Chess System Tal 2.1 EAS

[Ponti]

Hi... does CST 2.1 EAS run under Linux ? Is there a binary already made, or do I have to compile it ?

[OneTrickPony]

Some of those games are just unreal, what an engine!
I downloaded the executable from the rating list site. Is it self contained or do I need a NNUE file to get the amazing chess it plays?
I've left save/load NNUE fields empty and it seems to work. No multiPV mode. Is all of that expected?

[Graham Banks]

Some of those games are just unreal, what an engine!
I downloaded the executable from the rating list site. Is it self contained or do I need a NNUE file to get the amazing chess it plays?
I've left save/load NNUE fields empty and it seems to work. No multiPV mode. Is all of that expected?

Yes. No separate NNUE file.

[Modern Times]

Hi... does CST 2.1 EAS run under Linux ? Is there a binary already made, or do I have to compile it ?

I think only Windows exes are available, and it is closed source. I might be wrong. Perhaps it runs under WINE.

[pohl4711]

And not to forget: CSTal does not support to use more threads on a machine than there are physical cores. And the max. number of threads is very limited (max 64). So, no TCEC, not only no multiPV analysis and no Linux and no Android binaries, no skill level for humans, closed sources and no support for Syzygybases.

But perhaps you try out Patricia 5 instead? Patricia is open source, plays incredible aggressive (and not to forget, around +100 Celo stronger than CSTal 2.1 EAS), has a working multiPV analysis, a skill-level system, supports real multithreading (runs fine in TCEC), supports Syzygybases and Jim Ablett made great binaries: Windows, Linux, Android.
https://drive.proton.me/urls/YTAHHXANQ4#HJ9O7qoQZIxH

So, the playing-style of Patricia 5 and CSTal 2.1 EAS is quite similar (and really great!), but in all other aspects, Patricia is way better.

[Peter Berger]

So, the playing-style of Patricia 5 and CSTal 2.1 EAS is quite similar (and really great!), ..

I don't think this is true at all. Patricia and CSTal play a completely different game of chess. What you personally like better, is a matter of taste of course.

I think both you and Ed do interesting things to try and make the question of style more measurable, but the personality of an engine can still be all different even if they score highly according to similar parameters.

A very classical though most mysterious parameter that people tend to appreciate and value highly is to play "human-like". I personally feel CSTal is way ahead in this regard.

[IanO]

Darn, I thought someone made a CS Tal module for my Fidelity Elite Auto-Sensory board. Oh well.

[mclane]

Many chess engines have huge problems to successfully play a gambit opening. If you force that gambit opening line, the engine often tries to get back material and to get back material what completely ruins the idea of the gambit.

Humans have not so much problems to follow the idea of a gambit opening.



Maybe CSTal could, in opposite to other engines, successfully play gambit openings.

[jefk]

Maybe CSTal could, in opposite to other engines, successfully play gambit openings.

maybe, but if i look at the evals which i see during games with gambits, i notice that Patr-5
seems much more comfortable with sacrificing pawns (or similar, exchange vs 1 pawn).