Crafty 25.0 Release

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
User avatar
JVMerlino
Posts: 1003
Joined: Wed Mar 08, 2006 9:15 pm
Location: San Francisco, California

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by JVMerlino » Mon Dec 28, 2015 8:45 pm

bob wrote:
JVMerlino wrote:
bob wrote:
JVMerlino wrote:
bob wrote:Is this windows or linux? All of my output above is from Linux machines... (or OS X, still unix).
This is 64-bit Windows 7, using Daniel's compiles (I tried both popcnt and non-popcnt). I also made sure that Crafty 24.1 did not exhibit the same behavior, just in case my eyes were playing tricks on me.

jm
Is it happening consistently? If so that points to the windows version as I ran SMP/non-SMP and EGTB/non-EGTB multiple times just now and while there is a bit of variability in the SMP stuff, no draw scores are showing up for me, ever...
It is happening consistently, regardless of whether I use the default of 1 thread or use "cores 4" (which is also a little weird, because it says that the option is disabled but it DOES set the maximum threads). Of course, the output is different because of time-to-depth, but the fundamental results are the same.

jm
When you see "something" is disabled from crafty, it means that either you had something like something=xx on the command line, or something=xx in the crafty.rc/.craftyrc file. If you set something explicitly via crafty.rc or command line, it assumes you mean it and it won't let xboard/winboard override something you explicitly specified...

But it is really looking like the windows version is simply broken. I don't know if microsoft has added an update to their MSVC compiler or what. Would be interesting for someone to compile 24.1 on the same compiler that the broken 25.0 executable comes from, that would reveal whether it is something compiler-related or something 25.0 related...

Those odd draw scores point toward something broken with hashing. What I have no idea, since that code is the same for windows or unix, except for allocation. I'll take a quick look at the windows allocation code to see if there is something different there...
I'm getting that "option disabled" message even if I run Crafty with no rc file and with no commandline options. For example, if I type "cores 4", the message says:

Warning-- xboard 'cores' options disabled
max threads set to 4.

What's even weirder now is that I've tried all previous compiles (including Dann's new build) on my Win 10 laptop. Daniel's and Jose's builds perform identically to my Win 7 desktop.

HOWEVER, on my laptop, Dann's build works perfectly on Fine 70. But on my desktop, the build crashes immediately on launch. :?

I'm now thoroughly confused....

EDIT: Some further info on Dann's build crashing on my Win7 desktop. It crashes slightly differently depending on whether or not I use an rc file. If I use the rc file, I do not see any output before the crash. If I do not use an rc file, I only see:

EPD Kit revision date: 1996.04.21

...and then it crashes.... :cry:

jm

User avatar
Werner
Posts: 2417
Joined: Wed Mar 08, 2006 9:09 pm

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by Werner » Mon Dec 28, 2015 9:09 pm

JVMerlino wrote:But your output points to another possibility. Could you try it with a crafty.rc file, but without TBs? I do not have any TBs that Crafty supports.
jm
The result is the same, only the output is funny: always the same time for a lot of depths:
FEN: 8/k7/3p4/p2P1p2/P2P1P2/8/8/K7 w - - 0 1

Crafty-25.0:

66 01:38 485.550.427 4.915.970 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
66 01:38 485.550.601 4.915.972 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
67 01:38 485.551.951 4.915.986 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
67 01:38 485.552.106 4.915.987 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
68 02:34 760.672.645 4.930.787 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
68 02:34 760.672.697 4.930.788 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
69 02:34 760.674.944 4.930.802 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
69 02:34 760.675.041 4.930.803 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
70 02:34 760.677.115 4.930.816 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
70 02:34 760.677.238 4.930.817 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
Werner

User avatar
JVMerlino
Posts: 1003
Joined: Wed Mar 08, 2006 9:15 pm
Location: San Francisco, California

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by JVMerlino » Mon Dec 28, 2015 9:10 pm

