Cornfed wrote: ↑Fri Nov 13, 2020 6:06 pm
Still about 80 points behind a similar current Stockfish. Not bad.
The performance of Dragon is -14 elo according to Stockfish 12 at 6 cores, 2 min + 0.5 sec TC
The performance of Dragon is -18 elo according to Stockfish 12 at 6 cores, 1 min + 0.5 sec TC
The performance of Dragon is -46 elo according to Stockfish 12 at 1 core, 10 sec + 0.2 sec TC http://talkchess.com/forum3/viewtopic.p ... &start=130
No one should be surprised if Dragon beats Stockfish in the TCEC Tournament.
Re: Dragon by Komodo Chess
Posted: Fri Nov 13, 2020 6:49 pm
by AdminX
Playing a LTC game that is not yet completed, but now at move 79 and still in the middle game.
[d]5k2/1pnr1p2/p1pr2p1/P3q2p/1PR4P/4P1P1/Q1R1BPK1/8 w - - 34 78
Cornfed wrote: ↑Fri Nov 13, 2020 6:06 pm
Still about 80 points behind a similar current Stockfish. Not bad.
The performance of Dragon is -14 elo according to Stockfish 12 at 6 cores, 2 min + 0.5 sec TC
The performance of Dragon is -18 elo according to Stockfish 12 at 6 cores, 1 min + 0.5 sec TC
The performance of Dragon is -46 elo according to Stockfish 12 at 1 core, 10 sec + 0.2 sec TC http://talkchess.com/forum3/viewtopic.p ... &start=130
No one should be surprised if Dragon beats Stockfish in the TCEC Tournament.
I not sure as of now. I saw scaling issues with Stockfish 12. Like your results, and form other testing sites. This issues is now better with the current stockfish.
My question is if Dragon is now #2 and is better then Lc0. That will be my next test.
Regardless Dragon can fight for the best engine spot. As this is the first release of Dragon. And is very close to the first release of Stockfish NNUE.
Cornfed wrote: ↑Fri Nov 13, 2020 6:06 pm
Still about 80 points behind a similar current Stockfish. Not bad.
The performance of Dragon is -14 elo according to Stockfish 12 at 6 cores, 2 min + 0.5 sec TC
The performance of Dragon is -18 elo according to Stockfish 12 at 6 cores, 1 min + 0.5 sec TC
The performance of Dragon is -46 elo according to Stockfish 12 at 1 core, 10 sec + 0.2 sec TC http://talkchess.com/forum3/viewtopic.p ... &start=130
No one should be surprised if Dragon beats Stockfish in the TCEC Tournament.
Right, but about 80 (think it was actually 75) behind the current development version most everyone has, right?
Or did I read the chart wrong?
Cornfed wrote: ↑Fri Nov 13, 2020 6:06 pm
Still about 80 points behind a similar current Stockfish. Not bad.
The performance of Dragon is -14 elo according to Stockfish 12 at 6 cores, 2 min + 0.5 sec TC
The performance of Dragon is -18 elo according to Stockfish 12 at 6 cores, 1 min + 0.5 sec TC
The performance of Dragon is -46 elo according to Stockfish 12 at 1 core, 10 sec + 0.2 sec TC http://talkchess.com/forum3/viewtopic.p ... &start=130
No one should be surprised if Dragon beats Stockfish in the TCEC Tournament.
I not sure as of now. I saw scaling issues with Stockfish 12. Like your results, and form other testing sites. This issues is now better with the current stockfish.
My question is if Dragon is now #2 and is better then Lc0. That will be my next test.
Regardless Dragon can fight for the best engine spot. As this is the first release of Dragon. And is very close to the first release of Stockfish NNUE.
For now I don' t think that Dragon is stronger than Stockfish Dev. But I think elo difference is very little at TCEC conditions. Komodo scales better with increasing time according to Stockfish. Komodo has been using net for a short time. They have the resources of Chess.com.
I predict that much better nets will come out until the next TCEC tournament. Because of that I think Dragon can beat Stockfish in the next TCEC tournament.
Cornfed wrote: ↑Fri Nov 13, 2020 6:06 pm
Still about 80 points behind a similar current Stockfish. Not bad.
At 5m+5s on a 32 thread 2950x. Dragon is doing much better then 80 Elo. Very good for the first release of Dragon.
Mark L. asked my to change the default setting for hash memory from default. I do not know if this changed the results. Default was 64, and was changed to 512. The suggested best setting for this many threads.
