INFO-PC@USC-ISIB (02/12/83)
From: Dick Gillmann <INFO-PC@USC-ISIB> Info-PC Digest Friday, 11 February 1983 Volume 2 : Issue 9 Today's Topics: Microsoft Flight Simulator Basic Compiler Bug C Language and Pointers (3 msgs) HERCULES Graphics Card (2 msgs) Mailing List Program The Dangers of Coffee PC Uses for the Handicapped Assembly Language MODEM7 Aparrat Memory Board ---------------------------------------------------------------------- Date: 8 Feb 83 07:40:33 EST From: SHOLAR@CMU-CS-C Subject: Re: Microsoft Flight Simulator To: INFO-PC@USC-ISIB The Microsoft Flight Simulator seems to do a remarkable job of simulating a Cessna 182. If you ignore the window display and look at the instruments, it seems to behave much like the GAT I use once in a while. In VFR flight, it lacks a bit of realism because you must manipulate the keyboard in order to look out of the windows at different angles -- you can't easily turn your head, in other words. I can only compare it to the GAT and a C-152, and find that it doesn't do too badly. A flight instructor friend tried it, and with only 30 seconds of familiarization with the controls, he seemed to feel right at home. Well worth the $50. Bill ------------------------------ Date: 7 Feb 83 10:06:06-PST (Mon) To: info-pc@isib From: harpo!floyd!vax135!cornell!roger (Roger Hoover)@Berkeley.arpa Subject: Basic Compiler Bug Has anyone else had the IBM Basic compiler give the error Internal error -- String space corrupt when another error (trapped or untrapped) was encountered? The problem was noticed on several large chained segments using the run time library. ------------------------------ Date: 8 February 1983 2217-mst From: Paul Schauble <Schauble @ M.PCO.LISD.HIS> Subject: C Language and Pointers To: Info-PC @ USC-ISIB Pardon me for responding this way, but you have stepped on one of my very sore toes. Subject: Re: Optimized C86 C Compiler From: Billy <BRACKENRIDGE@USC-ISIB> Would you explain to me how Computer Innovations can support 20-bit address space and still maintain compatibility with any other C language? It is fundamental to C that integers and pointers be the same size. I can't see any way out of this dilemma except the approach taken by Lattice. You just ain't going to have a 32-bit C on the 8086 until Intel announces a 32-bit successor to the 8086/8. Wrong! Wrong! Wrong! "5.6 Pointers are not integers You may notice in older C programs a rather cavalier attitude toward copying pointers. It has generally been true that on most machines a pointer may be assigned to an integer and back again without changing it; no scaling or conversion takes place, and no bits are lost. Regrettably, this has led to the taking of liberties with routines that return pointers which are then merely passed to other routines -- the requisite pointer declarations are often left out... This will work on many machines, since the default type for functions is <int> and <int> and pointer can usually be safely assigned back and forth. Nevertheless this kind of code is inherently risky, for it depends on details of implementation and machine architecture which may not hold for the particular compiler you use. It's wiser to be complete in all declarations. (The program <lint> will warn of such constructions, in case they creep in inadvertently.)" "The C Programming Language" Kernighan & Ritchie Notice that this was written in 1978. The equivalence happened to hold for the PDP-11, by accident. The authors acknowledge that it works, but is not proper use of the language. As C has been implemented on more and more machines, the ability to use the equivalence has been less and less. I am currently working with C on three machines. Of those, the equivalence between pointers and ints holds on one, fails on another, and probably (I haven't seen the compiler yet) on the third. One out of three. Having recently spent time trying to get Unix to run on a machine where the equivalence did NOT hold, I feel that code that depends on this "feature" is inherently brain-damaged. Paul ------------------------------ Date: 10 Feb 83 08:31:51 PST (Thu) From: npois!inuxd!stevens@Berkeley Subject: C Language and Pointers To: INFO-PC@ISIB It certainly is NOT fundamental to the C language that integers and pointers be the same size. It may be true of most implementations, but it was never guaranteed in the C Bible (Kernighan & Ritchie). Programs that do weird things like store pointers into integers are NON-PORTABLE. -- Scott Stevens -- npois!inuxd!stevens@berkeley ------------------------------ Date: 11 Feb 1983 2303-PST Subject: C Language and Pointers From: Dick Gillmann <GILLMANN@USC-ISIB> To: Info-PC cc: Brackenridge Paul and Scott, I must begin by saying that I'm not a C programmer. I read Kernighan and Ritchie's book on the language and I'm eager to try it on my PC (I've been writing in assembly language). To my amazement, I was unable to find a compiler that supported 20-bit addresses. This mystifies me. Why should I have to live in 64K? Not supporting the 20-bit address space seems like a major flaw. In fact, now that memory is so cheap, 20 bits seems rather limited. It's certainly true that writing programs that depend on coincidences of representation is bad practice. So why don't C compilers support 20-bit addresses as a matter of course? /Dick ------------------------------ Date: 7 Feb 83 9:05:16-PST (Mon) To: info-pc@isib From: teklabs!keithe@Berkeley.arpa