jshah@andromeda.rutgers.edu.rutgers.edu (Jigish Shah) (09/28/88)
does any one have a preference between MS C5.1 and Turbo C 2.0
holtz@beowulf.ucsd.edu (Fred Holtz) (09/28/88)
In article <876@galaxy> jshah@andromeda (Jigish Shah) writes: > >does any one have a preference between MS C5.1 and Turbo C 2.0 It's a pretty sad state of affairs when the first thing that came to mind upon reading the original subject line was "great, another lawsuit..." As for a preference, Turbo C is a much better deal, maybe the primary concern if you are paying for it personally. MSC may be a more mature product with a larger user base, but that doesn't make up for its cost and MicroSoft's reputation for poor support. But what do I know, I've been using Lattice C V3.1 for over two years without any upgrades! But seriously, if I were in the market for a new compiler I would go with the Borland product without hesitation (I have only looked at MSC 5.0 and TC 1.5, though...) Asbestos suit donned and return flamethrower manned!-)
bobmon@iuvax.cs.indiana.edu (RAMontante) (09/28/88)
Fred Holtz originally thought the subject implied "lawsuit". I'm not a big fan of Microsoft, but I do wanna tip my hat to them for responding to the Borland challenge by bringing out QuickC, "interrupt" function type, etc. _instead_ of just bringing out the lawyers. There's too much competition-by-legalistic-assault going on these days. -- -- bob,mon (bobmon@iuvax.cs.indiana.edu) -- "Aristotle was not Belgian..." - Wanda
williamo@hpcupt1.HP.COM (William O'Shaughnessy) (09/28/88)
Zortech's C++ compiler should definitely be in consideration when deciding on a C compiler. Zortech's C++ product is actually two compilers the C++ compiler and a C compiler. I've heard their support is good. I've never been happy with Borland's support, although I purchased Turbo C and payed an exorbident price for the upgrade to Turbo C 1.5. The above opinions are my personal opinions and do not necessarily reflect those of my employer. Bill O'Shaughnessy
cbenda@unccvax.UUCP (carl m benda) (09/29/88)
In article <876@galaxy>, jshah@andromeda.rutgers.edu.rutgers.edu (Jigish Shah) writes: > > does any one have a preference between MS C5.1 and Turbo C 2.0 Well let me think... which of the above runs under OS/2 and DOS?? hint, it IS NOT turbo 2.0... /Carl cbenda@unccvax
brb@akgua.ATT.COM (Brian R. Bainter) (09/29/88)
I have been very happy with Turbo C 1.0 and 1.5 and with Borland in
general. Borland's support has been very good for me. When I have had
a problem (which has been seldom) I have been able to talk to a real
person who is able to either give me a work around or a patch.
While working for a certain company, we were using the Microsoft
compiler. When we found bugs or problems, we were not treated very
well and often could not even talk to anyone. They were fairly
unresponsive.
Borland also has my vote when it comes to speed and price. The execution
speed is generally as fast if not faster, the compilation speed is faster,
and the code is often smaller. These are of course my observations and may
not reflect the accepted norm. The price is a lot better too for the product
that is received. When I was teaching a college C course, I was able to get
coupons for my students for the full compiler for $44.95. The listed price
is very competitive as well.
Hope this helps,
--
Brian R. Bainter KA7TXA
AT&T Technologies Atlanta Works
{cbosgd, gatech, ihnp4, moss, mtune, ulysses}akgua!brbwilliamo@hpcupt1.HP.COM (William O'Shaughnessy) (09/29/88)
TurboC does have some very positive attributes: 1. An informative and friendly interactive user interface. 2. A simplified, easy to use MAKE called "Project" along with a regular Make. 3. An integrated debugger available with TurboC 2.0. The above are my personal opinions and are not necessarily those of my employer. Bill O'Shaughnessy
noren@dinl.uucp (Charles Noren) (09/29/88)
In article <5930017@hpcupt1.HP.COM> williamo@hpcupt1.HP.COM (William O'Shaughnessy) writes: |Zortech's C++ compiler should definitely be in consideration when |deciding on a C compiler. Zortech's C++ product is actually two |compilers the C++ compiler and a C compiler... Does Zortech (or MS C) provide all the graphics functions of Turbo C? I like Turbo C because it supports my clone Herc Card. -- Chuck Noren NET: ncar!dinl!noren US-MAIL: Martin Marietta I&CS, MS XL8058, P.O. Box 1260, Denver, CO 80201-1260 Phone: (303) 971-7930
andrews@hpcupt1.HP.COM (Edward E. Andrews) (09/29/88)
But can I write WINDOWS applications with Turbo C?
