Dragon by Komodo Chess

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Uri Blass
Posts: 8969
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Dragon by Komodo Chess

Post by Uri Blass » Mon Nov 16, 2020 7:49 am

lkaufman wrote:
Mon Nov 16, 2020 7:11 am
JJJ wrote:
Mon Nov 16, 2020 12:18 am
lkaufman wrote:
Sun Nov 15, 2020 8:55 pm
JJJ wrote:
Sun Nov 15, 2020 6:37 pm
I would like to see penguin against Komodo at Bullet. It would be great !
What would you think would be a fair match? Knight odds at 1' + 1" perhaps?
I think it would be nice. I think So already player at Knight handicap against Stockfish some years ago at 1 min and lost badly :

https://www.youtube.com/watch?v=xu-pSaN ... el=KchessK
Of course no human has much chance against Dragon with knight odds at one minute without increment, the one second inc makes all the difference. Also I note that in the three games of SF vs So, Wesley never played the obviously correct reply 1...d5! to 1e4 when the b1 knight is missing, he probably didn't even look at knight odds before the games. In fact 1.e4 is just a bad move for White unless Black is forbidden to reply 1...d5. That's a big problem with knight odds. I suspect that many of Morphy's opponents either played with the restriction of answering 1e4 with 1...e5 or else did so on grounds of sportsmanship. We would either have to alternate knights and just avoid 1e4 when b1 is off, or add the 1e4 e5 restriction.
Or maybe they did not know that 1...d5 is the best reply against 1.e4 with knight odds.

I do not see why do you think that it should be obvious that 1...d5 is the correct reply when knight b1 is missing.

BrendanJNorman
Posts: 2360
Joined: Sun Feb 07, 2016 11:43 pm
Full name: Brendan J Norman

Re: Dragon by Komodo Chess

Post by BrendanJNorman » Mon Nov 16, 2020 8:02 am

Uri Blass wrote:
Mon Nov 16, 2020 7:49 am
lkaufman wrote:
Mon Nov 16, 2020 7:11 am
JJJ wrote:
Mon Nov 16, 2020 12:18 am
lkaufman wrote:
Sun Nov 15, 2020 8:55 pm
JJJ wrote:
Sun Nov 15, 2020 6:37 pm
I would like to see penguin against Komodo at Bullet. It would be great !
What would you think would be a fair match? Knight odds at 1' + 1" perhaps?
I think it would be nice. I think So already player at Knight handicap against Stockfish some years ago at 1 min and lost badly :

https://www.youtube.com/watch?v=xu-pSaN ... el=KchessK
Of course no human has much chance against Dragon with knight odds at one minute without increment, the one second inc makes all the difference. Also I note that in the three games of SF vs So, Wesley never played the obviously correct reply 1...d5! to 1e4 when the b1 knight is missing, he probably didn't even look at knight odds before the games. In fact 1.e4 is just a bad move for White unless Black is forbidden to reply 1...d5. That's a big problem with knight odds. I suspect that many of Morphy's opponents either played with the restriction of answering 1e4 with 1...e5 or else did so on grounds of sportsmanship. We would either have to alternate knights and just avoid 1e4 when b1 is off, or add the 1e4 e5 restriction.
Or maybe they did not know that 1...d5 is the best reply against 1.e4 with knight odds.

I do not see why do you think that it should be obvious that 1...d5 is the correct reply when knight b1 is missing.
Pretty self-evident isn't it?

The only drawback of the Scandinavian (to counterweight the overwhelming strength of a centralized queen) is that the queen can immediately be attacked with 3.Nb1-c3, costing black time and kicking the queen out of her nice post.

Without a knight on b1... :shock:

Another point in addition to this is that black having a more open game (Scandi is very good for this) allows him to trade pieces more readily, bringing the knight up ending closer and closer.

Madeleine Birchfield
Posts: 311
Joined: Tue Sep 29, 2020 2:29 pm
Location: Dublin, Ireland
Full name: Madeleine Birchfield

Re: Dragon by Komodo Chess

Post by Madeleine Birchfield » Mon Nov 16, 2020 8:33 am

BrendanJNorman wrote:
Mon Nov 16, 2020 8:02 am
Another point in addition to this is that black having a more open game (Scandi is very good for this) allows him to trade pieces more readily, bringing the knight up ending closer and closer.
Couldn't white avoid all that by pushing his pawn to e5 on move 2, yielding a French Defence/Advanced Caro-Kann type positions?

