My "official" request to top engine programmers

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

Moderator: Ras

Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: My "official" request to top engine programmer

Post by Lyudmil Tsvetkov »

duncan wrote:
Lyudmil Tsvetkov wrote:

you can evaluate without searching, but you can not search without evaluating, even a single node.

so, evaluation is more important.
a 30 move searcher does not need any evaluation.
why so?
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: My "official" request to top engine programmer

Post by Lyudmil Tsvetkov »

carldaman wrote: Thanks a lot, Daniel -- if there's a will... :)
there is a bill? :)
Rodolfo Leoni
Posts: 545
Joined: Tue Jun 06, 2017 4:49 pm
Location: Italy

Re: My "official" request to top engine programmer

Post by Rodolfo Leoni »

duncan wrote:
peter wrote:Are we both talking about this one from Jeremy Bernstein?

http://www.open-chess.org/viewtopic.php?f=7&t=2663

The download-link given in that posting even works still
I tried this. hash is only 4 meg. do you know how does one change that?
4 MBs is persistent hash file start size. It grows while using the engine. As well as Critter, moves, scores and depths are saved after a move is made, either your move in infinite analysis mode or engine move in game.

Persistent hashes are additional hashes, so if you set 1 GB hash for the engine it'll be 1 GB + those 4 MB, at start. Persistent hash entries are the pv nodes which respect the depth setting. If depth is set to 20 (default) and engine goes to depth 30, 11 positions are saved into PHs (from depth 20 to 30). This is why Jeremy says, "The lower the depth, the larger the file".
F.S.I. Chess Teacher
peter
Posts: 3410
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: My "official" request to top engine programmer

Post by peter »

Hi Duncan!
duncan wrote:
peter wrote:Are we both talking about this one from Jeremy Bernstein?

http://www.open-chess.org/viewtopic.php?f=7&t=2663

The download-link given in that posting even works still
I tried this. hash is only 4 meg. do you know how does one change that?
If you really mean hash, no problem here with the SSE4.2- compile and 16G, didn't have to set anything in engine's parameters.

If you mean the "Persistent Hash Size", it can be raised to 1024M max.
Peter.
Rodolfo Leoni
Posts: 545
Joined: Tue Jun 06, 2017 4:49 pm
Location: Italy

Re: My "official" request to top engine programmer

Post by Rodolfo Leoni »

cdani wrote:......
I have tested it a few times. If anything strange happens let me know.
......
Well, it's not related with PHs, but it seems there's no multipv output. At least, here with Arena 3.5.1. It happens with 0.91 version too. Can somebody confirm it?

As for PHs, they seem to work fine. I'll try some automatic analysis of a pgn file, in backward analysis mode.
F.S.I. Chess Teacher
peter
Posts: 3410
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: My "official" request to top engine programmer

Post by peter »

Rodolfo Leoni wrote:
cdani wrote:......
I have tested it a few times. If anything strange happens let me know.
......
Well, it's not related with PHs, but it seems there's no multipv output. At least, here with Arena 3.5.1. It happens with 0.91 version too. Can somebody confirm it?
Yes, Rodolfo, indeed in Arena (3.0 here) MV- mode doesn't work as expected. There was more than a single output- line for a short time right after starting analysis now and then, but any more line then a single one disappeared always soon again.

Maybe it's a matter of showing the multiple pvs by the GUI, in Deep Fritz 14
the problem didn't arise till now
Peter.
Guenther
Posts: 4718
Joined: Wed Oct 01, 2008 6:33 am
Location: Regensburg, Germany
Full name: Guenther Simon

Re: My "official" request to top engine programmer

Post by Guenther »

cdani wrote:
Rodolfo Leoni wrote:If it's a little effort as to make hashes persistent in infinite analysis or to introduce a load and save command button than we can expect he'll do it, but when things become more complicated it'd be unfair to ask Daniel or anybody else to let his developement work apart. At least, for now. :)
:-)

I added to Andscacs the capability of saving the full hash to file, to allow the user to recover a previous analysis session and continue it.
The saved hash file will be of the same size of the hash memory, so if you defined 4 GB of hash, such will be the file size. Saving and loading such big files can take some time.

To be able to do it I have added 4 new uci parameters:

option name NeverClearHash type check default false
option name HashFile type string default hash.hsh
option name SaveHashtoFile type button
option name LoadHashfromFile type button

You can set the NeverClearHash option to avoid that the hash could be cleared by a Clear Hash or ucinewgame command.

The HashFile parameter is the full file name with path information. If you don't set the path, it will be saved in the current folder. It defaults to hash.hsh.

To save the hash, stop the analysis and press the SaveHashtoFile button in the uci options screen of the GUI.

To load the hash file, load the game you are interested in, load the engine withouth starting it, and press the LoadHashfromFile button in the uci options screen of the GUI. Now you can start the analysis.

(windows 64 popcnt)
www.andscacs.com/downloads/andscacs091148.zip

I have tested it a few times. If anything strange happens let me know.
I hope is useful to analyzers!
Would be nice also for no popcount machines :)

Guenther
https://rwbc-chess.de

[Trolls n'existent pas...]
carldaman
Posts: 2287
Joined: Sat Jun 02, 2012 2:13 am

Re: My "official" request to top engine programmer

Post by carldaman »

Lyudmil Tsvetkov wrote:
carldaman wrote: Thanks a lot, Daniel -- if there's a will... :)
there is a bill? :)
:lol:

Actually, with DJ, it is his treat entirely, so he picks up the tab.
Nice guy, really. It is the fellows who bill us for their programs
that should have taken the lead on this one. ;)
duncan
Posts: 12038
Joined: Mon Jul 07, 2008 10:50 pm

Re: My "official" request to top engine programmer

Post by duncan »

Lyudmil Tsvetkov wrote:
duncan wrote:
Lyudmil Tsvetkov wrote:

you can evaluate without searching, but you can not search without evaluating, even a single node.

so, evaluation is more important.
a 30 move searcher does not need any evaluation.
why so?
speaking as a patzer in chess, I thought 99.99% of all positional weaknesses would be discovered before 30 moves, so no evaluation needed.
User avatar
Marek Soszynski
Posts: 586
Joined: Wed May 10, 2006 7:28 pm
Location: Birmingham, England

Re: My "official" request to top engine programmer

Post by Marek Soszynski »

+1
Marek Soszynski