SF-Pro 15.07.2023

Discussion about chess-playing software (engines, hosts, opening books, platforms, etc...)
Post Reply
Eduard Nemeth
Posts: 617
Joined: Thu Jun 10, 2010 12:50 am

Re: SF-Pro 15.07.2023

Post by Eduard Nemeth » Wed Nov 22, 2023 5:18 pm

I don't need your games or books. I only use server games for myself. But if you do a tournament, then you should also offer the games for download, that's just my opinion. You can download all games of my tests on the testing page of my homepage. Next: What does this have to do with evaluating the engine? :roll: Nothing. If your engine shows 0.00 at depth 43 and mine shows +1.25 at depth 30, then that is a problem with the engine and not the book. This variant is a draw if both engines play well.
Peace be with you!

APOCALYPSE
Posts: 213
Joined: Sat Mar 26, 2022 9:37 am
Contact:

Re: SF-Pro 15.07.2023

Post by APOCALYPSE » Wed Nov 22, 2023 7:22 pm

Eduard Nemeth wrote:
Wed Nov 22, 2023 5:18 pm
I don't need your games or books. I only use server games for myself. But if you do a tournament, then you should also offer the games for download, that's just my opinion. You can download all games of my tests on the testing page of my homepage. Next: What does this have to do with evaluating the engine? :roll: Nothing. If your engine shows 0.00 at depth 43 and mine shows +1.25 at depth 30, then that is a problem with the engine and not the book. This variant is a draw if both engines play well.
Okay, I'll check for the error today.

Eduard Nemeth
Posts: 617
Joined: Thu Jun 10, 2010 12:50 am

Re: SF-Pro 15.07.2023

Post by Eduard Nemeth » Wed Nov 22, 2023 7:55 pm

Here are 2 analyses:

SolistaChess,SL-BETA - Mustafacanseven,SF-PRO 1-0 0.0, Rated game, 5 min 2023
k5r1/1pp1q3/p4b2/2p1p2r/1PN1Pp1p/P1BP1PpQ/2P3K1/4R1R1 w - - 0 31

Analysis by INCOGNITO 4-avx2:

31.Rb1 cxb4 32.axb4 b6 33.Rb2 Ka7 34.Rgb1 Rd8 35.Ra1 Kb7 36.Be1 Qe8 37.Rba2 Ra8 38.Nd2 Be7 39.Nb3 Qf8 40.c3 Qg8 41.c4 Qd8 42.Rd1 Qg8 43.Bc3 Rd8 44.Rb2 Ka8 45.c5 Rg5 46.b5
+/- (1.11) Depth: 29/46 00:00:30 311MN, tb=44
White is better

Analysis by Stockfish dev-20231120-b4e9ee72:

31.Rb1 cxb4 32.axb4 b6 33.Rb2 Kb7 34.Ra1 Rd8 35.Rba2 Ra8 36.Bd2 Rh7 37.Rh1 Rd8 38.Bc3 Rh5 39.Rha1
+/= (0.42 ++) Depth: 44/38 00:00:32 357MN, tb=2895
White is slightly better

INCOGNITO 4:
+/- (1.11) Depth: 29/46 (only Depth 29 needed for +1)
(Download here: https://solistachess.jimdosite.com/solista-news/)

STOCKFISH dev:
+/= (0.42 ++) Depth: 44/38

Cool Iris 11.50:

k5r1/1pp1q3/p4b2/2p1p2r/1PN1Pp1p/P1BP1PpQ/2P3K1/4R1R1 w - - 0 1

Analysis by Cool Iris 11.50_AVX2:

31.Re2
= (0.30 ++) Depth: 40/53 00:00:45 467MN, tb=829
White has an edge

Test it yourself.
Last edited by Eduard Nemeth on Wed Nov 22, 2023 8:08 pm, edited 1 time in total.
Peace be with you!

APOCALYPSE
Posts: 213
Joined: Sat Mar 26, 2022 9:37 am
Contact:

Re: SF-Pro 15.07.2023

Post by APOCALYPSE » Wed Nov 22, 2023 8:05 pm

Eduard Nemeth wrote:
Wed Nov 22, 2023 7:55 pm
Here are 2 analyses:

SolistaChess,SL-BETA - Mustafacanseven,SF-PRO 1-0 0.0, Rated game, 5 min 2023
k5r1/1pp1q3/p4b2/2p1p2r/1PN1Pp1p/P1BP1PpQ/2P3K1/4R1R1 w - - 0 31

Analysis by INCOGNITO 4-avx2:

31.Rb1 cxb4 32.axb4 b6 33.Rb2 Ka7 34.Rgb1 Rd8 35.Ra1 Kb7 36.Be1 Qe8 37.Rba2 Ra8 38.Nd2 Be7 39.Nb3 Qf8 40.c3 Qg8 41.c4 Qd8 42.Rd1 Qg8 43.Bc3 Rd8 44.Rb2 Ka8 45.c5 Rg5 46.b5
+/- (1.11) Depth: 29/46 00:00:30 311MN, tb=44
White is better

Analysis by Stockfish dev-20231120-b4e9ee72:

31.Rb1 cxb4 32.axb4 b6 33.Rb2 Kb7 34.Ra1 Rd8 35.Rba2 Ra8 36.Bd2 Rh7 37.Rh1 Rd8 38.Bc3 Rh5 39.Rha1
+/= (0.42 ++) Depth: 44/38 00:00:32 357MN, tb=2895
White is slightly better

INCOGNITO 4:
+/- (1.11) Depth: 29/46 (only Depth 29 needed for +1)
(Download here: https://solistachess.jimdosite.com/solista-news/)

STOCKFISH dev:
+/= (0.42 ++) Depth: 44/38

Test it yourself.
My engine comes with a special add-on that I did not give to cubal because these add-ons are only given when purchasing the engine.

