6 weeks delivery * All Currency in U.S. Dollars Continually Updated In 68 Micro Journal Monthly Computer Publishing Inc. 5900 Cassandra Smith Rd. Hixson, TN 37343 'FLEX la a trademark of Technical Systems Consultants *0S9 Is a trademark of Mlcroware and Motorola *68' Micro Journal la a trademark of Computer Publishing Inc. (61 5) 842-4601 Telex 5106006630 FAX (615)842.7990 68 Mraro Journal April '89 61 1 !!! Subscribe Now !!! 68 MICRO JOURNAL OK, PLEASE ENTER MY SUBSCRIPTION Bill My: Mastercard^ VISAj | Card* . Exp. Date. For 1 Year 2 Years 3 Years Enclosed: S Name Street. City State Zip Country_ My Computer Is: Subscription Rates U.S.A.: 1 Year $24.50, 2 Years $42.50, 3 Years $64.50 'Foreign Surface: Add $12.00 per Year to USA Price. •Foreign Airmail: Add $48.00 per Year to USA Price. 'Canada & Mexico: Add $9.50 per Year to USA Price. *U.S. Currency Cash or Check Drawn on a USA Bank ! \ / 68 Micro Journal 5900 Cassandra Smith Rd POB 849 Hixson.TN 37343 Telephone 615 842-4600 Telex 510 600-6630 62 -A' Reader Service Disks Poetry, Disk- 1 Filesoil, Minical, Minicopy \Minifms, "Lifetime "Foodlisl, ••Did. DUk- 2 Diskcdit w/ insi & fixe*. Plime. •Pnnod. ••Snoopy. "Football, "Hexpawn."Liretime. Disk- 3 Cbug09, Seel, S«2. Find. Table2. Inlext. Disk-exp, •Disksave. Disk- 4 Mailing Program, *Finddai, "Change, •Testdisk Disk. S 'DISKFDC 1, 'DISKFK 2, ••LETIER, "LOVESIGN. ••BLACKJAK, "BOWLING. Disk- 6 ••Purchase Order, Index (Disk file mdx). Disk- 7 Linking Loader, Rload, darkness. Disk. 8 Cncsl, Lanpher (May 82). Disk- 9 Dalecopy, Diskfix9 (Aug 82). Disk. 10 Home Accounting (July 82). Disk- 11 Dissembler (June 84). Disk- 12 Modcm68 (May 84). Dlsk-13 •Imlmf68, Testmf68,*Cleanup.*Dskalign.l4elp.Dale.Txl Dlsk-14 •Inil. 'Test. •Tenninal, 'Find, 'Diskcdii. Iiiil Lib. Dlsk-15 Modcm9 + Updates (Dec, 84 Gilchiisi) to Modon9 (April 84 Commo). Disk- 16 Copy.Txl, Copy. Doc. CaiTxl. Cai.Doc. Disk- 17 Match Utility, RATBAS. A Basic Preprocessor. Disk-18 ParscMod, Sizc.Cind (Sept. 85 Armsuong),CMDCODE. CMD.Txt (Sept. 85 Spray). Dlsk-19 Clock. Dale, Copy, Cat, PDE1. Asm A Doc, Errors Sys, Do, Log.Asm & Doc. Disk. 20 UNDC Like Tools (July A Sept 85 Taylor A Gilchrist) Diagoii.C, Grep.C, LS.C, FDUMP C. Uisk-21 Utilities A Games - Dale, Life, Madness, Touch, Goblin, Slarshol, A 15 more. Disk-22 Read CPM & Non-Fl.EX Disks. FraserMay 1984. Disk-23 ISAM, Indexed Sequential file Accessing Methods, Condon Nov. 1985. Extensible Table Driven, language Recognition Utility, Anderson March 1986, Disk-24 68° Micro Journal Index of Articles A Bit Bucket Items rrom 1979 - 1985, John Current. Dlsk-2S KERM1T for FLEX derived ftom the UNDC ver. Burg Feb. 1986. (2)-5" Disks or (0-8" Disk. Disk-26 Compacu UrriBoard review, code A diagram, Burlison March 86. Dlsk-27 ROTABlT.TXT, SUMSTEST.TXT. CONDATA.TXT. BADMEN.TXT. Disk-28 CT-82 Emulator, bit mapped. Disk -29 "Star Trek Dlsk-30 Simple Winchester, Dec. '86 Green. Disk-31 "• Read/Write MS/PC-DOS (SK'DOS) Disk-32 Heir-UNK Type upgrade 68MJ 2/87 Dlsk-33 Build the GT-4 Terminal - 68MJ 1 1/87 Condon. Disk 34 FLEX 6809 Diagnostics, Disk Drive Test. ROM Test, RAM Test - 68MJ 4/88Koipi NOTE: This is a reader seivice ONLYI No Warranty it offeicd or implied, they are as received by 68' Micro Journal, and are for reader convenience ONLY (some MAY include fixes or patches). Also 6800 and 6809 programs are mixed, as each is fairly simple (mostly) to convert to the other. Software is available to cross -assemble all. ♦ Denotes 6800 - ••Denotes BASIC • • • Denotes 68000 - 6809 no indicator. 8" disk $19.50 5" disk $16.95 Shipping & Handling -U.S.A. Add: - $3.50 Overseas add: $4.50 Surface - $7.00 Airmail 68 MICRO JOURNAL 5900 Cassandra Smith Rd. Hixson.TN 37343 (615) 842-4600 - Telex 510 600-6630 April '89 68 Micro Journal The Only 6809 BASIC to Binary Compiler for OS-9 FLEXorSK*DOS Even runs on the 68XXX SK*DOS Systems* Hundreds Sold at Suggested Retail: $100.00 • 6809 ■ 0S-9™ users can now transfer their FLEX™ Extended BASIC (XBASIC) source files to OS-9, com- pile with the OS-9 version and run them as any other OS-9 binary "CMD" program. Much faster than BASIC programs. • 6809 - FLEX users can compiler their BASIC source files to a regular FLEX "CMD" file. Much faster execu- tion. •68XXX - SK'DOS™ users running on 68XXX systems (such as the Mustang-08/A) can continue to execute their 6809 FLEX BASIC and compiled programs while getting things ported over to the 68XXX. SK'DOS allows 6809 programs to run in emulation mode. This is the only system we know of that will run both 6809 & 68XXX binary files. K-BASIC is a truecompiler. Compiling BASIC 6809 pro- grams to binary command type programs. The savings in RAM needed and the increased speed of binary execution makes this a must for the serious user. And the price is now RIGHT! Don't get caught up in the "Learn a New Lan- guage" syndrome ■ Write Your Program in BASIC, Debug it in BASIC and Then Com pile it to a .CMD Binary File. For a LIMITED time save over 65%... This sale will not be repeated after it's over! * SPECIAL Thank-You-Sale Only J\ rom: C S.E. Media P I 5900 Cassandra Smith Rd. Hlxson, Tn 37343 Telephone 615 642-6809 Telex 510 600-6630 A Division of Computer Publishing inc. Over 1,200 Titles- 6800-6809-66000 FAX (615)842-7990 * K.BASIC will ran under 6SXXX SK'DOS in emulation mode for ihc 6809. Price subject lo change wiihovl notice. 68 Mao Journal April '89 63 PT-68000 SINGLE BOARD COMPUTER The PT68K2 is Available in a Variety of Formats From Basic Kits to Completely Assembled Systems BASIC KIT (8 MHZ) - Board. 68000. HUMBUG MONITOR + BASIC in ROM. 4K STATIC RAM. 2 SERIAL PORTS, all Components $200 PACKAGE DEAL • Complete Kit with Board 68000 10 MHZ. SK'OOS. 51 2K RAM. and all Necessary Parts $575 ASSEMBLED BOARD (12 MHZ) Completely Tested. 1024K RAM, FLOPPY CONTROLLER. PIA. SK'DOS S899 ASSEMBLED SYSTEM - 10 MHZ BOARD. CABINET POWER SUPPLY. MONITOR + KEYBOARD. 