JJJ
Posts: 1344
Joined: Sat Apr 19, 2014 11:47 am

Re: Dragon by Komodo Chess

Post by JJJ » Mon Nov 16, 2020 9:48 am

lkaufman wrote:
Mon Nov 16, 2020 7:11 am
JJJ wrote:
Mon Nov 16, 2020 12:18 am
lkaufman wrote:
Sun Nov 15, 2020 8:55 pm
JJJ wrote:
Sun Nov 15, 2020 6:37 pm
I would like to see penguin against Komodo at Bullet. It would be great !
What would you think would be a fair match? Knight odds at 1' + 1" perhaps?
I think it would be nice. I think So already player at Knight handicap against Stockfish some years ago at 1 min and lost badly :

https://www.youtube.com/watch?v=xu-pSaN ... el=KchessK
Of course no human has much chance against Dragon with knight odds at one minute without increment, the one second inc makes all the difference. Also I note that in the three games of SF vs So, Wesley never played the obviously correct reply 1...d5! to 1e4 when the b1 knight is missing, he probably didn't even look at knight odds before the games. In fact 1.e4 is just a bad move for White unless Black is forbidden to reply 1...d5. That's a big problem with knight odds. I suspect that many of Morphy's opponents either played with the restriction of answering 1e4 with 1...e5 or else did so on grounds of sportsmanship. We would either have to alternate knights and just avoid 1e4 when b1 is off, or add the 1e4 e5 restriction.
Maybe you want to test Dragon Komodo before to have an idea how strong it ll play at bullet. In anycase, Penguin would love to challenge it. Also, you might consider many others handicap, even a 3 pawn handicap at bullet or maybe blitz to see how it does. Hikaru plays himself many handicap game against lower opponent, this time it would be the opposite.

User avatar
AdminX
Posts: 5611
Joined: Mon Mar 13, 2006 1:34 pm
Location: Acworth, GA
Contact:

Re: Dragon by Komodo Chess

Post by AdminX » Mon Nov 16, 2020 9:53 am

Only played 3 long-time control games so far, but I can see that this version of dragon is tenacious.

"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers

BrendanJNorman
Posts: 2360
Joined: Sun Feb 07, 2016 11:43 pm
Full name: Brendan J Norman

Re: Dragon by Komodo Chess

Post by BrendanJNorman » Mon Nov 16, 2020 11:14 am

Madeleine Birchfield wrote:
Mon Nov 16, 2020 8:33 am
BrendanJNorman wrote:
Mon Nov 16, 2020 8:02 am
Another point in addition to this is that black having a more open game (Scandi is very good for this) allows him to trade pieces more readily, bringing the knight up ending closer and closer.
Couldn't white avoid all that by pushing his pawn to e5 on move 2, yielding a French Defence/Advanced Caro-Kann type positions?
That's a smart idea, but not the same.

The difference is that you are already a piece down and wasting another tempo playing 2.e5.

So black can play 2...c5 before you've had a chance to play d4 and thus your advanced French structure isn't really happening.

Image

If you play 3.Nf3 then black gets 3...Bg4 when not only has be prevented your advanced French pawn chain but also has his "bad" bishop on its ideal square outside of the black pawn chain...this essentially means ZERO winning chances for white...Even without a piece down black is better in such positions.

Basically white is prevented from ever playing a favorable d4 advance, so it is hardly going to be a French OR Caro position, just a bad position.

Against a human GM, even Komodo would lose badly this semi-closed position, with an inferior position AND a piece down as white.