johnl@ima.ima.isc.com (John R. Levine) (09/29/88)
In article <1133@unccvax.UUCP> cbenda@unccvax.UUCP (carl m benda) writes: >In article <876@galaxy>, jshah@andromeda.rutgers.edu (Jigish Shah) writes: >> does any one have a preference between MS C5.1 and Turbo C 2.0 >Well let me think... which of the above runs under OS/2 and DOS?? >hint, it IS NOT turbo 2.0... Which of the above is available for $67? Sure as heck not MS C. I have both MS C 5.1 and Turbo 1.5 (since the 800 number for the 2.0 upgrade has been busy every time I've called.) Depending on what I'm doing, sometimes I use one, sometimes the other. MS C with the optimization turned all the way on has a more sophisticated optimizer, though it doesn't seem to generate code all that much better than Turbo. Turbo, on the other hand, supports multiple addressing models in the integrated environment and also allows in-line assembler as well as explicit reference to machine registers in C code, sometimes useful in grotty low-level stuff. For the price there's no doubt in my mind that Turbo is the way to go. If somebody else is paying for it, take your pick. -- John R. Levine, IECC, PO Box 349, Cambridge MA 02238-0349, +1 617 492 3869 { bbn | think | decvax | harvard | yale }!ima!johnl, Levine@YALE.something Rome fell, Babylon fell, Scarsdale will have its turn. -G. B. Shaw
mr@homxb.UUCP (mark) (09/30/88)
In article <876@galaxy>, jshah@andromeda.rutgers.edu.rutgers.edu (Jigish Shah) writes: > > does any one have a preference between MS C5.1 and Turbo C 2.0 For one, TC2.0 is much less expensive. I use both. mark homxb!mr
mr@homxb.UUCP (mark) (09/30/88)
In article <1133@unccvax.UUCP>, cbenda@unccvax.UUCP (carl m benda) writes: # In article <876@galaxy>, jshah@andromeda.rutgers.edu.rutgers.edu (Jigish Shah) writes: # > # > does any one have a preference between MS C5.1 and Turbo C 2.0 # # # # Well let me think... which of the above runs under OS/2 and DOS?? # hint, it IS NOT turbo 2.0... Most often, when I write C code, I write it to compile everywhere; On UNIX, DOS, OS/2, etc. With any compiler; UNIX cc, tc, msc, etc. (at least I try for this ideal) # # /Carl # cbenda@unccvax mark homxb!mr
suitti@haddock.ima.isc.com (Steve Uitti) (09/30/88)
In article <876@galaxy> jshah@andromeda (Jigish Shah) writes: > >does any one have a preference between MS C5.1 and Turbo C 2.0 I have MSC 5.2 and (currently) Turbo C 1.4. Turbo C 1.4: + Takes up far less disk space for a usable running compiler. + Costs less (was $65 vs > $300 for me) + Runs three times as fast + Produces code that runs about the same speed as MSC 5.1 + Produces code that is generally slightly smaller than MSC 5.1 + Has some of the best lint-like capabilites I've seen + Produces code that works. - Much C code has not been ported to it. [I see here on the net that newer versions have new features, and may fix things]. MSC 5.1 + Has a prototypes generator + Has a debugger (codeview) + Comes with a Turbo C-like compiler (quick C), which is unfortunately brain dead + Has a code omptimizer (which you can't use, because it breaks at least something when used in almost any largeish package) + Has supports OS/2 (I don't care) - Takes up to 5.5 MB of your hard disk for std libraries - Is slow - Often produces code that doesn't work - Doesn't always parse correct C code properly - Is amazingly complex - Costs an arm and a leg - Isn't better than Turbo C - Much C code has not been ported to it. This last point, "Much C code has not been ported to it", is a minus for both compilers. The truth is that MSC 4.0 has lots of stuff ported to it. None of the stuff I've tried works under MSC 5.0 or 5.1. The compiler has given me almost nothing but trouble. ** flame on (this is the point of this note) ** I would very much like to see us (the user community) support good inexpensive compilers, like Turbo C. Not (just) by buying it, but by using it (writing stuff in it, porting stuff to it). I only bought MSC so that I could compile other people's stuff. I bought the wrong compiler! MSC 5.1 != MSC 4.0 !!! I can't stand writing anything with the compiler, because it takes so damned long for it to get to the next spot where it won't compile my code. Don't you dare tell me to get a faster machine. I timed it. Turbo C compiles three times faster. If I had a faster machine, Turbo C would still be three times faster. When I upgrade to the new 137 MHz 80986 based box, Turbo C will compile 437,000 lines per minute and MSC 5.1 will only compile 145,000 lines per minute. Stephen.
