Chess System Tal 2 NNUE has been released

[Graham Banks]

Thorsten is conducting the same argument on a German language chess engine forum. They’ve slapped him around pretty thoroughly.

In the spirit of his tests, I’ve run matches on Linux (wine). CSTal2 lost 100% of games.

No time losses here though, but I'll now follow the instructions given anyway.

[Marek Soszynski]

Great News!!! Please Chris & Ed.... Release x64bit version without avx mandatory ... with 64-bit for old cpus, a lot of friends can use yours programs! My Q6600 is still in race! sadly cannot run Rebel 16.2 and Cstal 2 nnue . Thanks a Lot!

+1

[mclane]

Its not my job to setup the engines the way they work the best.

It is your job to tell the engine that you're using a f***ed up GUI that steals time from the engine.

If they want certain values beeing used, the could write it down in the readme file e.g.

Read here what they tell you about the Move Overhead option.

The problems stockfish has, especially with consecutive x in y time controls without increment go far deeper then move overhead problems in arena.

It's not even a sudden death time control.
It's a relatively easy x in y time control.
So no rocket science needed.

[mclane]

Its not my job to setup the engines the way they work the best.
What comes next ? Do i have to choose certain opening book that program x can win ? But why is there no opening book in the files for the engine ?

The problems with time controls in the arena gui is a new feature with Stockfish.
Older versions behave normal. So the programmers have changed something internally and this causes trouble.

If they want certain values beeing used, the could write it down in the readme file e.g.

You don't understand anything, otherwise you wouldn't write something like this. You live in the Stone Age of computer chess. You've probably never played chess online (there's not just the Fritz server that you don't like because it's from ChessBase), otherwise you wouldn't write like this. On game servers move overhead has to be set much higher because of LAG. This is the case everywhere where online games are played (also on Lichess). Otherwise your engine will lose almost every online game longer than 80 moves. Any 5-year-old learning chess online will know what to do. But the super expert Thorsten, who has been with computer chess since the beginning, doesn't know that. Isn't that funny! Move overhead has to be set correctly by the operator and nobody else. This applies to online chess, different GUIs, etc. You are responsible for it.

There is no "correct" move overhead. There is not even mentioned a value in the Readme file that comes with the engine. Not even in the explanation of the parameter is a value. All it says is that the value can be setup between 10 and 5000. But even with 1000 stockfish 16 has problems to play within the setup 40/5 or 40/10 time control.
It's correct that online chess does not interest me.

[mclane]

Thorsten is conducting the same argument on a German language chess engine forum. They’ve slapped him around pretty thoroughly.

In the spirit of his tests, I’ve run matches on Linux (wine). CSTal2 lost 100% of games.

They do attack any person who criticizes Stockfish. They do even have an own stockfish subforum but instead of posting there they spam the other forums or threads with this engine nonsense too. In a post about Chess System Tal they discuss about Stockfish. They are religious believers and they really believe that stockfish is like God or never loses or bullshit like this. They have similar religious symbol like the Christians and I do think they also believe stockfish walks over water and sits next to God in heaven.
In the moment another program beats stockfish in a game or many games these guys will either commit suicide or cancel their internet contract.

[marsell]

They do attack any person who criticizes Stockfish. They do even have an own stockfish subforum but instead of posting there they spam the other forums or threads with this engine nonsense too. In a post about Chess System Tal they discuss about Stockfish. They are religious believers and they really believe that stockfish is like God or never loses or bullshit like this. They have similar religious symbol like the Christians and I do think they also believe stockfish walks over water and sits next to God in heaven.
In the moment another program beats stockfish in a game or many games these guys will either commit suicide or cancel their internet contract.

Let God out of this discussion

[mclane]

Do you know fans of chess engines who would claim there engine never loses?!

Only the followers of the holy Stockfish believe so.

[RubiChess]

The problems stockfish has, especially with consecutive x in y time controls without increment go far deeper then move overhead problems in arena.

It's not even a sudden death time control.
It's a relatively easy x in y time control.
So no rocket science needed.

Thorsten, you already have proven in CSS that you are just too dumb for a proper setup even on a modern computer. So lets finsih this discussion. You won't learn anything.

