🚀 go-pugleaf

Thanks a lot to all who responded to my request.

The question was:
>    I need to write a system in C to acquire and process real-time
> data (from an A/D board) and also to display the results in a
> graphical fashion.  My proposed platform is an IBM compatible 386
> system with math coprocessor and VGA monitor running under MS-DOS.  I
> do not plan to run under Windows 3.0.  My question is simply which C
> should I use, Turbo or Microsoft?

The consensus seems to be that Turbo-C is the way to go.  The reasons
cited are listed below.  Since I'm only parroting information provided
by others, I can't guarantee the veracity of any of these remarks.

Pro Turbo C:

1) Compiles much faster than Microsoft.

2) Graphics library (non Windows) is a little better.

3) Superior setup and selection of options.

4) Friendlier and more functional debugging capability.

5) Friendlier tech support.

6) Library and language support for interrupts.

7) Ability to do direct video writes (a pro or a con depending on your
   religious convictions :-)).

Pro Microsoft C:

1) The executable code is probably slightly more efficient (time and
   space wise).

2) ? The Microsoft C library is slightly more unix compatible

3) The runtime library provides model independent forms of all the
   memXXX and strXXX functions.

Other comments:

1) Turbo C's ability to do direct video writes speeds up program
   execution so even if Microsoft C calculates faster, you get it
   all back if you do any graphics work.

2) One person indicated that I'm on the wrong track altogether.
   Rather than worry about which C to use, I should be deciding which
   C++ to use - Turbo or Zortech.

Lew Berman
bermanl@cs.yale.edu