english@stromboli.usc.edu (Joe English) (09/30/88)
In article <1133@unccvax.UUCP> cbenda@unccvax.UUCP (carl m benda) writes: >In article <876@galaxy>, jshah@andromeda.rutgers.edu (Jigish Shah) writes: >> does any one have a preference between MS C5.1 and Turbo C 2.0 >Well let me think... which of the above runs under OS/2 and DOS?? >hint, it IS NOT turbo 2.0... And another consideration: if you ever want to do any Windows programs, you *have* to use MSC. In order for Windows to do its code-swapping and multi-tasking stuff, every subroutine has to do some weird stack frame manipulation in the prologue. Which, of course, the MS compiler can do and Turbo can't (1.0 at least). Turbo 2.0 might have this, but I haven't heard so. Does anyone else know? BTW, I do *not* reccommend doing Windows programming unless you are going to get paid a lot of money for it :-) /|/| "If you think this is bad, you should see what it looks like -----< | | in *here*!" O \|\| english%lipari@oberon.usc.edu
erc@unisec.usi.com (Ed Carp) (09/30/88)
In article <2722@ima.ima.isc.com>, johnl@ima.ima.isc.com (John R. Levine) writes: > > Which of the above is available for $67? Sure as heck not MS C. I > > For the price there's no doubt in my mind that Turbo is the way to go. If > somebody else is paying for it, take your pick. Have you tried Quick C? It's aout $75, has an integrated environment, and the compiler is pretty quick! I am using it for all of my development work at the moment. Since I don't care about OS/2 (I think that it is even worse than MSDOS), and I don't believe that the market (software and job) is there for OS/2, a compiler that supports MSDOS and most of the UNIX system calls (those that can be supported on a single-user box, anyway) is important for me to have. I have written CURSES for DOS, signal(SIGALRM)-like functions (for both DOS ad UNIX), and have been eminently satisfied with Quick C. Any comments? -Ed
cbenda@unccvax.UUCP (carl m benda) (09/30/88)
In article <2722@ima.ima.isc.com>, johnl@ima.ima.isc.com (John R. Levine) writes: > In article <1133@unccvax.UUCP> cbenda@unccvax.UUCP (carl m benda) writes: > >In article <876@galaxy>, jshah@andromeda.rutgers.edu (Jigish Shah) writes: > >> does any one have a preference between MS C5.1 and Turbo C 2.0 > >Well let me think... which of the above runs under OS/2 and DOS?? > >hint, it IS NOT turbo 2.0... > > Which of the above is available for $67? Sure as heck not MS C. I > have both MS C 5.1 and Turbo 1.5 (since the 800 number for the 2.0 upgrade > has been busy every time I've called.) Depending on what I'm doing, The point I am trying to make is that if you would like to someday (today) use a C compiler that runs in OS/2 protect mode, It sure as heck is not Turbo. Try creating a 4Mb array using Turbo... This brings me to my second gripe, and that is with the misleading advertis- ing of the Turbo assembler which Borland claims is 100% compatable with MASM 5.1.. This is not exactly true because TASM won't run in OS/2 protect mode. Now there's compatablity right??? MASM will run, TASM won't. /Carl cbenda@unccvax
sullivan@marge.math.binghamton.edu (fred sullivan) (09/30/88)
In article <1133@unccvax.UUCP> cbenda@unccvax.UUCP (carl m benda) writes: >In article <876@galaxy>, jshah@andromeda.rutgers.edu.rutgers.edu (Jigish Shah) writes: >> does any one have a preference between MS C5.1 and Turbo C 2.0 >Well let me think... which of the above runs under OS/2 and DOS?? >hint, it IS NOT turbo 2.0... And do you know any more advantages of turbo 2.0? Fred Sullivan SUNY at Binghamton Dept. Math. Sciences Binghamton, NY 13903 sullivan@marge.math.binghamton.edu First you make a roux!
