Chess2u
Would you like to react to this message? Create an account in a few clicks or log in to continue.

Chess2uLog in

ANOTHER EXPERIMENT

descriptionANOTHER EXPERIMENT EmptyANOTHER EXPERIMENT

more_horiz
Hi dear chess engines fans! Inspired by the interesting @RSilvian experiment about learning,
SINCE I THINK THAT MIXING .EXP FILES BETWEEN STOCKFISH DERIVATIVES ISN'T SO BAD
(after that on 4nd learning run ProteusSF RBE 008a has won it has scored +7 against my @Mehmet1921 's nightmare engine Kayra 1.4 and it has also scored +7 on 5th learning run against the very strong @dorsz 's Swordfish 15.2 using always TCEC 22
Superfinal Jeroen Noomen's selected 50 positions - 100 games repeating every position on reverse side
(thanks @im48)
 
ANOTHER EXPERIMENT K14-3

All 100 Kayra 1.4's PGN GAMES (Please analyze them with me!)

ANOTHER EXPERIMENT Proteus-SF-RBE-008bvs-Swordfish15-2-5-MINS-2secs-7

All 100 SwordFish 15.2 PGN GAMES (Please analyze them with me!)

(Sorry @dorsz , It's only a test to demonstrate that mixed .exp files work fine. I know that is unfair to compare two engines if the second hasn't an experience file . Without Khalid Omar's learning feature they have the same strength wink )

I'VE STARTED THIS FINAL TEST:

5 mins + 2 secs / ALL | Tablebases Syzygy 6 man


ProteusSF RBE 008a + ProteusSF.exp (@Sarona + trained ProteusSF.exp + @Igor 1959 data)

versus

Eman 7.93 Eman 05-05-22 Depth4 original untouched Eman.exp
downloaded from @Anton101 FB page.


NOTE: The file not found on Eman config is the Eman.bin book, to be sure that it not interferes.

Since it's a private engine, unfortunately I can run Khalid Omar's and ProteusSF engines only on Windows 11 ARM64 4 CPUs 2 MN/s (supporting only popcnt SSE41). My full respect and admiration to the programmer that with his learning code is changing chess engines history ONCE AGAIN (after Stockfish 1 released as FREE Open Source and NNUE smile )

FINAL RESULT:  +1 for ProteusSF RBE 00Bb. An Hitchcock thrilling until the end!!!

ANOTHER EXPERIMENT EM1

ANOTHER EXPERIMENT EM3

ALL 100 PGN games ProteusSF RBE 008b versus Eman 7.93


WOW!! Thank you @Sarona & @Igor for giving me your data to compile my ProteusSF.exp milkshake  clap (I don't want glorify myself as somebody does  rolleyes) Dear Mr. Khalid Omar I'm available as betatester for Eman 8.xx . Your learning code is sublime. Your engine's source is the only one that I'd like to examine. I don't care about AI, Zeus, Fischerov, Raubfisch and Solidor's clones.

Maestro training method is deprecated!!  bigsmile (Just kidding, I will apply it rigorously for next September Proteus Aurora 009 release, starting the training from scratch (but releasing also the ProteusSF.exp Milkshake Edition  cool )


ANOTHER EXPERIMENT Don-t-feed

Last edited by AlexChess on Mon 06 Jun 2022, 4:14 pm; edited 66 times in total

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
Eman is very strong engine!!!

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
Edit

Last edited by deeds on Fri 03 Jun 2022, 1:18 pm; edited 1 time in total

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
Edit

Last edited by deeds on Sun 05 Jun 2022, 8:09 am; edited 2 times in total

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
Proteus 008a vs Swordfish 15.2 59-41 is a very good result so I had to download PGN to see the lost games.

Something strange is there. It was not a match like you said:
"Noomen's selected 50 positions - 100 games repeating every position on reverse side"
It was not a fair match. All openings were played twice but engines had the same colors.

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
Edit

Last edited by deeds on Fri 03 Jun 2022, 5:08 pm; edited 1 time in total

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
I don't accuse Alex of cheating. It had to be just a mistake.

deeds wrote:
Edit

How do you edit your post. I could not find this option.

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
dorsz wrote:
I don't accuse Alex of cheating. It had to be just a mistake.

deeds wrote:
Edit

How do you edit your post. I could not find this option.


Dear @dorsz, sorry!

There must be something on BanksiaGui tournament configuration (recently updated to 0.54). Luckily I can edit and remove crossboard and games, while I understand what is happening. Obviously if I would cheat I would not publish all games (not saying that to you but to the troll  bigsmile )

Kind regards!

Alex

PS: Just updated ProteusSF RBE 008b with a lot of changes in the code (and added also DOTS5) , I restart the match from scratch and If I have the same issue I'll repeat also Kayra and Eman tests. Thank you for reporting this problem.

Last edited by AlexChess on Fri 03 Jun 2022, 5:52 pm; edited 1 time in total

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
dorsz wrote:
I don't accuse Alex of cheating. It had to be just a mistake.

deeds wrote:
Edit

How do you edit your post. I could not find this option.


The was NO SPECIAL option instead of "one for 2 games". Now I try this way and verify before publishing games and results.

Repeating all the games, now I'll watch all your games. We can play on Mac M1 6 Mn/s insted 2 Mn/s finally!


ANOTHER EXPERIMENT New


PS: I have an EDIT option on the right. (And my personal Troll out of the door  bigsmile  bigsmile )


RESTARTED.


