Official Release of Ethereal 11.75, supporting MultiPV

[ThatsIt]

Thanks for the link.

You should reconsider your complete test setup.

There are so much games like that:

[Site "ChessOwl"]
[Date "2019.11.10"]
[Round "4"]
[White "Smarthink 1.98 x64"]
[Black "Ethereal 11.73 x64"]
[Result "0-1"]
[ECO "A00"]
[PlyCount "19"]
[EventDate "2019.??.??"]

1. g4 {book} 1... c5 {book} 2. Nc3 {-0.11/14 1.8s} 2... d5 {+0.80/16 2.6s} 3.
g5 {-0.43/15 4.7s} 3... h6 {+1.17/17 3.6s} 4. Bg2 {-0.57/14 2.8s} 4... e6 {
+1.27/17 4.7s} 5. g6 {-0.86/14 1.4s} 5... fxg6 {+1.36/16 2.7s} 6. d4 {
-1.17/14 4.7s} 6... Nc6 {+1.45/16 2.1s} 7. Qd3 {-1.29/14 2.3s} 7... cxd4 {
+2.49/18 3.6s} 8. Qxg6+ {-1.65/14 2.7s} 8... Kd7 {+1.98/15 2.3s} 9. Nb1 {
-1.79/13 2.6s} 9... Qf6 {+2.46/17 3.9s} 10. Qg3 {
-2.02/15 2.1s, Black wins by adjudication} 0-1

Best wishes,
G.S.
(CEGT team)

[Guenther]

And here is the link of Ethereal 11.75 gauntlet done a few hours just after the Ethereal 11.73 was finished:
http://chessowl.blogspot.com/2019/11/et ... gines.html

A more invasive way of offering a download, as done on 'chessowl', is nearly impossible.

I checked the pgn and the search depths are a lottery, which was expected.
Obviously the cpu load is distributed totally uneven over all games.
More later. Also your adjudication rules are horrible.

[ThatsIt]

Hundreds of games like that:


[Event "Ethereal 11.73 vs. Top Chess Engines"]
[Site "ChessOwl"]
[Date "2019.11.10"]
[Round "14"]
[White "Critter 1.6a x64"]
[Black "Ethereal 11.73 x64"]
[Result "1-0"]
[ECO "A40"]
[PlyCount "26"]
[EventDate "2019.??.??"]

1. d4 {book} 1... e6 {book} 2. e4 {+0.36/15 3.3s} 2... d5 {-0.38/15 3.0s} 3.
Nc3 {+0.35/14 1.7s} 3... Bb4 {-0.28/15 2.3s} 4. e5 {+0.19/13 1.9s} 4... c5 {
-0.11/15 3.7s} 5. dxc5 {+0.20/13 3.2s} 5... Nc6 {+0.18/16 2.3s} 6. Nf3 {
+0.11/14 4.5s} 6... Nge7 {0.00/17 5.5s} 7. Bd3 {+0.13/14 1.9s} 7... a6 {
+0.08/17 6.9s} 8. Be3 {+0.49/14 2.3s} 8... O-O {-0.56/17 3.8s} 9. Bxh7+ {
+1.49/13 3.1s} 9... Kxh7 {-2.34/17 5.9s} 10. Ng5+ {+1.49/11 0.032s} 10... Kg6 {
-2.64/16 1.8s} 11. h4 {+1.82/13 1.9s} 11... Qe8 {-2.91/16 1.7s} 12. h5+ {
+1.82/11 0.027s} 12... Kh6 {-2.80/19 5.6s} 13. Qg4 {+6.05/12 0.99s} 13... Nf5 {
-2.52/18 2.1s, White wins by adjudication} 1-0

[Event "Ethereal 11.73 vs. Top Chess Engines"]
[Site "ChessOwl"]
[Date "2019.11.10"]
[Round "17"]
[White "Ethereal 11.73 x64"]
[Black "Fizbo 2.0 x64"]
[Result "1-0"]
[ECO "A04"]
[PlyCount "22"]
[EventDate "2019.??.??"]

1. Nf3 {book} 1... d6 {book} 2. d4 {+0.50/14 2.0s} 2... Bf5 {+0.04/13 2.2s} 3.
c4 {+0.43/16 5.1s} 3... e6 {0.00/13 2.1s} 4. Nc3 {+0.58/15 3.2s} 4... c6 {
-0.27/14 5.6s} 5. e4 {+0.90/15 2.9s} 5... Bg6 {-0.27/13 3.3s} 6. Qb3 {
+1.05/16 2.3s} 6... Qc7 {-0.21/13 2.9s} 7. Bf4 {+0.90/17 4.6s} 7... Qd7 {
-0.45/13 2.5s} 8. Be2 {+1.70/16 4.8s} 8... Nf6 {-0.68/13 2.0s} 9. e5 {
+3.01/16 2.4s} 9... dxe5 {-1.27/15 1.8s} 10. Nxe5 {+2.72/16 2.4s} 10... Qc8 {
-1.27/13 2.6s} 11. h4 {+3.38/17 2.0s} 11... Nbd7 {
-2.29/14 3.2s, White wins by adjudication} 1-0

[Event "Ethereal 11.73 vs. Top Chess Engines"]
[Site "ChessOwl"]
[Date "2019.11.10"]
[Round "26"]
[White "Naum 4.6 x64"]
[Black "Ethereal 11.73 x64"]
[Result "1-0"]
[ECO "B02"]
[PlyCount "24"]
[EventDate "2019.??.??"]