Subject: HERCULES Graphics Card Responses to: "Why not just purchase the IBM monochrome card? The monochrome card works fine with monochrome monitors. The color card is awful with monochrome monitors. The Amdek 310 (A?) I saw was as good as the IBM monochrome monitor; the 310 was the display for the Columbia IBM PC look-alike." First, I find it difficult to believe that the IBM monochrome card will drive a standard (i.e. b&w version of NTSC rate video) monitor, because the IBM monochrome H-rate is 18.4 kHz (cf. 15.75 kHz for standard) and its V-rate is 50 Hz (cf 60). But I haven't tried it yet - I will this afternoon and if it DOES work I'll let you know. Second, I tried out the Hercules card and found it deficient for my used in its lack of compatibility with existing software. The only patch sent with it is to the BASIC interpreter; AND it only works with hi-res graphics (i.e. BASIC's SCREEN 2). So you can toss your compiled code out the window, and a lot of games (which use low-res, SCREEN 1) don't work. I called Hercules and they said they are going to supply additions to the BASIC compiler to improve compatibility, but what I THINK (but am not sure) they need to do is to supply a replacement BIOS ROM so that (almost?) everything will work with it. Enlightening criticisms/additions are welcomed. Keith Ericson ------------------------------ Date: 10 Feb 1983 0946-PST Subject: Re: HERCULES Graphics Card From: Randy Cole <COLE@USC-ISIB> To: teklabs!keithe@UCB-VAX, info-pc Keith: I believe your doubts about the IBM monochrome card driving a standard 15.75 kHz monitor are correct. However, my understanding is that the Amdek V310 monitor was designed (modified) to use the 18.4 kHz scan rate put out by the IBM monochrome board and is sold as a cheap alternative to the IBM monitor. Your comments on the Hercules card are very interesting. The lack of software is perhaps less drastic in the cases of languages like Pascal, Fortran, etc., which have no graphics primitives built in. I don't think a new ROM would be necessary to make such a card respond correctly to any BIOS call. What would be necessary would be to rewrite the graphics primitives and replicate any other code that used the same software interrupt vector, put that code into memory and protect it from being clobbered, and redirect the interrupt vector. Question: When in text mode, are the characters put out by the Hercules card identical to the higher-resolution characters the IBM monochrome card puts out, or are they 5x7 characters like the color/graphics card puts out? Randy Cole ------------------------------ Date: Thu 3 Feb 83 08:35:52-PST From: Hellmut Golde <GOLDE@WASHINGTON> Subject: Mailing List Program To: gillmann@USC-ISIB I have used PC-File from Dick Button in Bellevue; it is one of the Freeware programs. I find it simple to use, good to maintain mailing lists, and inexpensive. ($25 or $35). Jim Button, P.O. Box 5786, Bellevue, WA 98006. Just drop him a note and he will send you a floppy, I think. --Hellmut Golde ------------------------------ Date: 10 Feb 83 18:09:55-EST (Thu) From: D. J. Farber <farber@udel-relay> Subject: The Dangers of Coffee To: info-pc@isib I can report that PC keyboards are badly affected by coffee. Also diskettes do not work when they have been soaked with coffee. Dave ------------------------------ Date: Fri 11 Feb 83 10:59:49-PST From: Guillermo A. Loyola <CSL.Lantz.Gmo@SU-SCORE.ARPA> Subject: PC Uses for the Handicapped To: info-pc@USC-ISIB.ARPA cc: human-nets@RUTGERS.ARPA I'd like to hear from anybody doing work in the area of Personal Computer uses by handicapped persons. We have a coworker with cerebral palsy. Some software has been written for him using a speech synthesizer but a lot more is needed. The guy who wrote the software (who has no access to the net, but I can set up the contacts) would really like to start a dialog with people working in this area. Please replay to me directly with a U.S. Mail address and/or phone number. Thanks. Guillermo. ------------------------------ Date: 11 Feb 1983 1223-PST Subject: Assembly Language MODEM7 From: LEVYAL at USC-ISI To: info-pc at ISIB I once ran across a MODEM7 for the PC but can no longer locate it. I would appreciate any help. Need it in assembly language. Thanks, Allan ------------------------------ Date: 11 Feb 1983 22:35:11 EST (Friday) From: Martin Schoffstall <schoff at BBN-UNIX> Subject: Aparrat Memory Board To: info-pc at isib Being cheap, and also having about 30 64K Intel RAMs on hand, I ordered the Aparrat memory board with 64K installed and have now received it in Boston for $153. Included in the price is their pseudo disk software that allows you to implement a disk in memory in 64K * N (N=1 to 5) user selectable increments as drive D. They both work well except for some flakiness described below: With the board stuffed with 256K of RAM and all the proper switches set both on the system board and the RAM board at boot up the following memory errors showed up after four different boots 3caa201 3cf9201 3caa201 3ca0201 Not very consistent! Being a software person, I ignored it and used some canned software like 123 etc. for about an hour. I then booted the system three more times and guess what? NOTHING. Just a normal boot to DOS. Any ideas? How about cold memory chips!? schoff at bbn-unix ------------------------------ End of Info-PC Digest ****************************** -------