But I will definitely check again for possible errors.

My personal tests are carried out honestly, you can even watch it in the video clip that I attached in this post, and the tests are carried out with the following parameters:
Threads = 1
Hash = 512 MB
Ponder = ON
Time = 2min.+1sec.

Best regards.

Eduard Nemeth
Posts: 617
Joined: Thu Jun 10, 2010 12:50 am

Re: SF-Pro 15.07.2023

Post by Eduard Nemeth » Wed Nov 22, 2023 8:15 pm

As I said, your adds-on have nothing to do with the skill level. You also don't specify which settings are the best and which setting plays in your tournaments.

Threads = 1
Hash = 512MB
Ponder = ON
Time = 2min.+1sec.

And what should I do with it now? It's like kindergarten. Nobody plays publicly on servers with 1 thread, and almost nobody plays Bullet on PlayChess.com - because it's unimportant, that's why nobody plays it, we play Blitz, usually 5 mins.
Peace be with you!

APOCALYPSE
Posts: 213
Joined: Sat Mar 26, 2022 9:37 am
Contact:

Re: SF-Pro 15.07.2023

Post by APOCALYPSE » Wed Nov 22, 2023 8:23 pm

Eduard Nemeth wrote:
Wed Nov 22, 2023 8:15 pm
As I said, your adds-on have nothing to do with the skill level. You also don't specify which settings are the best and which setting plays in your tournaments.

Threads = 1
Hash = 512MB
Ponder = ON
Time = 2min.+1sec.

And what should I do with it now? It's like kindergarten. Nobody plays publicly on servers with 1 thread, and almost nobody plays Bullet on PlayChess.com - because it's unimportant, that's why nobody plays it, we play Blitz, usually 5 mins.
I am testing the engine on one thread, the Stockfish team also conducts tests on one thread, as for the engine settings, this is a secret.
I'm not interested in PlayChess or games on it at all.

APOCALYPSE
Posts: 213
Joined: Sat Mar 26, 2022 9:37 am
Contact:

Re: SF-Pro 15.07.2023

Post by APOCALYPSE » Thu Nov 23, 2023 11:40 am

Eduard Nemeth wrote:
Wed Nov 22, 2023 8:15 pm
As I said, your adds-on have nothing to do with the skill level. You also don't specify which settings are the best and which setting plays in your tournaments.

Threads = 1
Hash = 512MB
Ponder = ON
Time = 2min.+1sec.

And what should I do with it now? It's like kindergarten. Nobody plays publicly on servers with 1 thread, and almost nobody plays Bullet on PlayChess.com - because it's unimportant, that's why nobody plays it, we play Blitz, usually 5 mins.
Eduard Nemeth, provide me with your email and I will send you my private engine (without additional commercial files) with corrections, for your private tests only :?: :!:
Regards.

Eduard Nemeth
Posts: 617
Joined: Thu Jun 10, 2010 12:50 am

Re: SF-Pro 15.07.2023

Post by Eduard Nemeth » Thu Nov 23, 2023 12:59 pm

I'm not interested. I just wanted to discuss here. The result of a bullet tournament with only one thread says nothing about how well the engine would play on servers where games are played with longer time controls and more cores. The Incognito 4 engine, for example, is not strong in Bullet (not on 1 thread anyway). But if someone plays with it on 6 cores or more, then this engine is stronger than Stockfish. That's why I posted this PlayChess game as proof (Incognito 4 played with 12 threads = 6 cores). I want to have an engine that can win on servers and not in bullet and not offline on just 1 thread (that's a joke, sorry :x ).
Peace be with you!

APOCALYPSE
Posts: 213
Joined: Sat Mar 26, 2022 9:37 am
Contact:

Re: SF-Pro 15.07.2023

Post by APOCALYPSE » Thu Nov 23, 2023 1:12 pm

Eduard Nemeth wrote:
Thu Nov 23, 2023 12:59 pm
I'm not interested. I just wanted to discuss here. The result of a bullet tournament with only one thread says nothing about how well the engine would play on servers where games are played with longer time controls and more cores. The Incognito 4 engine, for example, is not strong in Bullet (not on 1 thread anyway). But if someone plays with it on 6 cores or more, then this engine is stronger than Stockfish. That's why I posted this PlayChess game as proof (Incognito 4 played with 12 threads = 6 cores). I want to have an engine that can win on servers and not in bullet and not offline on just 1 thread (that's a joke, sorry :x ).
And here what you are saying, you still did not understand me, I just wanted to say that I conduct my tests on 1 Threads.
And yet, I don’t need to show my emotions, they are of no use.
Thank you for drawing my attention to the engine error (I have already corrected it).

APOCALYPSE
Posts: 213
Joined: Sat Mar 26, 2022 9:37 am
Contact:

Re: SF-Pro 15.07.2023

Post by APOCALYPSE » Thu Nov 23, 2023 1:20 pm

Image

Post Reply