jmoore@pc.ecn.purdue.edu (James D Moore) (09/30/88)
In article <5930019@hpcupt1.HP.COM> andrews@hpcupt1.HP.COM (Edward E. Andrews) writes: >But can I write WINDOWS applications with Turbo C? Personaly I do not find WINDOWS (assuming you are speaking of MS stuff) to be all that great. I vote for Turbo-C. Besides, How much will the WINDOWS development cost you ? It is not cheap I am sure of that. Borlands products and support are by far nicer that any MS support I have ever had. Jim Moore Purdue University Of course this is only my oppinion and not that of my employer.
markd@proxftl.UUCP (Mark Davidson) (09/30/88)
I certainly hope this doesn't start a flame war, but I'd like to throw in some comments in the endless "Turbo C/MS C/Quick C" discussion. I own both MSC 5.1 and Turbo C 1.5 (2.0 soon, hopefully). Although MSC's documentation is many times better than Turbo C's, I still like Turbo C, especially when I'm initially working on a program (mainly because Turbo's compile/link times are so much better). My experience with Quick C has not been a happy one. I was working on a project at my last job (a few months ago) and I started out using Quick C. I ended up completing the job with my own copy of Turbo C. Quick C is extremely picky about equipment configuration (especially display cards), can only debug medium model code in the integrated environment (I wasn't using medium model), and on several occasions took valid C code and generated object code that was never going to work correctly. I was amazed at all the problems I had with it. I don't know about you people, but my jaw about hit the ground when I ran CodeView on a program I had compiled with Quick C and saw that it had turned an if statement into a comparison that had nothing to do with the values I was comparing! Call me naive, but I had never had a compiler I was using generate just plain wrong code. -- In real life: Mark E. Davidson uflorida!novavax!proxftl!markd Proximity Technology Inc., 3511 NE 22nd Ave, Ft. Lauderdale FL, 33308 #define STANDARD_DISCLAIMER <Quote construction site>
wsd@whuts.UUCP (DINSMORE) (09/30/88)
In article <5930019@hpcupt1.HP.COM> andrews@hpcupt1.HP.COM (Edward E. Andrews) writes: >But can I write WINDOWS applications with Turbo C? You most certainly can! If you're talking strictly Mirosoft Windows, I don't know and don't care since Microsoft windows are useless to me. I currently use C-Spot-Run (CSR) windows library. Damn easy to use and produces excellent code. I now have V3.0 and since it is shareware I might be able to post it, but even easier I can give you the BBS number to get all by yourself! C Spot Run A User-Supported C Add-on Library Version 3.0 (Supports Turbo-C V1.5, MSC V5.x, and Quik-C) July 12,1988 Bob Pritchett New Dimension Software 23 Pawtucket Dr. Cherry Hill, NJ 08003 Voice: (609) 424-2595 Data: (609) 354-9259 FidoNet 107/414 $15 to register, $50 gets source, $75 for commercial license. I have absolutely no affiliation with New Dimension Software, just a very satisfied user. -- | Wayne S. Dinsmore Make it in Massachusetts, | AT&T Bell Labs (ihnp4,att,moss)!whuts!wsd Spend it in New Hampshire. | 20 Shattuck Road Room 4A-118 | Andover,Mass 01810
