Big-SF 200324

Post Reply
Solista
Posts: 202
Joined: Thu Jul 06, 2023 11:47 am
Location: Germany
Has thanked: 7 times
Been thanked: 293 times
Contact:

Big-SF 200324

Post by Solista »

After getting a lot of letters from all over the world and some other people also being dissatisfied with Stockfish dev, I decided to release another engine. The meaning is: The influence of the small network was reduced and the large network was implemented with dimension 3072 (instead of dimension 2560). I use this engine for Blitz and Bullet as well as for quick analysis. I prefer to use the large network on Lichess because it allows me to win more games against weaker engines because it values fewer positions 0.00.

Big-SF 200324 Win 64-Bit (avx2, avx512, bmi2, sse41, source code, external networks):
https://pixeldrain.com/u/xK1DY2NH

Implemented features "Minimum Thinking Time" and "Slow Mover". By the way, new D 3072 nets are currently being tested every day: https://tests.stockfishchess.org/nns. You can use these networks with the engine by copying the network into the folder and replacing the large network in the engine options. Have fun!

https://solistachess.jimdosite.com/
User avatar
massimilianogoi
Site Admin
Posts: 369
Joined: Thu Aug 04, 2022 1:42 pm
Has thanked: 569 times
Been thanked: 608 times
Contact:

Re: Big-SF 200324

Post by massimilianogoi »

Solista wrote: Wed Mar 20, 2024 9:24 am After getting a lot of letters from all over the world and some other people also being dissatisfied with Stockfish dev, I decided to release another engine. The meaning is: The influence of the small network was reduced and the large network was implemented with dimension 3072 (instead of dimension 2560). I use this engine for Blitz and Bullet as well as for quick analysis. I prefer to use the large network on Lichess because it allows me to win more games against weaker engines because it values fewer positions 0.00.
Not evaluating a position 0.00 doesn't mean that the engine makes a better move. On the opposite: I got better move choices with Cfish, where the evaluations were near 0.00, instead of other engines where the evaluation was floating and so the move choices. A key factor on move choices is the coherence, that means choosing always the same move when the same position is proposed more and more times.
People who have lost the hope.
Solista
Posts: 202
Joined: Thu Jul 06, 2023 11:47 am
Location: Germany
Has thanked: 7 times
Been thanked: 293 times
Contact:

Re: Big-SF 200324

Post by Solista »

22. Mar 2024: Big-SF 210324. Latest SF updates and new improved search for the large network. As I already wrote, the SF developers are testing new D 3072 networks. Here is a test version with the network "nn-56e7225a7727.nnue". The network is now embedded because some users are having problems with external networks (even though it's easy). Features: Slow Mover, Minimum Thinking Time. Download Win 64-Bit (avx2, bmi2, sse41, source code): https://pixeldrain.com/u/vqWuhCtX
Solista
Posts: 202
Joined: Thu Jul 06, 2023 11:47 am
Location: Germany
Has thanked: 7 times
Been thanked: 293 times
Contact:

Re: Big-SF 200324

Post by Solista »

Big-SF 220324 (dimension 3072). There is a new developer test version with new network nn-31636a26c4f6.nnue. I haven't changed anything this time, but I have implemented Minimum Thinking Time and Slow Mover for everyone who wants to play online.

Changed developer code:
https://github.com/official-stockfish/S ... f7c29444d4

The version with Dimension 3072 network against D 2560 is currently being tested on Fishtest, live:
https://tests.stockfishchess.org/tests/ ... 0526c52815

Download 64-bit (avx2, bmi2, sse41, source code):
https://pixeldrain.com/u/NMM14eoP

Have fun testing.
Fringe407
Posts: 20
Joined: Fri Mar 15, 2024 10:48 pm
Has thanked: 34 times
Been thanked: 36 times

Re: Big-SF 200324

Post by Fringe407 »

Very good work! Thanks for your contribution to Stockfish, Eduarth.

Best regards
Fringe
Solista
Posts: 202
Joined: Thu Jul 06, 2023 11:47 am
Location: Germany
Has thanked: 7 times
Been thanked: 293 times
Contact:

Re: Big-SF 200324

Post by Solista »

Fringe407 wrote: Sat Mar 23, 2024 8:42 am Very good work! Thanks for your contribution to Stockfish, Eduarth.

Best regards
Fringe
Have fun testing! By the way, congratulations to you on your clear victory against ProteusSF on Lichess! :greetings-clappingyellow:
Fringe407
Posts: 20
Joined: Fri Mar 15, 2024 10:48 pm
Has thanked: 34 times
Been thanked: 36 times

Re: Big-SF 200324

Post by Fringe407 »

Solista wrote: Sat Mar 23, 2024 9:14 am
Fringe407 wrote: Sat Mar 23, 2024 8:42 am Very good work! Thanks for your contribution to Stockfish, Eduarth.

Best regards
Fringe
Have fun testing! By the way, congratulations to you on your clear victory against ProteusSF on Lichess! :greetings-clappingyellow:
Dear Eduarth,

Congratulations on being the first to report a cheater.I agree with your actions and have proven your suspicions to be correct. I do not tolerate cheating and have no respect for those who do. I only lost one match due to my internet being disconnected when I was in a winning position of +4.0. Despite being humiliated, he still challenged me to a duel. He seems to enjoy losing. Eduarth, I enjoy following your work on engines and books. I sent you an email, but I heard you were in seclusion for a while. I would like to try the Light engine for my analyses. Thank you for your work.

Best regards,
Fringe.
Solista
Posts: 202
Joined: Thu Jul 06, 2023 11:47 am
Location: Germany
Has thanked: 7 times
Been thanked: 293 times
Contact:

Re: Big-SF 200324

Post by Solista »

Yes, I am currently only answering a few emails. There are many that I get. Please send me an email with the subject Light 26.3.
Fringe407
Posts: 20
Joined: Fri Mar 15, 2024 10:48 pm
Has thanked: 34 times
Been thanked: 36 times

Re: Big-SF 200324

Post by Fringe407 »

Dear Eduard,

I'll e-mail you right away. You're marvellous.

My best regards,
Fringe
Post Reply