Question about Stockfish 17

[chessica]

How do you know whether the "NUMA" function works? Output to the console
would be helpful for me. Can someone show it here?

[RubiChess]

You've told us a thousand times that SF17 is not stronger than 16, so why do you care about NUMA in Sf17?

[Frank Quisinsky]

Not relevant, it is a question!

Thinking on Aaron Tay (Singapur) and his work in winboard times.
On the other hand, about NUMA a lot of information can be found with a little bit work only.

[chessica]

You've told us a thousand times that SF17 is not stronger than 16, so why do you care about NUMA in Sf17?

Hi Andreas, I think you're confusing someone. I've never written anything against stockfish, in contrast I'm a big fan of stocki. In my MEA test that I published here, it's also relatively high up.

[chessica]

Not relevant, it is a question!

Thinking on Aaron Tay (Singapur) and his work in winboard times.
On the other hand, about NUMA a lot of information can be found with a little bit work only.

Hi Frank,

I don't want to know how it works, just how I can see from the engine
output whether numa or cpu is being used.

[Jouni]

You have multiple CPUs or CPUs with multiple NUMA domains? For normal user zero importance.

[syzygy]

How do you know whether the "NUMA" function works? Output to the console
would be helpful for me. Can someone show it here?

In the console, type this:

Code: Select all

setoption name Threads value 128
go depth 1

I picked 128 threads because with too few threads NUMA might not be play a role (I'm not sure).

My laptop, which has no NUMA and "only" 16 cores, I get this:

Code: Select all

Stockfish 17 by the Stockfish developers (see AUTHORS file)
setoption name Threads value 128
info string Using 128 threads
go depth 1
info string Available processors: 0-15
info string Using 128 threads
info string NNUE evaluation using nn-1111cefa1111.nnue (133MiB, (22528, 3072, 15, 32, 1))
info string NNUE evaluation using nn-37f18f62d772.nnue (6MiB, (22528, 128, 15, 32, 1))
info depth 1 seldepth 3 multipv 1 score cp 13 nodes 18479 nps 2639857 hashfull 2 tbhits 0 time 7 pv e2e4
bestmove e2e4 ponder e7e5
quit

If you have a NUMA machine, I think you should see something about NUMA in one of the "info string" lines.

[chessica]

How do you know whether the "NUMA" function works? Output to the console
would be helpful for me. Can someone show it here?

In the console, type this:

Code: Select all

setoption name Threads value 128
go depth 1

I picked 128 threads because with too few threads NUMA might not be play a role (I'm not sure).

My laptop, which has no NUMA and "only" 16 cores, I get this:

Code: Select all

Stockfish 17 by the Stockfish developers (see AUTHORS file)
setoption name Threads value 128
info string Using 128 threads
go depth 1
info string Available processors: 0-15
info string Using 128 threads
info string NNUE evaluation using nn-1111cefa1111.nnue (133MiB, (22528, 3072, 15, 32, 1))
info string NNUE evaluation using nn-37f18f62d772.nnue (6MiB, (22528, 128, 15, 32, 1))
info depth 1 seldepth 3 multipv 1 score cp 13 nodes 18479 nps 2639857 hashfull 2 tbhits 0 time 7 pv e2e4
bestmove e2e4 ponder e7e5
quit

If you have a NUMA machine, I think you should see something about NUMA in one of the "info string" lines.

Hello, it looks like this to me:


Stockfish 17 by the Stockfish developers (see AUTHORS file)
uci
id name Stockfish 17
id author the Stockfish developers (see AUTHORS file)

option name Debug Log File type string default <empty>
option name NumaPolicy type string default auto
option name Threads type spin default 1 min 1 max 1024
option name Hash type spin default 16 min 1 max 33554432
option name Clear Hash type button
option name Ponder type check default false
option name MultiPV type spin default 1 min 1 max 256
option name Skill Level type spin default 20 min 0 max 20
option name Move Overhead type spin default 10 min 0 max 5000
option name nodestime type spin default 0 min 0 max 10000
option name UCI_Chess960 type check default false
option name UCI_LimitStrength type check default false
option name UCI_Elo type spin default 1320 min 1320 max 3190
option name UCI_ShowWDL type check default false
option name SyzygyPath type string default <empty>
option name SyzygyProbeDepth type spin default 1 min 1 max 100
option name Syzygy50MoveRule type check default true
option name SyzygyProbeLimit type spin default 7 min 0 max 7
option name EvalFile type string default nn-1111cefa1111.nnue
option name EvalFileSmall type string default nn-37f18f62d772.nnue
uciok
setoption name Threads value 128
info string Using 128 threads
go depth 1
info string Available processors: 0-3
info string Using 128 threads
info string NNUE evaluation using nn-1111cefa1111.nnue (133MiB, (22528, 3072, 15, 32, 1))
info string NNUE evaluation using nn-37f18f62d772.nnue (6MiB, (22528, 128, 15, 32, 1))
info depth 1 seldepth 2 multipv 1 score cp 13 nodes 82 nps 9111 hashfull 0 tbhits 0 time 9 pv e2e4
bestmove e2e4 ponder e7e5

[chessica]

Cfish 060821 64 NUMA by Syzygy based on Stockfish
uci
id name Cfish 060821 64 NUMA
id author The Stockfish developers

option name Contempt type spin default 24 min -100 max 100
option name Analysis Contempt type combo default Off var Off var White var Black
option name Threads type spin default 1 min 1 max 512
option name Hash type spin default 16 min 1 max 33554432
option name Clear Hash type button
option name Ponder type check default false
option name MultiPV type spin default 1 min 1 max 500
option name Move Overhead type spin default 10 min 0 max 5000
option name Slow Mover type spin default 100 min 10 max 1000
option name nodestime type spin default 0 min 0 max 10000
option name UCI_AnalyseMode type check default false
option name UCI_Chess960 type check default false
option name SyzygyPath type string default <empty>
option name SyzygyProbeDepth type spin default 1 min 1 max 100
option name Syzygy50MoveRule type check default true
option name SyzygyProbeLimit type spin default 7 min 0 max 7
option name SyzygyUseDTM type check default true
option name BookFile type string default <empty>
option name BookFile2 type string default <empty>
option name BestBookMove type check default true
option name BookDepth type spin default 255 min 1 max 255
option name EvalFile type string default nn-62ef826d1a6d.nnue
option name Use NNUE type combo default Hybrid var Hybrid var Pure var Classical
option name LargePages type check default true
uciok
setoption name Threads value 128
go depth 1
info string Unable to allocate large pages for the transposition table.
info string Hybrid NNUE evaluation using nn-62ef826d1a6d.nnue enabled.
info depth 1 seldepth 1 multipv 1 score cp 48 nodes 2727750 nps 2347461 hashfull 1000 tbhits 0 time 1162 pv e2e4
bestmove e2e4 ponder e7e5

[RubiChess]

You've told us a thousand times that SF17 is not stronger than 16, so why do you care about NUMA in Sf17?

Hi Andreas, I think you're confusing someone. I've never written anything against stockfish, in contrast I'm a big fan of stocki. In my MEA test that I published here, it's also relatively high up.

You are right. I'm sorry, I confused you with Jouni.