klotz@ihlpg.ATT.COM (Dave Klotzbach) (09/30/88)
> although I purchased Turbo C and payed an exorbident price for > the upgrade to Turbo C 1.5. > Bill O'Shaughnessy Compared to MS upgrades how can you possibly consider $49 for an upgrade and a Graphics Subroutine Package, exorbident? -- Dave Klotzbach ihnp4!ihlpl!klotz IE 2F-518 (312) 416-7437
suitti@haddock.ima.isc.com (Steve Uitti) (10/01/88)
In article <839@proxftl.UUCP> markd@proxftl.UUCP (Mark Davidson) writes: >...Although MSC's documentation is many >times better than Turbo C's, I still like Turbo C,... What?! With MSC, you get lots of documentation. You need to read it all. It doesn't read like a book. It can be hard enough to find the right section to look stuff up in an index. The manual comes in ring binders. The 5.1 updates came without binders, but the original ring binders were already full! Ring binders are a pain to deal with, if for example, you want to read the thing while commuting to work on the train... With Turbo C (1.4), the manuals are two edge bound soft cover books. If you already know C, you only have to read about three or four chapters from one book. The other book describes the libraries. You can read the one book, and index the other. You can get started quickly, and continue quickly. Thats what documentation is for. More is not better. (Although some say "less" is "more", but that's a lie too). I've only written letters to Microsoft to complain about support and ask stuff. I received a reply. Though it answered my questions it was not especially helpful. I've never required support for Turbo. The compiler has NEVER crashed. The best support is not having to use it. I no longer develop stuff in MSC. Stephen.
DOHC@TUCCVM.BITNET (Bob Roberds) (10/01/88)
>But can I write WINDOWS applications with Turbo C?
No. But can you write WINDOWS apps w/o the obscenely expensive Windows
Development Kit, either?jjboritz@violet.waterloo.edu (Jim Boritz) (10/01/88)
I have had most of my experience with MSC 5.1, and I have "played" with
Turbo C a bit.
Turbo is faster.
Turbo C 1.5 is/was cheap, and didn't come with a debugger.
Turbo C 2.0 is more expensive, and a debugger still costs more on top of
that.
MSC 5.1 supports MS Windows and it supports OS/2.
MSC has had Codeview for a long time now. There are also versions available
for Windows and for OS/2.
Compilers in general are not toys. They are very complex pieces of
software. They have large manuals and lots of options.
If all you are interested in is DOS and porting from *NIX systems, then the
choice is tough to make, but Turbo would probably win right now because of
speed.
If you are looking towards the future and towards other environments, then
there is no choice. MSC is the the only candidate. It is a complete,
robust compiler with lots of flexibility.
But wait, there is now another contender in the C compiler market.
Watcom C 6.5. It has received consistently good reviews and produces code
that is without a doubt faster that either MSC or Turbo C.