1. e4 {book} 1... Nf6 {book} 2. e5 {+0.44/14 2.2s} 2... Ng8 {-0.79/17 5.6s} 3.
Nf3 {+0.91/13 1.8s} 3... d6 {-0.88/16 2.7s} 4. Bb5+ {+0.80/12 1.9s} 4... c6 {
-0.44/17 3.5s} 5. Bc4 {+0.57/13 5.2s} 5... Bf5 {-0.57/16 2.2s} 6. d4 {
+0.89/12 2.4s} 6... dxe5 {-0.31/15 1.6s} 7. Nxe5 {+0.80/12 1.6s} 7... e6 {
-0.20/16 3.9s} 8. O-O {+0.57/12 2.8s} 8... Nd7 {-0.20/16 1.6s} 9. Qe2 {
+0.34/13 3.3s} 9... Ngf6 {-0.24/15 2.0s} 10. Nc3 {+0.42/12 1.0s} 10... Be7 {
-0.11/17 2.8s} 11. Rd1 {+0.42/12 2.3s} 11... O-O {-0.31/16 1.6s} 12. g4 {
+2.06/12 1.4s} 12... Nxg4 {-2.49/17 5.2s, White wins by adjudication} 1-0

Best wishes,
G.S.
(CEGT team)

[Guenther]

A year ago, I was using Arena 3.5 as my primary chess engines tournament program using a maximum of 7 instances to run a tournament. I got the inspiration to use 32 concurrencies with cutechess when Ethereal became very strong and Andrew revealed that he was using 32 bit concurrencies in cutechess to run million games to test Ethereal. I was intrigued, tested it myself and was blown that it could even run when I increased it to 64 concurrencies under Linux. This was a huge resource saver. Instead of using five 5 core AMD computers to run tournaments for rating list publication, I could just use one 4 core computer which is what I am having now.

I am also surprised that Andrew is not fully aware that Ethereal with Windows binary could run on Linux where WINE is installed. Perhaps, I am the only one running chess engine tournaments with all Windows binary executables under Linux. The best way to verify this is for the unbelievers to test it for themselves and then share their experiences.

The proof that Ethereal is working under cutechess and Linux is to visit my website chessowl.blogspot.com and download all the gauntlet pgn's then look the contents. All other primary engines running under the gauntlet also run with cutechess and Linux including all the Stockfish and Clones. The pgn's are unaltered and you could see what cutechess is commenting for each move. All of these engines are Windows binaries with .exe extension which you can also download in the tournament score board links.

Obviously Andrew has the hardware to do so. You don't, sorry.
The pgn reveals the expected behaviour.
All games are just random noise and for the trash bin.

[ThatsIt]

My advices:

delete all PGN files you have ever produced.

Set options for:

threads (option.Threads=1, sometimes Core or Cores)

hash (option.Hash=256, sometimes hash=256, case sensitivity)

concurrency 3 (-concurrency 3)

and if your are working on the machine:
concurrency 2 (-concurrency 2)

And please change your adjust rules.

Best wishes,
G.S.
(CEGT team)

[Ovyron]

Yes, if you want more games, reduce the engines' times, like, if instead of 1 minute games you want what concurrency 32 would give, give one core and 1.875 seconds per game to the engines, at least results will be fair. Otherwise the results are as good as random flips, where the coins are biased in favor of the engines that are the most aggressive in their resource allocation (to see an engine win all its games, set its Priority to Above Normal) and completely unusable.

[drj4759]

Thank you very much for your suggestions not related to Ethereal.

This is about Ethereal 11.75 and let's focus on the problem reported. I would appreciate if you could run Ethereal 11.75 on Linux with cutechess with concurrencies you desire and report if you have the same problem that I posted like "connection stalls". I would miss Ethereal 11.75 on my rating list in the current publication.

[RubiChess]

Thank you very much for your suggestions not related to Ethereal.

This is about Ethereal 11.75 and let's focus on the problem reported. I would appreciate if you could run Ethereal 11.75 on Linux with cutechess with concurrencies you desire and report if you have the same problem that I posted like "connection stalls". I would miss Ethereal 11.75 on my rating list in the current publication.

Why don't YOU correct your (obviously very bad) setup and see if the problem persists? It wouldn't be a big surprise to me if cutechess reports "connection stalls" just because the engine doesn't get any processing slot for quite some time.
You're rating list is probably quite useless, so missing E11.75 isn't your biggest problem.

Regards, Andreas

[drj4759]

It's probably better if you could set the perfect setup for Ethereal 11.75 under Linux/cutechess and then if it's running perfect, my "obviously bad setup" is causing the problem.

[RubiChess]

It's probably better if you could set the perfect setup for Ethereal 11.75 under Linux/cutechess and then if it's running perfect, my "obviously bad setup" is causing the problem.

Why should I? I'm not interested in testing Ethereal like you seem not interested in understanding what processor cores and concurrency mean.

Bye.