This is really interesting, thanks for doing this analysis.

My interpretation is Open Watcom is about on par with Turbo C/C++. For
this test, OW is 8 seconds (6%) faster on 8088, 3 seconds (11%) slower
on 286, and 1 second (10%) slower on 386. Also 1 second slower on 486,
but the "twice as slow" (50%) difference looks bigger because the
times were so fast: 2 seconds vs 4 seconds.

It's really interesting that IA-16 GCC is so fast across all CPUs. 38
seconds (32%) faster on 8088, 12 seconds (46%) faster on 286, 5
seconds (52%) on 386, and 1 second on 486.

I know this is emulated in DOSBox (not real CPU) but it's still a good
comparison.


It's also interesting to see the EXE size differences:

IA16-GCC: 19728 Bytes (19 kiB)
Watcom: 20544 Bytes (20 kiB)
Turbo C++: 17510 Bytes (17 kiB)

Turbo C/C++ was very good at keeping the EXE size small, but the
counter is that Open Watcom meets a later C standard so probably has
more in the runtime. But IA-16 GCC is also more modern, yet its EXE
size is in the middle (smaller than OW). This makes me think I should
build more of my FreeDOS programs using IA-16 GCC than I have been.


> > Am 21.07.2026 um 19:47 schrieb Jim Hall :
> >
> > There's also the RUNTIME tool, included in FreeDOS in the
> > "Development" package group.

On Wed, Jul 22, 2026 at 1:57 PM Bernd Boeckmann wrote:
>
> Thanks, I did some more benchmarks (this time in 86box) and used RUNTIME:
>
> 8088 @ 4.7 MHz:
>   IA16-GCC: 01:19.04m
>   Watcom: 01:53.59m
>   Turbo C++: 02:01.66m
>
> 286 @ 6 MHz:
>   IA16-GCC: 00:14.00m
>   Watcom: 00:27.51m
>   Turbo C++: 00:24.49m
>
> 386 @ 16 MHz:
>   IA16-GCC: 00:05.00m
>   Watcom: 00:11.15m
>   Turbo C++: 00:10.00m
>
> 486DX @ 25 MHz:
>   IA16-GCC: 00:01.43m
>   Watcom: 00:03.95m
>   Turbo C++: 00:02.36m
>
> Watcom performs significantly worse than even Turbo C++ on 386 and
> 486. That is not something I expected.
>
> Also: 286 and 386 are both ~4 times as fast per clock cycle than the
> 8088, and the 486 is even ~10 times as fast per clock cycle than the
> 8088, and >2 times than the 386.
>


_______________________________________________
Freedos-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/freedos-devel

Reply via email to