Madeleine Birchfield
Posts: 311
Joined: Tue Sep 29, 2020 2:29 pm
Location: Dublin, Ireland
Full name: Madeleine Birchfield

Re: Dragon by Komodo Chess

Post by Madeleine Birchfield » Mon Nov 16, 2020 12:03 pm

BrendanJNorman wrote:
Mon Nov 16, 2020 11:14 am
If you play 3.Nf3 then black gets 3...Bg4 when not only has be prevented your advanced French pawn chain but also has his "bad" bishop on its ideal square outside of the black pawn chain...this essentially means ZERO winning chances for white...Even without a piece down black is better in such positions.
I was taking a look at 3.c3 but that actually transposes into the Sicilian Alapan and I missed that black could play 3...d4 in response (I was taking a look at the 3...Bf5 4.d4 line which does lead to Caro/French structures).

lkaufman
Posts: 4569
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Dragon by Komodo Chess

Post by lkaufman » Mon Nov 16, 2020 3:57 pm

JJJ wrote:
Mon Nov 16, 2020 9:48 am
lkaufman wrote:
Mon Nov 16, 2020 7:11 am
JJJ wrote:
Mon Nov 16, 2020 12:18 am
lkaufman wrote:
Sun Nov 15, 2020 8:55 pm
JJJ wrote:
Sun Nov 15, 2020 6:37 pm
I would like to see penguin against Komodo at Bullet. It would be great !
What would you think would be a fair match? Knight odds at 1' + 1" perhaps?
I think it would be nice. I think So already player at Knight handicap against Stockfish some years ago at 1 min and lost badly :

https://www.youtube.com/watch?v=xu-pSaN ... el=KchessK
Of course no human has much chance against Dragon with knight odds at one minute without increment, the one second inc makes all the difference. Also I note that in the three games of SF vs So, Wesley never played the obviously correct reply 1...d5! to 1e4 when the b1 knight is missing, he probably didn't even look at knight odds before the games. In fact 1.e4 is just a bad move for White unless Black is forbidden to reply 1...d5. That's a big problem with knight odds. I suspect that many of Morphy's opponents either played with the restriction of answering 1e4 with 1...e5 or else did so on grounds of sportsmanship. We would either have to alternate knights and just avoid 1e4 when b1 is off, or add the 1e4 e5 restriction.
Maybe you want to test Dragon Komodo before to have an idea how strong it ll play at bullet. In anycase, Penguin would love to challenge it. Also, you might consider many others handicap, even a 3 pawn handicap at bullet or maybe blitz to see how it does. Hikaru plays himself many handicap game against lower opponent, this time it would be the opposite.
Could you give me some specifics about handicap games Hikaru has played against lower rated human opponents, such as what handicaps, what time controls, and what chess.com ratings the opponents had? I'm especially interested in games he played this way against titled players. I don't expect complete details, just a summary of whatever you know.
Komodo rules!

ProfessorShadoko
Posts: 1
Joined: Sun Sep 23, 2018 10:29 am
Full name: Jean-Philippe Karr

Re: Dragon by Komodo Chess

Post by ProfessorShadoko » Mon Nov 16, 2020 7:31 pm

lkaufman wrote:
Tue Nov 10, 2020 10:01 pm
Alayan wrote:
Tue Nov 10, 2020 7:42 pm
May the people rejoice while originality fades away.
lkaufman wrote:
Tue Nov 10, 2020 6:42 pm
When both engines were set to use MultiPV=4 for standard chess, Dragon MCTS was able to defeat SF12 decisively in a 3600 games blitz match, so we believe that on most computers it is the world's strongest engine when looking at four or more lines of play at once.
This marketing claim "forgets" to acknowledge that this increased strength of the first move is achieved by reducing how deep the engines looks at lines beyond the first one. Indeed, the ability to report multiple PVs with eval, and dedicating similar resources to explore each of the lines, are separate concepts.

Play with MultiPV=4 only reflects the quality of the first move, not that of the 2nd, 3rd or 4th, and is a fundamentally flawed way to evaluate how useful a MultiPV implementation is.

The increased quality of the first move in Komodo (and Leela) MCTS comes at the cost of a lowered quality and reliability for the moves, PV and evals of the 2nd and further lines come from smaller search trees. Turning MultiPV on in these engines doesn't increase the likelihood that they discover an initially disliked move is much better than first thought.

Most people that do use MultiPV for analysis expressed strong opposition to this trade-off when reducing 2nd and further PV lines was considered for Stockfish, though in some circumstances (moves that are much worse than the main move) it can be useful.

