Discussion of anything and everything relating to chess playing software and machines.
Moderators: hgm, Harvey Williamson, bob
Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
-
asanjuan
- Posts: 206
- Joined: Thu Sep 01, 2011 3:38 pm
- Location: Seville, Spain
Post
by asanjuan » Thu Nov 26, 2015 9:57 am
It's been a hard year for me and I could only do minor changes and corrections. But these changes added 40 elo points in self test.
You can download the new version from
www.chessrhetoric.com (page in spanish)
Dedicated for those players who want an "almost human" engine and play interesting games.

-
jpqy
- Posts: 503
- Joined: Thu Apr 24, 2008 7:31 am
- Location: Belgium
Post
by jpqy » Thu Nov 26, 2015 10:02 am
Thanks Alberto
A little question..is it possible to compile a bmi2 compile?!
JP.
-
Frank Quisinsky
- Posts: 4819
- Joined: Wed Nov 18, 2009 6:16 pm
- Location: Trier, Germany
-
Contact:
Post
by Frank Quisinsky » Thu Nov 26, 2015 10:14 am
Hi Alberto,
good News!
At the moment a lot to do in testing engines but you can be sure ... I will test it (wait of it).
Best
Frank
I like computer chess!
-
asanjuan
- Posts: 206
- Joined: Thu Sep 01, 2011 3:38 pm
- Location: Seville, Spain
Post
by asanjuan » Thu Nov 26, 2015 11:10 am
jpqy wrote:Thanks Alberto
A little question..is it possible to compile a bmi2 compile?!
JP.
Not yet. Actually this is the first version that adds a popcount compile. My code is still not ready to add this extension.
Sorry.
Looks like it affects to mobility calculation. Do you know how much does it improve the performance? Popcount was a relatively poor speedup (5% or so).
At the moment I'm interested in improving the algorithm, so hardware speedups of 4-5% are not in the pipeline. Anyway, I will consider it for future versions.
-
SzG
- Posts: 2447
- Joined: Fri Mar 10, 2006 6:20 am
- Location: Szentendre, Hungary
Post
by SzG » Thu Nov 26, 2015 5:55 pm
Hi Alberto,
Thanks for the updated engine. Are you sure it is only 40 Elos better than 1.4.1? The size suggests you've made a lot of changes.
Gabor Szots
CCRL testing group
-
cdani
- Posts: 2047
- Joined: Sat Jan 18, 2014 9:24 am
- Location: Andorra
-
Contact:
Post
by cdani » Thu Nov 26, 2015 6:15 pm
¡Gracias! Thanks!
-
MikeB
- Posts: 2522
- Joined: Thu Mar 09, 2006 5:34 am
- Location: Pen Argyl, Pennsylvania
Post
by MikeB » Thu Nov 26, 2015 8:03 pm
Do you have a Mac compile?
-
asanjuan
- Posts: 206
- Joined: Thu Sep 01, 2011 3:38 pm
- Location: Seville, Spain
Post
by asanjuan » Thu Nov 26, 2015 8:25 pm
SzG wrote:Hi Alberto,
Thanks for the updated engine. Are you sure it is only 40 Elos better than 1.4.1? The size suggests you've made a lot of changes.
Yes, I am quite sure. The reason of the size is because
I changed the compiler. This is the first versión built with Mingw. Version 1.4.1 was built using an old VS2005
-
carldaman
- Posts: 1543
- Joined: Sat Jun 02, 2012 12:13 am
Post
by carldaman » Thu Nov 26, 2015 8:27 pm
asanjuan wrote:It's been a hard year for me and I could only do minor changes and corrections. But these changes added 40 elo points in self test.
You can download the new version from
www.chessrhetoric.com (page in spanish)
Dedicated for those players who want an "almost human" engine and play interesting games.

Good news, this is much appreciated, Alberto!
Thanks,
CL
-
asanjuan
- Posts: 206
- Joined: Thu Sep 01, 2011 3:38 pm
- Location: Seville, Spain
Post
by asanjuan » Thu Nov 26, 2015 9:21 pm
MikeB wrote:Do you have a Mac compile?
Last week I was looking for a cross compiler targetting Mac, but I couldn't find any.
It would be nice to find someone who can build the binary for me.
I can send him the sources by mail.
