jeffreyan11/laser-chess-engine

Laser, a UCI-compliant chess engine.

laser-chess-engine's Issues

Segfault when issued "quit" during a search

We are having an issue right know where on Linux machines cutechess does not kill every copy of your engine, only the 50% that are not searching when cutechess is closed down.

It is possible that this is on my end, but since Ethereal works as expected here, I am looking into differences with Laser first.

I'm not sure this is exactly the issue, but something I just noticed.

Happens under go depth, go infinite, and go wtime ...,

Feature Request: support some missing uci parameters

Hi,
i split this in two issues, and close the first #5.

Please consider to support "go searchmoves", and "ponder" flag & "ponderhit"
Some comments already done in this old issue.

Jürgen

Laser for android

Hi,

Can you make it for android?
Thanks!

Laser 1.6 blundered in no time presure

Hello,
I have created an Gautlet Tournament for Laser 1.6 and in the last game Laser 1.6 made an incredible blunder. I want to mention that Laser 1.6 it wasn't in time pressure, it had 16 Minutes or something like that. I see I couldn't attache the pgn to see.
Thanks.

Laser for android

I have a question: I'm porting the engines to android and they all work fine, but with the laser engine problem. It's easy to compile, but when playing for example black, on the 3rd payday for blacks and for whites constantly makes wrong moves and loses (For example 1.e2 -e4 ​​e7-e5 2.kg1-f3 2 ... here Laser does wrong move and loses). Here are examples:

[Event "Chess for Android Tournament"]
[Date "2018.01.11"]
[Round "2"]
[White "Laser 1.6 beta"]
[Black "Chiron 4"]
[Result "0-1"]
[TimeControl "60+1"]
[PlyCount "2"]
[Device "ZTE BLADE A210"]

1. d4 {[11/18]25/d7d5 2.469} 1... d5 {[14/24]-39/c2c4 4.117} 2. --
   {[%playeraction resign]} 0-1

[Event "Chess for Android Tournament"]
[Date "2018.01.11"]
[Round "1"]
[White "Chiron 4"]
[Black "Laser 1.6 beta"]
[Result "1-0"]
[TimeControl "60+1"]
[PlyCount "3"]
[Device "ZTE BLADE A210"]

1. e4 {[16/31]32/d7d5 13.683} 1... c5 {[11/22]-24/g1f3 2.445} 2. Nf3
   {[14/27]46/b8c6 2.318} 2... -- {[%playeraction resign]} 1-0

"PrintStatistics() & Searchmoves" problems

Hi,
your program prints some statistics on "bestmove". Looking at the code its comes from this PrintStatistics() function.
Can this be removed? I also checked Laser11 (=same). Or at least have a flag for it in the makefile to disable/enable it (should be off by default).
It gives some problems to python-chess and from the uci api, it seems to not be allowed this way?!?

Next:
it seems to me that the engine doesnt support "searchmoves" as parameter of "go". Is that correct?
I need this to use "alternative move". Which doesnt work with your engine (SF f.e. is OK).
If its not inside, how are chances it will come (soon)?

Jürgen

Please add ARMv9.2, SVE2, Dot Product and Neon support

It's sad that it's not possible to compile the newest version on Android and Linux devices.

Question: UCI960 Support?

Hello,
im looking around for different uci engines for the project "picochess", which is also part of these DGT Pi chess clocks.
Mainly, im searching for engines with UCI960 support and/or Levels.

Now: Your side saying "todo uci960" ...can y tell me if that's planed in near future, and if yes when this can be expected? Also what about level support (similar to SF).

Jürgen

Laser 1.6b for Linux on SDChess website from Russia

Hello,

is http://www.sdchess.ru/engines/laser_1_6b.zip an official release or not? What´s about 1.6 win64 bit engine?

Potential Tuning data

No contact info listed on git athSM

https://www.dropbox.com/s/nwgla3evxn7ypn3/EtherealFens.zip?dl=0

Played 400,000 games with aggressive adjudication. Threw out a couple ten thousand, selected 4 random positions from each.

Restructure project

Move all source and the syzygy directroy into a /src/ folder at the top level.
Additionally, move the makefile into the /src/ directory.

Too many system calls

http://talkchess.com/forum3/viewtopic.php?f=7&t=65726

Long ago I fixed this locally by limiting the check rate for calling systemTime.

You've changed things, so I don't know where to look this time.

@noobpwnftw gets less than 50% of expected NPS while running your engine in parallel.

It is an issue with server-grade Intel CPUs, so we cannot reproduce on our thread rippers.

Laser 1.4 Syzygy tablebases

Hello,
About Laser 1.4 in Aquarium GUI, now the engine is thinking and is great, but I have completed the Syzygy path but unfortunetly at TB hits it shows 0.
Scuse me for my bad english, I hope to understand what I said.
Thanks.

laser 1.3 moves instantly no matter what thinking time is set

Hello,
I have downloaded Laser 1.3 chess engine and it always moves instantly, no matter what tkinking time is set. And if I may ask you another thing: The EvalCache UCI option that is having the maximum value 1048576 what represents this number (Kb, Mb...)?
The GUI which I am using is Aquarium 2016.
My PC specs if it matters are:
CPU: INTEL I5 6600K
RAM: 8GB DDR4 @ 2400MHz
MB: MSI Z170A PC MATE
Please help me.
Thanks a lot.

v1.5 Windows executables including source code

Hello Jeffrey and Michael,,

on the Russian SDChess website http://www.sdchess.ru/download_engines.htm v1.5 for Linux is mentioned, but not on GitHub https://github.com/jeffreyan11/uci-chess-engine/

• why? Are existing Windows binaries 32/64 bit (anywhere) of edition 1.5?

Can you build a Korean chess engine?

Can you build a Korean chess engine?

Laser Chess Engine Enhancement, Playing Chess Variants, Musketeer Chess

Hi
Nice to see inspiring and open source work.

Can you please consider a fork of your engine playing chess variants.

I'm especially interested in variants i created (and naturally other popular ones). Mine is Musketeer Chess. It is discussed in many forums and also there is Musketeer Stockfish a fork of Stockfish.

Please find details here:

https://github.com/ianfab/Musketeer-Stockfish

http://talkchess.com/forum3/viewtopic.php?f=7&t=72572

https://github.com/fsmosca/musketeer-chess

i'd like to create a chess server playing chess and chess variants. The majority of the servers that exist now are based on Stockfish or LC0. It's certainly much more interesting to have servers with other engines. Please mail me to musketeerchess (a) gmail .com to discuss this and make a good development plan;

Best regards
Zied

PVS() may return false MATE scores

It appears to me that it is possible, when no ttMove is present, for every move to be pruned without being searched in PVS(), thus resulting in movesSearched = 0, and then a mate score being returned.

I tested locally adding this condition to moveIsPrunabled : movesSearched > 0
889 - 883 - 1596 [0.501] 3368 @ 5s+.05s

Seems unintended.