For deep analysis, multiPV isn't going to be used on positions with one clear forced move anyway, and the effective time-handicap that makes KMCTS look good at blitz matter much less while the search/eval deficit remain, making Stockfish the best MultiPV engine for e.g. correspondence chess.
It is true that with Komodo MCTS (or Lc0 and other MCTS engines) the second, third, and later lines are analyzed less than the first line. However it's based on how close they are to each other. In positions where several moves appear to be nearly equally good, the analysis of the later moves will be nearly as good as the analysis of the best move. In positions where there are say two plausible moves and the rest are obvious blunders, then the two plausible moves will be analyzed properly and the others will get short analysis. This is how it should work in my opinion. Spending equal time on obvious blunders is obviously not a sensible use of resources. Occasionally it will pay off, but that's not the usual case.

This is an interesting discussion, and perhaps for many users (including myself), a more important one than to determine which engine is the strongest one. It would be worth spending efforts to find out which engine is the best analysis tool.

Using (for example) MultiPV=3, this would mean being able to evaluate the strength of the 3 displayed moves. I have no idea about what would be the best way to do this, though. Maybe, from the positions after each of the 3 moves, play many games at a fast time control between high-level engines ?
Further, such tests might also be used to assess how accurate the engine's evaluation is ('accurate' meaning that it would conform to a given relationship between the evaluation score and the win probability). Which is also interesting for people using these engines as analysis tools.

JJJ
Posts: 1344
Joined: Sat Apr 19, 2014 11:47 am

Re: Dragon by Komodo Chess

Post by JJJ » Tue Nov 17, 2020 4:44 am

lkaufman wrote:
Mon Nov 16, 2020 3:57 pm
JJJ wrote:
Mon Nov 16, 2020 9:48 am
lkaufman wrote:
Mon Nov 16, 2020 7:11 am
JJJ wrote:
Mon Nov 16, 2020 12:18 am
lkaufman wrote:
Sun Nov 15, 2020 8:55 pm
JJJ wrote:
Sun Nov 15, 2020 6:37 pm
I would like to see penguin against Komodo at Bullet. It would be great !
What would you think would be a fair match? Knight odds at 1' + 1" perhaps?
I think it would be nice. I think So already player at Knight handicap against Stockfish some years ago at 1 min and lost badly :

https://www.youtube.com/watch?v=xu-pSaN ... el=KchessK
Of course no human has much chance against Dragon with knight odds at one minute without increment, the one second inc makes all the difference. Also I note that in the three games of SF vs So, Wesley never played the obviously correct reply 1...d5! to 1e4 when the b1 knight is missing, he probably didn't even look at knight odds before the games. In fact 1.e4 is just a bad move for White unless Black is forbidden to reply 1...d5. That's a big problem with knight odds. I suspect that many of Morphy's opponents either played with the restriction of answering 1e4 with 1...e5 or else did so on grounds of sportsmanship. We would either have to alternate knights and just avoid 1e4 when b1 is off, or add the 1e4 e5 restriction.
Maybe you want to test Dragon Komodo before to have an idea how strong it ll play at bullet. In anycase, Penguin would love to challenge it. Also, you might consider many others handicap, even a 3 pawn handicap at bullet or maybe blitz to see how it does. Hikaru plays himself many handicap game against lower opponent, this time it would be the opposite.
Could you give me some specifics about handicap games Hikaru has played against lower rated human opponents, such as what handicaps, what time controls, and what chess.com ratings the opponents had? I'm especially interested in games he played this way against titled players. I don't expect complete details, just a summary of whatever you know.
He played too many different handicap games and mostly against opponent with a very low rating compared to him. But he wins sometimes against higher players with strange odds like in this game : https://youtu.be/0sWXhA_I3G0?t=482, starting the video at : 8:02.

Also he wons against 2479 players here with different odds, but on time : https://www.youtube.com/watch?v=JYohwKpu5K8

It is not the same handicap you like to propose, but perhaps it would help you calculate your odds of winnings. You can write "Hikaru odds game" on youtube and find more exemple if you like.

Post Reply