80 TRACK FLOPPY DRIVE, CABLES $1299 For A 20 MEG DRIVE, CONTROLLER and CABLES Add $295 PROFESSIONAL OS9 "SK'DOS It a Tradema* o( STARK SOFTWARE SYSTEMS CORP. "OS9 to a TiBdama* d MictowatB SS 00 FEATURES MC68000 Processor, 8 MHZ Clock (optional 10,12.5 MHZ) 51 2K or 1024K of DRAM (no wait slates) 4K of SPAM (61 16) 32K.64K or 128K of EPROM Four RS 232 Serial Ports Floppy disk controller will control up to lour 5 1/4", 40 or 80 track. Clock with on-board battery. 2 - 8 bit Parallel Poits Board can be mounted in an IBM type PC/ XT cabinet and has a power oonnector to match the IBM type power supply Expansion ports - 6 IBM PC/XT compatible I/O ports The HUMBUG" monitor supports monochrome anoVor color adaptor cards and Western Digital Winchester interlace cards PERIPHERAL TECHNOLOGY 1710 Cumberland Point Dr., Suits 8 Marietta, Georgia 30067 404/984-0742 VISA/MASTERCARD/CHECK/C.O.D. Send For Catalogue For Complete Information On All Products (r r DATA-COMP special Heavy Duty Power Supplies For A limited time our HEAVY DUTY SWITCHING POWER SUPPLY. These are BRAND NEW units. Note that these prices are less than 1/4 the normal price for these high quality units. r Make: Boschert Size: I0.S x 5 x 2.5 inches Including heavy mounting bracket and hcaisink. Rating: in 1 10/220 volts ac (strap change) Out: 130 watts Output: +5v-10amps + 12v. 4.0 amps + I2v - 2.0 amps -12v . 0.5 amps Mating Connector. Terminal strip Load Reaction: Automatic short circuit recovery SPECIAL: $59.95 each 2 or more $49.95 each Add: $7.50 each S/H V J Make: Boschert Size: 10.75 x 6.2 x 2.25 inches Rating: 1 10/220 ac (strap change) Out: 81 watts Outputs: +5v - 8.0 amps + 12v - 2.4 amps + 12v - 2.4 amps + 12v . 2.1 amps -12v - 0.4 amps Mating Connectors: Molex Load Reaction: Automatic short circuit recovery SPECIAL: $49.95 each 2 or more $39.95 each Add: $7.50 S/H each ^ 5900 Cassandra Smith Rd., Hixson, Tn. 37343 Telephone 615 842-4600 Telex 51 600-6630 Fax (615)842-7990 64 April '89 68 Micro Journal Now Offering: *FLEX'" (2 Versions) AND 'STAR-DOS PLUS+ '" For Ordering Call (615)842-4600 FROM - DATA-COMP, C.P.I. A Family of 100% 68XX Support Facilities The Folks who FIRST Put FLEX' V on The CoCo TSC Editor NOW $3 SOT STAR DOS PLUS + • Functions Same as FLEX • Reads - writes FLEX Disks • Run FLEX Programs • Just type: Run * STAR-DOS" • Over 3100 utilities & progiams to choose from. PLUS ALL VERSIONS OF FLEX & STAR-DOS* INCLUDE Read-Write-Dir RS Disk + Run RS Basic from Both More Free Utlities External Terminal Program + Test Disk Program + Disk Examine & Repair Program ■f Memory Examine Program + Many Many More!!! TSC Assembler Kr|i»00 NOW $35.09 UPS Battery Backup 330 VA UPS 4 Outlets 520 VA UPS 4 Outlets 800 VA UPS 6 Outlets 1200 VA UPS 6 Outlets Disks 309.00 5" Box of ten ( 1 0) 599.00 DS-DD 5.99 40 Track 859.00 DS-DD 14.95 80 Track 999.00 3.5 DS-DD $ 15.95 Printers EPSON LQ-500 24 Pin 350. 00 LX-800 9 Pin 239. 