--
--------------------------------------------------------------------------------
Jim Boritz jjboritz@violet.waterloo.edu
University of Waterloo {uunet,utai,clyde}!watmath!violet!jjboritzmadd@bu-cs.BU.EDU (Jim Frost) (10/02/88)
In article <1133@unccvax.UUCP> cbenda@unccvax.UUCP (carl m benda) writes: |In article <876@galaxy>, jshah@andromeda.rutgers.edu.rutgers.edu (Jigish Shah) writes: |> does any one have a preference between MS C5.1 and Turbo C 2.0 | |Well let me think... which of the above runs under OS/2 and DOS?? |hint, it IS NOT turbo 2.0... "Compatibility Box" comes to mind. I believe you meant to say "which of the above makes executables for OS/2 and DOS", in which case you'd be right. People I respect that are forced to use the MSC compiler have no nice things to say about it. If you're not interested in OS/2, I'd go for something else. I personally like the Turbo environments (better than ever with a debugger, finally) but I keep hearing good things about the Zortec (?) compiler. jim
bobmon@iuvax.cs.indiana.edu (RAMontante) (10/03/88)
jjboritz@violet.waterloo.edu (Jim Boritz) writes: > >Turbo C 2.0 is more expensive, and a debugger still costs more on top of >that. Turbo C v2.0, plus the Turbo Debugger and the (claimed) MASM-compatible Turbo Assembler, is being listed for $250 and will probably retail for comfortably less than $200 if past experience is a guide. (It'll cost me $100 to upgrade to these from my v1.5). The code compares in performance to MSC v5.1, with the development environment of QuickC. It does support EMS, although not OS/2. I have an EMS board, I'm not running OS/2. How much will you spend for MSC v5.1 (with a QuickC that only supports one memory model), plus MASM ?? I don't see MSC's future portability being any better than Turbo C's, but then I do deal with UNIX and I don't (won't) deal with OS/2. -- -- bob,mon (bobmon@iuvax.cs.indiana.edu) -- "Aristotle was not Belgian..." - Wanda
jpn@teddy.UUCP (John P. Nelson) (10/03/88)
In article <1142@unisec.usi.com> erc@unisec.usi.com (Ed Carp) writes: >In article <2722@ima.ima.isc.com>, johnl@ima.ima.isc.com (John R. Levine) writes: >> For the price there's no doubt in my mind that Turbo is the way to go. If >> somebody else is paying for it, take your pick. > >Have you tried Quick C? Yes. It's pretty terrible. >It's aout $75, has an integrated environment, In which only one memory model is available. That makes it a TOY, in my opinion, not a serious development tool! >and the compiler is pretty quick! In the tests that I did, the compiler was CONSISTENTLY slower (by as much as 2 to 1) than Turbo C, and the code generated was CONSISTENTLY slower and larger! >I am using it for all of my development work at the moment. There's not accounting for taste. >Since I don't care about OS/2 (I think that it is even worse than MSDOS), At least we agree on something :-) About the only advantage Quick C had over Turbo C 1.5 was the debugger capability. However, with Turbo 2.0, this advantage is gone. -- john nelson UUCP: {decvax,mit-eddie}!genrad!teddy!jpn smail: jpn@teddy.genrad.com
pjh@mccc.UUCP (Pete Holsberg) (10/03/88)
In article <13426@iuvax.cs.indiana.edu> bobmon@iuvax.UUCP (RAMontante) writes: ...jjboritz@violet.waterloo.edu (Jim Boritz) writes: ...> ...>Turbo C 2.0 is more expensive, and a debugger still costs more on top of ...>that. Jim, Turbo C 2.0 has a built-in debugger. Borland has a wonderful policy for educational sales that will bring any student or faculty member TC 2.0 for $49.95 and the Assembler+Debugger for another $49.95. THis is NOT an upgrade price. What's this OS/2 you're talking about? One of those vaporware products? :-) Pete Holsberg UUCP: {...!rutgers!}princeton!mccc!pjh Technology Division ...!att!jonlab!mccc!pjh Mercer College CompuServe: 70240,334 1200 Old Trenton Road GEnie: PJHOLSBERG Trenton, NJ 08690 Voice: 1-609-586-4800
prime@druhi.ATT.COM (Anthony Davis) (10/04/88)
In article <8325@haddock.ima.isc.com>, suitti@haddock.ima.isc.com (Steve Uitti) writes: > .......(stuff deleted) > With Turbo C (1.4), the manuals are two edge bound soft cover ^ Do you mean 1.5 or was there a release I missed? Tony Davis AT&T Bell Labs, Denver Co. local: druhi!prime net: att!druhi!prime
