Stockfish 2.2.1 running for the IPON

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

User avatar
natasha
Posts: 145
Joined: Tue Jan 25, 2011 3:10 pm

Re: Critter 1.4 time losses ...

Post by natasha »

rvida wrote:
rvida wrote:
mcostalba wrote:
rvida wrote: Without the delay the engine does nothing because the search is stopped before it has started:
But why when you receive a ponder hit you don't simply switch to normal search ? SF stops on ponder hit only if the flag stopOnPonderhit is raised.

Just curious.
I do...

But there is a case when 'go ponder' command signals the "thinking" thread that we've got I job, but before the OS scheduler wakes the thread we already cancel the request because of a subsequent ponderhit :(
Btw it is hard to explain (at least for me with my poor english) without giving out my source code...

Now I've got an idea... If you are interested enough, we can exchange our source codes ;). (I already have the sources of SF 2.0 but I am ready for an upgrade)
I can send you my sources in the hope that we can share in the spirit of Fabien's "readme.txt" :)
Anyway, I have decided to launch a new initiative. What's more boring
than reading one's own code at 3am tracking down a bug that might not
even exist, know what I mean? I have the solution: let's fix
each others bugs!

The new operation is called "Fix my Bugs and I Fix Yours!" (patent
pending). It works as follows:

1) You fix one of my bugs (excluding null move) before 2005/09/01
00:00 UTC (the acronym that does not mean anything in either
English or French, so that both parties are equally disappointed).

2) I select the most artistic bug fix after the date limit. A jury
will be nominated if necessary.

3) I fix a bug of your choice in your program (excluding "it plays bad
moves"), it's that simple!

+1
User avatar
rvida
Posts: 481
Joined: Thu Apr 16, 2009 12:00 pm
Location: Slovakia, EU

Re: Critter 1.4 time losses ...

Post by rvida »

IWB wrote:
OK, does this mean there will be a Critter 1.4.1 soon? :-)
I am not sure. You helped me with a bug-report, and you gave me a very good clue what might be wrong with my engine. I am very thankful for that. The fix itself is rather trivial, but I cannot be sure if this is the primary cause of the time losses.

There is a lot of "logistic" issues with a new release. I need to do a (PGO) recompile for Win32/64, Linux32/64, Android. Then sending the binaries to Emil V. who is the owner/maintainer of the website. Not to talk about the Mac compile - which is "outsourced" to a volunteer (Jeremy B.).

If I may ask you for a favor... I would send you (someday next week) a Critter 1.4a with the proposed fix. If you happen to have some free CPU time, can you please play a few hundred games under IPON conditions? If the fix would work and prevents losses on time, I will make a new 'official' release.
lucasart
Posts: 3241
Joined: Mon May 31, 2010 1:29 pm
Full name: lucasart

Re: Stockfish 2.2.1 running for the IPON

Post by lucasart »

IWB wrote:Here: http://www.inwoba.de you can follow.

Have fun
Ingo
Does StockFish still lose on time in this version ? At fast time control I experienced quite a few time losses with 2.1.1, especially when removing the resigning option.
IWB
Posts: 1539
Joined: Thu Mar 09, 2006 2:02 pm

Re: Critter 1.4 time losses ...

Post by IWB »

rvida wrote:
IWB wrote:
OK, does this mean there will be a Critter 1.4.1 soon? :-)
I am not sure. You helped me with a bug-report, and you gave me a very good clue what might be wrong with my engine. I am very thankful for that. The fix itself is rather trivial, but I cannot be sure if this is the primary cause of the time losses.
I realized that it is even worse. If you play ponder games on one computer, the imediate answer might be that fast that you code works. If you play ponder ON games on a web server the imediate answer is slow enough that it works again. It is my special setup (a local network drive), which happen to show the problem ...

rvida wrote: There is a lot of "logistic" issues with a new release. I need to do a (PGO) recompile for Win32/64, Linux32/64, Android. Then sending the binaries to Emil V. who is the owner/maintainer of the website. Not to talk about the Mac compile - which is "outsourced" to a volunteer (Jeremy B.).
Sorry, I know. It is never easy ... :-)
rvida wrote: If I may ask you for a favor... I would send you (someday next week) a Critter 1.4a with the proposed fix. If you happen to have some free CPU time, can you please play a few hundred games under IPON conditions? If the fix would work and prevents losses on time, I will make a new 'official' release.
Go for it, I'll send a PM ... but: The problem is, that Critter played 2400 games with a lower "Loss On Time" LOT rate than the version before. It happens only with the new Stockfish (whatever it is that makes the new version that fast). But yes, if you send a fix I will see how it performes vs Stockfish 2.2.1!