00 Gender Changer SERIAL Male to Male 8.95 Female to Female 8.95 Centronics Male to Male 9.95 Female to Female 9.95 Switch Boxes Serial/Parallel Converter 69.00 Parallel Serial Converter 69.00 Serial 2 Position Switch Box 27.00 Parallel 2 Position Switch Box 27.00 Serial 4 Position Switch Box 35.00 Parallel 4 Position Switch Box 33.00 Serial 4 PoslUon Crossover Switch 44.00 Parallel 4 Position Crossover Switch 49.00 Serial 2 Position 9 Pin Switch Box 44.00 DATA-COMP 5900 CASSANDRA SMITH RD. HDCSON. TN. 37343 SHIPPING USA ADD 2% FOREIGN ADD 5% MJN $3.60 (615)842-4600 FOR ORDERING TELEX 5 1O60O6630 An Ace of a System in Spades! The New MUSTANG-08/A : TM Now with 4 aerial porta standard At speed increase to 12 Mhz CPU + on board battery backup and Includes the PROFESSIONAL OS*9 package - wwii^g the $500.00 OS-9 C compUerl This offer wont last foreverl NOT 128K, NOT 512K FULL 768K No wait ram Now even faster! with 12 Mhz CPU I The MUSTANGOB™ system took every hand from al I other 68006 syaema we taated. running OS-9 6SK1 I The MUST ANG08 rdudes OS^6fiK~ and** Polar | Stark* SrCDOS™ 9CDOS is a single user, single tasking | system that takes up where ^RBC™ left off. SKDQS is | actuary a 68XXX FLEX type system r Not a ISC ptoducL) m ^rs sH m OS-9 6 8 K a v amrn i sjLBi l btawrnaaal LMtiii-S iut'- , CCantpOe timec OS« 68K Hani Disk MU5TANG0B 8 Urn CPU 0n*t-32sK Other poputor ffiOOe sytarn Imh-Oaase MUSTANG420 OiAi-Hase -^ System Includes OS-9 68K or SK-DOS - Your Choice Specifications: CPU MC68D0B 12 Mhz HAM 7sax No Wats** 25K CMpj PORTS 4-RS32 UC8BB1 QUART 2-fit*PaaU MOSB21 PIA CLOCK MK48T02 Real Trne Cta* B* BAJ EPflOM t6K,3»orMK jrtititto FLOPPY WDim SlftDmaa HUB DISK Mates Port WD1CCB Bod 25 Megabyte Hard Disk System V Now more serial ports * faster CPU i J Battery B/U - and $850.00 OS-9 Profes- 1 [ v slonal with C compiler Included! 5 J *$400.00 See Mustangs M for trade-in delate MUSTANG-08 LOOK Smsxfe 32 bk Rt^utrr C...1B.0...9.0 i ».a...6.3 rwi-,v register long I; tor (bO; I < 096999; *4t); $2,398.90 Complete with PROFESSIONAL OS-9 includes the $500.00 C compiler, PC style cabinet, heavy duty power supply, 5" DDDS 60 track floppy, 25 McgByte V Hard Disk - Ready to Run j Uha\e other 68006 cystoma there are several significant drterences. The MUSTANG-08 is a M 12 Megahertz sydam. The RAM in*) NO wait Sates, this means M bore MUSTANG type performance. Also, aJb*ty for addressable RDM4>ROM the RAM b the maximum slowed for a 68008. The 66008 can only address a total of 1 Megabytes of RAM The desrjn afbws al the RAM apace (tar al practral purposes) to be utlfead What Is not avslabie to the user b required and reserved for the system A RAM dak of 480K can be easry ajrfQured. leaving 288K free for pnjrarrvByaterri RAM space. The RAM OISK can be configured to any siae your ^ la stf i x i requirea (system must have 128K In addition to Is other requirernents). Leaving the mrardtr of the ongtoal 7B8K for program use. Sutraari source included (dnvera, etc) must an gjs k in. h, iii rfcn Data-Comp Division 9^^g^ fl Decade of Quality Service' y j W^pjjjfP Systems WaktWkie Computer Publishing, Inc. 5900 Cassanfra SmNh Road Tsterfione 615842-4601 • Telex 510 600-6630 rtaxi Tn 37343 S , * These will SWPC r*<ten*y FLEX 5" - Col tor spectt Mb.