Here is the results playing the best Stockfish.
Dragon by Komodo Chess vs Stockfish 101120 (TC=5m+5s)
With 198 games played so far. Dragon vs Stockfish 101120 +6 =172 -20. 46.5% TP = -25 Elo.
Live Stream
I read on TCEC that they have NOT received the Komodo Dragon version which they ae interested in testing it on FRC very soon, I guess that Mr. Kaufman has NOT sent them the program yet.
Chess.com asked that we not give it to competing events (which basically means TCEC) for thirty days, and since chess.com has financed much of the development of Dragon that is a reasonable request. In my opinion we would probably be favored over SF in such events in FRC (960) now, but not yet in standard chess.
Re: Dragon by Komodo Chess
Posted: Fri Nov 13, 2020 8:30 pm
by mehmet123
lkaufman wrote: ↑Fri Nov 13, 2020 8:21 pm
Chess.com asked that we not give it to competing events (which basically means TCEC) for thirty days, and since chess.com has financed much of the development of Dragon that is a reasonable request. In my opinion we would probably be favored over SF in such events in FRC (960) now, but not yet in standard chess.
The request of Chess.com is very logical. But I don' think TCEC Premiere league begins in 30 days. There will be Qualification League, League 3,League2 and League 1
Re: Dragon by Komodo Chess
Posted: Fri Nov 13, 2020 8:48 pm
by Madeleine Birchfield
mehmet123 wrote: ↑Fri Nov 13, 2020 8:30 pm
The request of Chess.com is very logical. But I don' think TCEC Premiere league begins in 30 days. There will be Qualification League, League 3,League2 and League 1
The TCEC FRC bonus 2 will be starting soon, and the chess.com request prevents Dragon from playing in the TCEC FRC bonus 2.
Jhoravi wrote: ↑Thu Nov 12, 2020 4:07 am
Hi Larry, Is there any chance of NNUE specialized for losing positions for use in handicap matches?
We actually did include some such positions in the training which might help a bit in such matches, probably not a lot though. The problem is that if the handicap is high enough to give humans a good chance (like knight odds or even two pawns), the stronger side almost always wins in self play so it can't really learn how to trick the opponent. I suppose we could train it by playing against a crippled version of itself, but that might mess up normal play so we haven't tried that. We are considering possible opponents for such a match. I wonder, which players would readers most like to see facing Dragon in a handicap match (other than Magnus Carlsen who has never shown interest in playing such matches with engines)?
How does Dragon fare at giving knight odd in slow bullet? My dream is to see Hikaru Nakamura against Dragon at knight odds in many games 2 '+ 1 " in a row, like 12 or 24 games.
While perhaps Hikaru would still be the favorite, it could be a periodic event to track Komodo's progress.
Jhoravi wrote: ↑Thu Nov 12, 2020 4:07 am
Hi Larry, Is there any chance of NNUE specialized for losing positions for use in handicap matches?
We actually did include some such positions in the training which might help a bit in such matches, probably not a lot though. The problem is that if the handicap is high enough to give humans a good chance (like knight odds or even two pawns), the stronger side almost always wins in self play so it can't really learn how to trick the opponent. I suppose we could train it by playing against a crippled version of itself, but that might mess up normal play so we haven't tried that. We are considering possible opponents for such a match. I wonder, which players would readers most like to see facing Dragon in a handicap match (other than Magnus Carlsen who has never shown interest in playing such matches with engines)?
How does Dragon fare at giving knight odd in slow bullet? My dream is to see Hikaru Nakamura against Dragon at knight odds in many games 2 '+ 1 " in a row, like 12 or 24 games.
While perhaps Hikaru would still be the favorite, it could be a periodic event to track Komodo's progress.
It is a dream to me as well.
I would like to see this too, but I don't think that even Dragon is strong enough to make this a good fight yet. 2' + 1" is about the same as 3' with no inc when playing a computer, as 60 move games are probably normal. This is just standard blitz, and regular Komodo did poorly in some 3' + 2" games with GM Lenderman about six months ago. He was about 2800 on chess.com blitz, while Hikaru is well over 3200 (he beat a 3000+ GM by 21 to 5 today). I'd want to see Dragon decisively win a test blitz match at knight odds against Lenderman or another GM of similar strength before offering it to Hikaru. The basic problem is that knight odds is really equal to four pawn handicap, not three, since knights are worth about four pawns in the opening position. Giving four pawns to the world's number 1 or 2 blitz player in blitz is a Herculean task!