cbenda@unccvax.UUCP (carl m benda) (10/04/88)
In article <25171@bu-cs.BU.EDU>, madd@bu-cs.BU.EDU (Jim Frost) writes: > In article <1133@unccvax.UUCP> cbenda@unccvax.UUCP (carl m benda) writes: > |In article <876@galaxy>, jshah@andromeda.rutgers.edu.rutgers.edu (Jigish Shah) writes: > |> does any one have a preference between MS C5.1 and Turbo C 2.0 > | > |Well let me think... which of the above runs under OS/2 and DOS?? > |hint, it IS NOT turbo 2.0... > > "Compatibility Box" comes to mind. I believe you meant to say "which > of the above makes executables for OS/2 and DOS", in which case you'd NO!!!! I don't even have the compatability box set up!!! It takes up too much memory! MSC 5.1 is specifically created to run in OS/2 PROTECT mode. This is why it is ludicrous to even make the comparison. While we're at it lets just through in the VAX VMS C compiler for comparison on a price only basis since you people insist on comparing the compilers on that level instead of which OS they R*U*N on!! > jim Sorry for the minor flame. I will concede that I use and LIKE turbo on an old XT I use and would never consider loading 5.1 onto it but please let Microsoft charge a little more for a compiler that has to DO much more namely work under OS/2.. This is a feat in itself and diserves the gold star for Intel 286 protected mode programming excellence. /Carl cbenda@unccvax
jpn@teddy.UUCP (John P. Nelson) (10/04/88)
In article <1141@unccvax.UUCP> cbenda@unccvax.UUCP (carl m benda) writes: >Sorry for the minor flame. I will concede that I use and LIKE turbo on >an old XT I use and would never consider loading 5.1 onto it but please >let Microsoft charge a little more for a compiler that has to DO much more >namely work under OS/2.. This is a feat in itself and diserves the gold >star for Intel 286 protected mode programming excellence. Hang on a second there! I didn't ask Microsoft to implement a single compiler that runs/targets two different operating systems simultaneously. Frankly, I think it's a silly thing for them to have done. It makes each tool much bulkier than otherwise, because each task must include the libraries for both OS's. Frankly, I don't understand why anyone would consider this a "good thing". If I want to use OS/2, I'll buy an OS/2 compiler. If I want to use MSDOS, I'll buy an MSDOS compiler. I don't use OS/2, and I don't PLAN to use OS/2 in the near future. So why should I pay for all the OS/2 compiler development costs? Actually, I DO have MSC 5.1, but only because microsoft offered a $25 MSDOS-only upgrade from 5.0 (which was MSDOS only). Of course, the MSDOS-only 5.1 upgrade package was identical to the $75 OS/2 upgrade (which makes no sense to me either). In any case, I can't afford 5.5 Meg of disk for the MSC compiler package, so I use Turbo C exclusively now. Anyone want to buy an OS/2 compiler cheap? -- john nelson UUCP: {decvax,mit-eddie}!genrad!teddy!jpn smail: jpn@teddy.genrad.com
ralf@b.gp.cs.cmu.edu (Ralf Brown) (10/05/88)
In article <3632@druhi.ATT.COM> prime@druhi.ATT.COM (Anthony Davis) writes: }In article <8325@haddock.ima.isc.com>, suitti@haddock.ima.isc.com (Steve Uitti) writes: }> With Turbo C (1.4), the manuals are two edge bound soft cover }Do you mean 1.5 or was there a release I missed? I believe Steve is referring to TC 1.0 + three patch sets, also known as 1.0D. You can tell how many patches have been applied by looking at the third character of the serial number: My TC 1.0 was 1.0B because the serial number was ?1B??????; it had the first patch applied. BTW, TC 1.5 serial numbers are ?1E?????. -- {harvard,uunet,ucbvax}!b.gp.cs.cmu.edu!ralf -=-=- AT&T: (412)268-3053 (school) ARPA: RALF@B.GP.CS.CMU.EDU |"Tolerance means excusing the mistakes others make. FIDO: Ralf Brown at 129/31 | Tact means not noticing them." --Arthur Schnitzler BITnet: RALF%B.GP.CS.CMU.EDU@CMUCCVMA -=-=- DISCLAIMER? I claimed something?
jlh@loral.UUCP (Physically Pffft) (10/06/88)
In article <5034@teddy.UUCP> jpn@teddy.UUCP (John P. Nelson) writes:
=I don't use OS/2, and I don't PLAN
=to use OS/2 in the near future. So why should I pay for all the OS/2
=compiler development costs?