Until then I leave the games in, as it is not that significant seen the overall statistic, and will replace the games with fixed Critter asap.

Bye
Ingo
Last edited by IWB on Sun Jan 08, 2012 9:38 am, edited 1 time in total.
mcostalba
Posts: 2684
Joined: Sat Jun 14, 2008 9:17 pm

Re: Critter 1.4 time losses ...

Post by mcostalba »

rvida wrote: Now I've got an idea... If you are interested enough, we can exchange our source codes ;). (I already have the sources of SF 2.0 but I am ready for an upgrade)
I can send you my sources in the hope that we can share in the spirit of Fabien's "readme.txt" :)
Anyway, I have decided to launch a new initiative. What's more boring
than reading one's own code at 3am tracking down a bug that might not
even exist, know what I mean? I have the solution: let's fix
each others bugs!

The new operation is called "Fix my Bugs and I Fix Yours!" (patent
pending). It works as follows:

1) You fix one of my bugs (excluding null move) before 2005/09/01
00:00 UTC (the acronym that does not mean anything in either
English or French, so that both parties are equally disappointed).

2) I select the most artistic bug fix after the date limit. A jury
will be nominated if necessary.

3) I fix a bug of your choice in your program (excluding "it plays bad
moves"), it's that simple!
Of course I accept :-) :-) :-)

I have very clear in my mind the bugs that bugged me most: I will check how you have done with them. I will send you a pm with my email address.

Thanks Richard !!!!
IWB
Posts: 1539
Joined: Thu Mar 09, 2006 2:02 pm

Re: Stockfish 2.2.1 running for the IPON

Post by IWB »

Stockfish 2.2.1 run is finished.

You can find the result here: http://www.inwoba.de

If you dont find it, it is your browsercache, hit F5 a few times and it will show up.

Bye
Ingo
mcostalba
Posts: 2684
Joined: Sat Jun 14, 2008 9:17 pm

Re: Stockfish 2.2.1 running for the IPON

Post by mcostalba »

IWB wrote:Stockfish 2.2.1 run is finished.

You can find the result here: http://www.inwoba.de

If you dont find it, it is your browsercache, hit F5 a few times and it will show up.

Bye
Ingo
Thanks Ingo ! But I don't find the 7zip file with the individual results. where I have to browse ?
IWB
Posts: 1539
Joined: Thu Mar 09, 2006 2:02 pm

Re: Stockfish 2.2.1 running for the IPON

Post by IWB »

mcostalba wrote:
Thanks Ingo ! But I don't find the 7zip file with the individual results. where I have to browse ?
Ups, I forgot to upload the most recent one, that is fixed and you find it in "Archive", scroll down a bit, then "IPON - Individual statistics".

Thanks for the new version!
Ingo
mcostalba
Posts: 2684
Joined: Sat Jun 14, 2008 9:17 pm

Re: Stockfish 2.2.1 running for the IPON

Post by mcostalba »

IWB wrote: Thanks for the new version!
Thanks to you for testing, well very good results!

Code: Select all

7 Stockfish 2.2.1 JA SSE42  2954 2200.0 (1674.0 : 526.0)
                                    100.0 ( 42.0 :  58.0) Houdini 2.0 STD           3016
                                    100.0 ( 54.0 :  46.0) Komodo 4 SSE42            2973
                                    100.0 ( 56.5 :  43.5) Deep Rybka 4.1 SSE42      2955
On direct matches SF won against all the engines but Houdini ! This was really unexpected :-)
IWB
Posts: 1539
Joined: Thu Mar 09, 2006 2:02 pm

Re: Stockfish 2.2.1 running for the IPON

Post by IWB »

mcostalba wrote:

Code: Select all

7 Stockfish 2.2.1 JA SSE42  2954 2200.0 (1674.0 : 526.0)
                                    100.0 ( 42.0 :  58.0) Houdini 2.0 STD           3016
                                    100.0 ( 54.0 :  46.0) Komodo 4 SSE42            2973
                                    100.0 ( 56.5 :  43.5) Deep Rybka 4.1 SSE42      2955
On direct matches SF won against all the engines but Houdini ! This was really unexpected :-)
And it seems that everyone is falling into that trap :-). 100 games are statisticaly meaningless AND my 50 openings hardly represent "Chess" ... !

Bye
INgo