Werner wrote:
JVMerlino wrote:But your output points to another possibility. Could you try it with a crafty.rc file, but without TBs? I do not have any TBs that Crafty supports.
jm
The result is the same, only the output is funny: always the same time for a lot of depths:
FEN: 8/k7/3p4/p2P1p2/P2P1P2/8/8/K7 w - - 0 1

Crafty-25.0:

66 01:38 485.550.427 4.915.970 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
66 01:38 485.550.601 4.915.972 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
67 01:38 485.551.951 4.915.986 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
67 01:38 485.552.106 4.915.987 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
68 02:34 760.672.645 4.930.787 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
68 02:34 760.672.697 4.930.788 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
69 02:34 760.674.944 4.930.802 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
69 02:34 760.675.041 4.930.803 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
70 02:34 760.677.115 4.930.816 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
70 02:34 760.677.238 4.930.817 0,00 0. ... () 1. Kb1 Kb7 2. Kc2 Kc8 3. Kb3 Kc7 4. Kc4 Kb6 5. Kc3 Kb7 6. Kd3 Kc7 7. Ke2 Kd7 8. Kf3 Ke7 9. Kf2 Ke8 10. Ke1 Kd8 11. Kd2 Kc8 12. Kc2 Kb7 13. Kc3 Kc7 14. Kc4 <3-fold>
Well, all I can say at this point is "thank God it's not just me". :)

jm

Peter Berger
Posts: 390
Joined: Thu Mar 09, 2006 1:56 pm

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by Peter Berger » Mon Dec 28, 2015 9:20 pm

JVMerlino wrote:
Werner wrote:
JVMerlino wrote:But your output points to another possibility. Could you try it with a crafty.rc file, but without TBs? I do not have any TBs that Crafty supports.
jm
The result is the same, only the output is funny: always the same time for a lot of depths:
Well, all I can say at this point is "thank God it's not just me". :)

jm
I get the same behaviour ( Fine 70 not solved either with/without TBs and with/without EGTB and with/without crafty.rc) - Windows 10, mt=2, Crafty25_64.exe

I also noticed that with hashp=128M it now sets hashp to 96M - similar with hash=384M it sets hash to 256M - might be new intended behaviour, just as no one mentioned it so far.

bob
Posts: 20557
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by bob » Mon Dec 28, 2015 9:42 pm

Peter Berger wrote:
JVMerlino wrote:
Werner wrote:
JVMerlino wrote:But your output points to another possibility. Could you try it with a crafty.rc file, but without TBs? I do not have any TBs that Crafty supports.
jm
The result is the same, only the output is funny: always the same time for a lot of depths:
Well, all I can say at this point is "thank God it's not just me". :)

jm
I get the same behaviour ( Fine 70 not solved either with/without TBs and with/without EGTB and with/without crafty.rc) - Windows 10, mt=2, Crafty25_64.exe

I also noticed that with hashp=128M it now sets hashp to 96M - similar with hash=384M it sets hash to 256M - might be new intended behaviour, just as no one mentioned it so far.
That's correct. A pawn hash entry is now only 24 bytes long, so it has to be 3/4 of a power of 2. A regular hash entry is 16 bytes, which makes the normal trans/ref size a perfect power of 2.

But the fact that this produces bogus stuff for everyone under windows, while it works perfectly under linux (fine #70) means there's some debugging to do to figure out what is going wrong. I've tried multiple compilers, gcc 4.7 and 4.8 as well as various Intel compilers, all with the same (correct/normal) results...

bob
Posts: 20557
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by bob » Mon Dec 28, 2015 9:55 pm

JVMerlino wrote:
bob wrote:
JVMerlino wrote:
bob wrote:
JVMerlino wrote:
bob wrote:Is this windows or linux? All of my output above is from Linux machines... (or OS X, still unix).
This is 64-bit Windows 7, using Daniel's compiles (I tried both popcnt and non-popcnt). I also made sure that Crafty 24.1 did not exhibit the same behavior, just in case my eyes were playing tricks on me.