[Damir]

Its not my job to setup the engines the way they work the best.

It is your job to tell the engine that you're using a f***ed up GUI that steals time from the engine.

If they want certain values beeing used, the could write it down in the readme file e.g.

Read here what they tell you about the Move Overhead option.

The problems stockfish has, especially with consecutive x in y time controls without increment go far deeper then move overhead problems in arena.

Thorsten,

Arena is rubbish. I would never use that GUI for engine testings, as I have written long time ago...

[mclane]

very good performance of CSTal:

Code: Select all

    Motor                                    Punkte  St Dr CS Lc Sh Be Re Et Cr Dr CS Ko De Se Wa Ch Bo De    S-B
01: Stockfish16 [t8, mo1000]                 25,0/34 ·· == == == 11 == == 1= 1= =1 =1 1= 11 == 1= 11 11 11  393,75
02: Dragon-3.1-64bit-avx2 [t8]               22,0/34 == ·· == == 1= == =0 11 == == 1= 1= =1 =1 =1 0= 11 11  354,25
03: CSTal2-v21-E1162-130-EASavx2 [t8*]       22,0/34 == == ·· == == == 1= == == == == =1 1= 11 1= 1= =1 11  346,75
04: Lc0 0.29.0 Nvidia RTX 3060 [j92-100]     21,5/34 == == == ·· == == == == == 1= 1= =1 == 1= == =1 11 11  339,00
05: ShashChess31-x86-64-avx2 [t8, Petrosian] 21,5/34 00 0= == == ·· 1= == == 11 1= 1= == =1 1= 11 1= 1= =1  335,00
06: Berserk-11.1 [t8]                        20,0/34 == == == == 0= ·· =1 0= == =1 1= == =1 == == 1= 11 =1  317,75
07: Rebel-16.2 [t8]                          18,5/34 == =1 0= == == =0 ·· == 1= 0= == 1= == == =1 =1 == 1=  301,25
08: Ethereal-14.00-pext-avx2 [t8]            18,5/34 0= 00 == == == 1= == ·· 0= == == == 1= == == 11 11 1=  284,00
09: Crystal_x64_060622_Tactical_E_AVX [t8]   17,5/34 0= == == == 00 == 0= 1= ·· == == == == 11 10 11 1= 0=  278,25
10: Dragon-3.1-64bit-avx2 [t8, mcts]         17,5/34 =0 == == 0= 0= =0 1= == == ·· == == =0 =1 1= 1= == 11  274,00
11: CSTal2-v21-E1019-Elo.avx2 [t8*]          17,5/34 =0 0= == 0= 0= 0= == == == == ·· =1 == == =1 =1 11 1=  266,25
12: Komodo-14.1-64bit [t8, mcts]             15,5/34 0= 0= =0 =0 == == 0= == == == =0 ·· == 1= 1= == =1 ==  245,75
13: DeepHiarcs15 [t8]                        14,0/34 00 =0 0= == =0 =0 == 0= == =1 == == ·· == == == == 01  223,75
14: Seer_v2.6 [t8]                           14,0/34 == =0 00 0= 0= == == == 00 =0 == 0= == ·· == 1= == 11  216,75
15: Wasp600-windows-avx [t8]                 14,0/34 0= =0 0= == 00 == =0 == 01 0= =0 0= == == ·· 11 == 1=  215,25
16: Chiron-5-64bit [t8]                      9,5/34  00 1= 0= =0 0= 0= =0 00 00 0= =0 == == 0= 00 ·· 01 ==  156,25
17: Booot6.5 [t8]                            9,0/34  00 00 =0 00 0= 00 == 00 0= == 00 =0 == == == 10 ·· ==  134,25
18: DeepShredder13UCIx64 [t8]                8,5/34  00 00 00 00 =0 =0 0= 0= 1= 00 0= == 10 00 0= == == ··  129,25

306 Partien gespielt / Turnier beendet
Name des Turniers: Ryzen9 3950x 2023
Ort/ Land: DESKTOP-D269HUC, Deutschland
Spielstufe: Turnier 40/10
Hardware: AMD Ryzen 9 3950X mit Nvidia Geforce RTX 3060 mit 12 GB Speicher