I recently posted a problem using big numbers that ran quite a bit slower than
the current Racket. Peter Bex supplied 2 patches that were a great improvement.
While trying to increase the speed I used the csc option -heap-size 1000M
where I varied the size from 1000M to 1M. I have 32 GB
With patch 0001 the elapsed time went from 33.7 seconds to 24.5 seconds.
With patch 0002 the elapsed time went to 23.4 seconds.
Good work -- Doug
On Wednesday, May 22, 2024 at 08:54:49 AM MDT, Peter Bex
wrote:
On Wed, May 22, 2024 at 02:42:38PM +0200, Peter Bex wrote:
> Attached are
Hello Mario,
Yes, please add the program to the chicken-benchmarks.
Regards,Doug
On Wednesday, May 22, 2024 at 12:50:56 PM MDT, Mario Domenech Goulart
wrote:
Hi Doug,
On Tue, 21 May 2024 21:35:33 + (UTC) "T.D. Telford"
wrote:
> Thanks for the reply. The elapsed timings for the
Hi Doug,
On Tue, 21 May 2024 21:35:33 + (UTC) "T.D. Telford"
wrote:
> Thanks for the reply. The elapsed timings for the program rho3rec are:
>
> chicken 5.3.0: 33.6 seconds
> Racket v8.2 [cs] : 18.1 seconds
> Dr Racket : 20.6 seconds (1 MB memory)
>
> The program uses the Pollard
> On Wed, May 22, 2024 at 02:42:38PM +0200, Peter Bex wrote:
> > Attached are two patches, one which has this bigger improvement, and
> > another which is a minor improvement which translates to shaving about
> > a second of runtime off your program (at least on my machine).
>
> The minor patch
On Wed, May 22, 2024 at 02:42:38PM +0200, Peter Bex wrote:
> Attached are two patches, one which has this bigger improvement, and
> another which is a minor improvement which translates to shaving about
> a second of runtime off your program (at least on my machine).
The minor patch was
Hello Doug and CHICKEN hackers,
Thanks for the benchmarking program. Somehow your e-mail is a bit
garbled, hence the top-posting, for which I apologize.
The benchmark in question deals mostly with small-sized bignums, which
means we're not even trigger the fancier division algorithms. This is
