schanck@salmon.cis.ohio-state.edu (Christopher Schanck) (01/15/89)
There has been a lot of discussion of the TC debuggers, integrated and standalone, some of it started by me. Regardless, as a user of TC since 1.0, I have a healthy respect for it as a compiler. Currently I am using version 1.5, and I can get 2.0 for $50 through OSU (can't get the debugger on an educational discount, boo hiss!). What I want to know is could someone(s) post a convincing reason or set of reasons for me to upgrade to 2.0; remember, all I really care about is the compiler. My system is an 8088 (ok, it's a V20) with a hard drive, and a CGA card. So convince me! :-) Thanks, Chris -=- "My brain is NOT a deadlock-free environment!!!!" --- Christopher Schanck, mammal at large. schanck@flounder.cis.ohio-state.edu
johnm@trsvax.UUCP (01/16/89)
>the debugger on an educational discount, boo hiss!). What I want to >know is could someone(s) post a convincing reason or set of reasons >for me to upgrade to 2.0; remember, all I really care about is the >compiler. My system is an 8088 (ok, it's a V20) with a hard drive, and >a CGA card. So convince me! :-) I'm sure others will post a variety of helpful reasons but here is one that I'll bet will be missed. The help system. THELP is a resident TSR that takes up only 8K and stays resident on ALL my machines. It gives me the handy C support that I need. Every day (almost) we end up asking ourselves what is the order of arguments in fread? What are the names of the structure elements in the ANSI time structure? What include file do I need when I use the findfirst() call... THELP gives me that INSTANTLY without having to take my hands off the keyboard. I love it. The organization of the help system is a hypertext one. That allows you to be looking at one of the ANSI time calls and then easily flip the cursor down to the other listed calls at the bottom of the screen, press ENTER, and view the syntax of the function. Everything is linked. If you take a look at it, it will definitely sway you. John Munsch
pjh@mccc.UUCP (Pete Holsberg) (01/17/89)
In article <31304@tut.cis.ohio-state.edu> Christopher Schanck <schanck@cis.ohio-state.edu> writes:
=am using version 1.5, and I can get 2.0 for $50 through OSU (can't get
=the debugger on an educational discount, boo hiss!). What I want to
Sure you can! Call Educational Sales at Borland. The
Debugger/Assembler package is also $49.95 to qualified educational folks.
--
Pete Holsberg UUCP: {...!rutgers!}princeton!mccc!pjh
Mercer College CompuServe: 70240,334
1200 Old Trenton Road GEnie: PJHOLSBERG
Trenton, NJ 08690 Voice: 1-609-586-4800rchen@m.cs.uiuc.edu (01/17/89)
>/* Written 12:39 am Jan 15, 1989 by schanck@salmon.cis.ohio-state.edu */ >/* ---------- "TC 2.0 Pros/Cons" ---------- */ >... What I want to >know is could someone(s) post a convincing reason or set of reasons >for me to upgrade to 2.0; remember, all I really care about is the >compiler. My system is an 8088 (ok, it's a V20) with a hard drive, and >a CGA card. So convince me! :-) > I'd say hold your horse on TC 2.0. I am having several problems to convert my TC 1.5 programs to TC 2.0 right now. The dummiest one is that in some cases (I don't know why) all TC graphics fonts look different (mainly in size) from before when using TC 2.0. In some other cases (I don't know why either), the fonts look the same but the textwidth() function returns different values under different compilers. Believe me, several people in our office experienced the same problem. We checked the compilers and the libraries we used. Still, no one can tell what's going on. By the way, we are using PS-2/50's with VGA board. We had no trouble with TC 1.5. -Ron Chen Office for Information Management University of Illinois at Urbana-Champaign
rja@edison.GE.COM (rja) (01/18/89)
In article <8000033@m.cs.uiuc.edu>, rchen@m.cs.uiuc.edu writes: > I'd say hold your horse on TC 2.0. I am having several problems > to convert my TC 1.5 programs to TC 2.0 right now. There are definitely some differences between TC 1.5 and 2.0. The changed behaviour of cprintf() with respect to '\n' is a good example. It isn't a bug since it is clearly documented and cprintf() isn't portable anyway. The main things I've noticed are changes to make TC more nearly ANSI C and more portable to non-DOS systems. I also use C on VMS and UNIX VAXen and am finding it much easier to port programs amongst all 3 with TC 2.0 as compared with before. I would say to get TC 2.0 and which is a much better, faster compiler that is more portable to ANSI/POSIX systems. ______________________________________________________________________________ rja@edison.GE.COM or ...uunet!virginia!edison!rja via Internet (preferable) via uucp (if you must) ______________________________________________________________________________
bkbarret@sactoh0.UUCP (Brent K. Barrett) (01/23/89)
In article <1779@edison.GE.COM>, rja@edison.GE.COM (rja) writes: > In article <8000033@m.cs.uiuc.edu>, rchen@m.cs.uiuc.edu writes: > > > I'd say hold your horse on TC 2.0. I am having several problems > > to convert my TC 1.5 programs to TC 2.0 right now. > > There are definitely some differences between TC 1.5 and 2.0. [CHOMP!] Another problem when upgrading to TC 2.0 is how 2.0 deals with bit fields. In 1.5, bit fields are expanded out to the byte, in 2.0 they are expanded out to the word. A project I maintain (a Citadel BBS system) uses bit fields for most user flags, and upgrading was a pain until I discovered this little item. It seemed to want to add a byte to each log record. Luckily I was informed by a friend who also ran into the problem. Wish Borland would have told us. -- "Somebody help me! I'm trapped in this computer!" Brent Barrett ..pacbell!sactoh0!bkbarret GEMAIL: B.K.BARRETT
cww@ndmath.UUCP (Clarence W. Wilkerson) (01/23/89)
I like this aspect of Turbo C, but use of it makes my programs harder to port to VAX or SUN unix c. GCC is ok though.
jsilva@cogsci.berkeley.edu (John Silva) (01/25/89)
In article <642@sactoh0.UUCP> bkbarret@sactoh0.UUCP writes: > > Another problem when upgrading to TC 2.0 is how 2.0 deals with bit >fields. In 1.5, bit fields are expanded out to the byte, in 2.0 >they are expanded out to the word. > > A project I maintain (a Citadel BBS system) uses bit fields for >most user flags, and upgrading was a pain until I discovered this >little item. It seemed to want to add a byte to each log record. >Luckily I was informed by a friend who also ran into the problem. > > Brent Barrett ..pacbell!sactoh0!bkbarret GEMAIL: B.K.BARRETT Perhaps bit fields would be byte expanded if the compilation switch pertaining to byte/word alignment were set to byte alignment. -J. --- John P. Silva INTERNET : jsilva@cogsci.berkeley.edu "You don't know what you're UUCP : {backbone}!ucbvax!cogsci!jsilva getting into, friend..."
bkbarret@sactoh0.UUCP (Brent K. Barrett) (01/26/89)
In article <9089@pasteur.Berkeley.EDU>, jsilva@cogsci.berkeley.edu (John Silva) writes: > Perhaps bit fields would be byte expanded if the compilation switch pertaining > to byte/word alignment were set to byte alignment. That's obvious, but wasn't the point of my message. They should have TOLD us the default had changed, and they didn't. -- "Somebody help me! I'm trapped in this computer!" Brent Barrett ..pacbell!sactoh0!bkbarret GEMAIL: B.K.BARRETT