ANOTHER EXPERIMENT RESET

Last edited by AlexChess on Fri 03 Jun 2022, 7:01 pm; edited 2 times in total

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
Edit

Last edited by deeds on Sat 04 Jun 2022, 3:00 am; edited 1 time in total

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
dorsz wrote:
I don't accuse Alex of cheating. It had to be just a mistake.

deeds wrote:
Edit

How do you edit your post. I could not find this option.



ANOTHER EXPERIMENT EDIT

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
@dorsz Yesterday I was feeling that there was something wrong +18 / 100 is abnormal. But I've checked engine settings, not games

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
deeds wrote:
dorsz wrote:
I don't accuse Alex of cheating. It had to be just a mistake.

Would you recognize a wolf in sheep's clothing in a flock of sheep ?
On the menu :
- pretend that it's the same engine in the next version when the source code has nothing to do
- replace default/best settings by bad/weaker settings
- rename source codes without any change
- attribute a result to its engine when it comes from an experience file trained by another stronger engine
- merge experience files from different engines
- publish non-fairplay results
- invent friendships
- etc.
We are no longer in "just a mistake"...


Partly true, but Alex will get better.

ANOTHER EXPERIMENT EDIT[/quote]

I don't have that button in the centre.

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
AlexChess wrote:
@dorsz Yesterday I was feeling that there was something wrong +18 / 100 is abnormal. But I've checked engine settings,  not games

Instead of +18/100 maybe use 59/100. In my opinion it looks clearer.

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
dorsz wrote:
AlexChess wrote:
@dorsz Yesterday I was feeling that there was something wrong +18 / 100 is abnormal. But I've checked engine settings,  not games

Instead of +18/100 maybe use 59/100. In my opinion it looks clearer.


Could you send me the games, for the hurry I've deleted Pixeldrain link and was my only copy rolleyes

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
mediafire.com file/qgnua3zkdcwhfft/P8a-SW15.2_pgn%25281%2529.pgn/file

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
(https://www.mediafire.com/file/qgnua3zkdcwhfft/P8a-SW15.2_pgn%25281%2529.pgn/file)

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
It has played all 50 positions C15-C65-E90-C42-A16, then repeated all 50 positions but not switching sides C15-C65-E90-C42-A16, So on second game it applies what it has just learned. Seems truly diabolical, you are right!  lol  lol  lol  lol

It's not me... is a wrong setting bigsmile Now it's ok C15-C15 ....C65-C65 switching sides. Luckily you have noticed that soon, since I want to test al top engines with TCEC and DOTS5. Until now I have used only bin books without switching sides on same position, so all other tests should be ok

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
ANOTHER EXPERIMENT P8B

ALL 31 SWORDFISH 15.2 GAMES UNTIL NOW

@dorsz I have always a lot of issues with BSG beta (crashed & timeout games, full tournaments  and setting of all engines suddenly disappearing, so I must often restart from scratch) but I still prefer it to ChessX and Cutechess on Mac, Pham is a dear friend and I want to support his work without criticizing him wink

Please let me know if it's all ok now. ProteusSF RBE 008b plays differently than ProteusSF 008a, it seems more Blue Marlin  smile

Reading my personal troll's comments from your quoted message: I'm more Willy Coyote than a wolf lol lol lol

ANOTHER EXPERIMENT Willy-coyote


ProteusSF RBE 008b DOWNLOAD LINKS

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
Edit

Last edited by deeds on Sun 05 Jun 2022, 8:09 am; edited 1 time in total

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
UPDATE 3

It's becoming a ProteusSF RBE 008b - Swordfish 15.2 match but is always ANOTHER EXPERIMENT about learning and I like to analyze it with you, since I don't truly know Khalid Omar (only exchanged some messages months ago asking for an Eman M1 build) ...I will repeat the match also against @Mehmet1921 'S KAYRA 1.4 against my new build ProteusSF RBE 008b (same bad setting "BSG not switching sides"  cry) before challenging Eman 7.93. Does somebody has Eman 8.01 and an updated Eman.exp , to be fair? @RSilvian @Anton101 @Mark62 @Zamchess @Zippo @Homayoun @Sarona @Igor @Solista ?

ANOTHER EXPERIMENT Sw3

ALL 45 GAMES UNTIL NOW: https://pixeldrain.com/u/mXrc182y


@dorsz Have you found the EDIT button? Which browser are you using? On Chrome Mac | Android it works.

Kind regards, Alex

PS: I'm restarting also the two DOTS 5 900 games tests ProteusSF RBE 008b vs SFMZ230522 and Kayra 1.4 on Android... BUT THIS TIME ONLY 5 mins 2 secs / ALL or it will never end bigsmile

Last edited by AlexChess on Sat 04 Jun 2022, 7:26 am; edited 5 times in total

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
I didn't have edit option, maybe because I am too new member or have too small number of posts?

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
dorsz wrote:
I didn't have edit option, maybe because I am too new member or have too small number of posts?



I don't know, but there should be a supervisor (if not a moderator) here to enable it to all members, since something has changed lately.


descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
UPDATE 4


+5

ANOTHER EXPERIMENT Sw4

ALL 55 GAMES UNTIL NOW: https://pixeldrain.com/u/ZCaZP81Z

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
Just Eman 7.93 and the exp file you know. Sorry.

descriptionANOTHER EXPERIMENT EmptyRe: ANOTHER EXPERIMENT

more_horiz
Permissions in this forum:
You cannot reply to topics in this forum