CCRL Discussion Board

Graham:

as requested, i removed search of SYZYGY tables in default path of "H:/SyzygyTablebases". Updated executables are on my site.

I also made a small bug fix to parallel search only (single CPU search is unchanged).

Youri.

I added option "SyzygyPath". It does the same thing as "NalimovPath". The engine only supports syzygy EGTBs, but their path can be passed to engine under either of those two options. If there are no syzygy files in that path, then syzygy EGTBs will not be used. Updated executable...

The way to pass tablebase path is: “setoption name NalimovPath value c:\chess\tb\4;c:\chess\tb\5” - path separated by semicolon. This is straight from the UCI specs. In addition, at start-up engine looks for SYZYGY tables in H:/SyzygyTablebases (hard-coded path specifically for CCRL), and in environ...

Version 1.5 of Fizbo chess engine has been released, and is available for download from https://sites.google.com/site/fizbochessengine/. Please use this version for all future CCRL tournaments. Changes from prior version: 1. I switched it to UCI protocol: this should make it more compatible with mod...

I see.
Then i guess the GUI will pass the path to my engine, and will have the logic to use that path in next version of my engine.
Thanks.

I recently added logic to handle 5-piece syzygy bases to my engine (Fizbo), and wonder if i can use that logic in CCRL games. I know that most if not all CCRL games terminate when there are only 5 pieces on the board, but looking-up EGBB scores from within the search, when search starts with >5 piec...

OK, i finally found the bug and fixed it. Now version 1.31 posted at https://sites.google.com/site/fizbochessengine/ should not crash. Aside from bug fix, it is unchanged from what it was, except that now it reports version number 1.31. And it is still different from version 1.3, but playing strengt...

I looked into this, and here is what i found: 1. i reproduce the problem; starting with FEN of "r1b1k2r/p2n1p2/3Pp2p/q2p4/1p1Q4/6BR/PPp1BPP1/1K1R4 w kq - 0 0" crashed the engine at search depth 15. 2. unfortunately, i did not save the source code that vent into the release. Since then i ma...

Please use the latest version, the one that displays version number 1.3. The one posted on the website right now.

It has different evaluation, and should play slightly better.

Youri.

Sorry, i forgot about version number that the engine reports.

I fixed it now - if you download the engine, it will now say that it is version 1.3.

Youri.

Fizbo version 1.3 has just been released. It can be downloaded from https://sites.google.com/site/fizbochessengine/. This release increases playing strength by around 80 ELO. Please use it for the 50th amateur series (division 2). Here are release notes: Changes in version 1.3, released November 11,...

Great, thanks!

OK, i found the problem - 2 occurrences of "popcount" sneaked into the "no_pc" version. I removed them, and verified by reviewing disassembly that there are no more of them in the "no_pc"executable. Sorry about that - i did not carry the code to automatically replace al...

let me look into this and see what i can find.

Thanks, Youri.

I don't know what is going on. The only difference between "popcount" and "non popcount" versions is that instead of "__popcnt64()" manually coded version of that function is used in "_no_pc" executable. I just tried running both files on my laptop and on my w...

Sorry about that - i don't know why they don't run. I recompiled and uploaded new binaries - could you please try them again? If i had some switches accidentally set to nonportable values, that should be fixed now. I think now i did exactly the same thing i did for previous version. Also, parallel s...

it can use 8 cores. But i only tried it on 4 cores, since i only have 4 core computer. And i would not expect 8 cores to be much better than 4, since parallel search algorithm is simple and should not scale well.

Youri.

i just released new version (1.2) of Fizbo engine. It can be downloaded from https://sites.google.com/site/fizbochessengine/. In this version: 1. improved search increases strength by around +70 ELO 2. now Fizbo supports parallel search. Self play using 4 cores shows strength increase of around +90 ...

I just added a 64-bit executable that does not use "popcount" instruction. This should work on any 64-bit CPU.
It is posted at "https://sites.google.com/site/fizbochessengine/".

Youri.

I live in New York, USA.

This is it! Thanks for pointing it out to me. I realized that the executable was created with "/AVX" compiler switch. I recompiled it without it, and i think now it should work on any CPU supporting "popcount" instruction - any Intel "i" CPU. The new executable is poste...

I don't know what is causing the problem. I just tried to run it on my laptop (i5+Win7) - it works fine. And for this compile i make some changes to ensure portability - turned off endgame bit-bases, turned off all logging, included runtime C++ libraries into the executable. I'm all out of ideas. Yo...

It actually is 64 bit executable, i just incorrectly listed it is as x86. I even called the executable fizbo11x64, to remember it. I apologize for the confusion.

Youri.

I have created a chess engine - Fizbo - could you please test it and add it to CCRL? Executable can be downloaded from: https://sites.google.com/site/fizbochessengine/ Features: uses Winboard protocol runs on 1 CPU core requires x86 Windows requires "popcount" and "bitscanforward"...