Well, if microsoft spread the develpement costs among the OS/2 users then
the compilers would be around $100,000 per copy (only half a :-) there).
Jim
--
Jim Harkins
Loral Instrumentation, San Diego
{ucbvax, ittvax!dcdwest, akgua, decvax, ihnp4}!ucsd!sdcc6!loral!jlhcarroll@s.cs.uiuc.edu (10/06/88)
About porting - I have Zortech and MSC 5.1. I've been having
some problems with a particular program that I compiled with MSC, so
I thought "Hey, I'll just use Zortech, and see if it's a compiler
problem". False. Zortech totally choked on the code. I couldn't find
a way to turn off its desire to do function prototype checking. It
was also missing include files such as "sys/types.h". So I'd have to
give only marginal marks as a C compiler. On the other hand, although
I've only done some small programs, I like it as a *C++* compiler,
which is what I bought it for anyway.
P.S. The code was Doug McDonald's DVIVGA program, which is based on
Nelson Beebe's DVIxxx programs.
Alan M. Carroll "How many danger signs did you ignore?
carroll@s.cs.uiuc.edu How many times had you heard it all before?" - AP&EW
CS Grad / U of Ill @ Urbana ...{ucbvax,pur-ee,convex}!s.cs.uiuc.edu!carrollmcdonald@uxe.cso.uiuc.edu (10/08/88)
>About porting - I have Zortech and MSC 5.1. I've been having >some problems with a particular program that I compiled with MSC, so >I thought "Hey, I'll just use Zortech, and see if it's a compiler >problem". False. Zortech totally choked on the code. I couldn't find >a way to turn off its desire to do function prototype checking. It >was also missing include files such as "sys/types.h". So I'd have to >give only marginal marks as a C compiler. On the other hand, although >I've only done some small programs, I like it as a *C++* compiler, >which is what I bought it for anyway. >P.S. The code was Doug McDonald's DVIVGA program, which is based on >Nelson Beebe's DVIxxx programs. ^^^^^^ I wrote that program. It is not a fair compiler test, not even for IBM-PC compilers. It is full of MSC-5.1-ism's (proprietary function calls), and will die during compilation without some bizarre compile-line switches ( -DANSI_LIBRARY -DANSI_PROTOTYPES). It is a nice program (advertisement), just blatently non-portable. All the Beebe programs are dependent on the compile-line things, depending on the operating system. Doug McDonald
bright@Data-IO.COM (Walter Bright) (10/11/88)
In article <213400010@s.cs.uiuc.edu> carroll@s.cs.uiuc.edu writes: >I couldn't find >a way to turn off its (Zortech C's) desire to do function prototype checking. Try the -p switch. >It was also missing include files such as "sys/types.h". types.h is not an ANSI standard file. It contains some definitions that are in other h files (such as time_t and size_t). It also has some pointless typedefs such as: typedef unsigned char u_char; Really, what's that worth? It also has some unix file system specific defs that are irrelevant on the PC.
todd@metheus.UUCP (Todd Stewart) (10/25/88)
In article <1709@dataio.Data-IO.COM> bright@dataio.Data-IO.COM (Walter Bright) writes: >... It also has some pointless typedefs such as: > typedef unsigned char u_char; >Really, what's that worth? Mostly it takes the sting out of saying what you really mean. "unisigned char" is so unwieldy that most go ahead and use default signed data types even when handling unsigned data e.g. array indices, etc. These subtle bugs go undetected by the original authors, but break in nasty and hard to debug ways when the code is later ported to a different configura- tion or host. I'm all for any typedef convention that encourages accurate programming by making it reasonable to say what you really mean. Todd Stewart Metheus Corporation