jm
Is it happening consistently? If so that points to the windows version as I ran SMP/non-SMP and EGTB/non-EGTB multiple times just now and while there is a bit of variability in the SMP stuff, no draw scores are showing up for me, ever...
It is happening consistently, regardless of whether I use the default of 1 thread or use "cores 4" (which is also a little weird, because it says that the option is disabled but it DOES set the maximum threads). Of course, the output is different because of time-to-depth, but the fundamental results are the same.

jm
When you see "something" is disabled from crafty, it means that either you had something like something=xx on the command line, or something=xx in the crafty.rc/.craftyrc file. If you set something explicitly via crafty.rc or command line, it assumes you mean it and it won't let xboard/winboard override something you explicitly specified...

But it is really looking like the windows version is simply broken. I don't know if microsoft has added an update to their MSVC compiler or what. Would be interesting for someone to compile 24.1 on the same compiler that the broken 25.0 executable comes from, that would reveal whether it is something compiler-related or something 25.0 related...

Those odd draw scores point toward something broken with hashing. What I have no idea, since that code is the same for windows or unix, except for allocation. I'll take a quick look at the windows allocation code to see if there is something different there...
I'm getting that "option disabled" message even if I run Crafty with no rc file and with no commandline options. For example, if I type "cores 4", the message says:

Warning-- xboard 'cores' options disabled
max threads set to 4.

What's even weirder now is that I've tried all previous compiles (including Dann's new build) on my Win 10 laptop. Daniel's and Jose's builds perform identically to my Win 7 desktop.

HOWEVER, on my laptop, Dann's build works perfectly on Fine 70. But on my desktop, the build crashes immediately on launch. :?

I'm now thoroughly confused....

EDIT: Some further info on Dann's build crashing on my Win7 desktop. It crashes slightly differently depending on whether or not I use an rc file. If I use the rc file, I do not see any output before the crash. If I do not use an rc file, I only see:

EPD Kit revision date: 1996.04.21

...and then it crashes.... :cry:

jm
I just noticed, those "warnings" should be suppressed if not in xboard mode. I had not paid much attention to them, simply adding them so that if a log file has the warnings, it would be helpful debugging info if someone reports a problem. I have modified the source to suppress these if not in xboard mode.

But that is not what is going on here. There is simply something broken in either windows, the windows compiler, or Crafty source when compiled only for windows...

I probably just should say "windows is deprecated" and move on. :)

User avatar
Jim Ablett
Posts: 1327
Joined: Fri Jul 14, 2006 5:56 am
Location: London, England
Contact:

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by Jim Ablett » Mon Dec 28, 2015 11:57 pm

Just updated my Crafty builds to latest version 25.0. You can get them here >

http://jimablett.net63.net/

Jim..

User avatar
JVMerlino
Posts: 1003
Joined: Wed Mar 08, 2006 9:15 pm
Location: San Francisco, California

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by JVMerlino » Tue Dec 29, 2015 12:20 am

These builds work perfectly for me, both with and without an rc file.

Thanks, Jim! :D

jm

User avatar
Jim Ablett
Posts: 1327
Joined: Fri Jul 14, 2006 5:56 am
Location: London, England
Contact:

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by Jim Ablett » Tue Dec 29, 2015 12:26 am

Hi John,

Glad there are functioning ok for you. Builds are compiled with all optional features enabled.

Jim.

bob
Posts: 20557
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Crafty 25.0 Release (Fine 70 problem?)

Post by bob » Tue Dec 29, 2015 12:30 am

Jim Ablett wrote:Hi John,

Glad there are functioning ok for you. Builds are compiled with all optional features enabled.

Jim.
Would be nice to know what was broken with the other compiles, but at least I am now going to stop looking at the windows code here...

Post Reply