Komodo 12.3 is out

[lkaufman]

Also please note that although 12.3 MCTS is still nearly a hundred elo below normal 12.3, it can already defeat 12.3 if both are in MultiPV = 2 (or more) mode. For this reason, I have switched to using the MCTS version myself when analyzing games or openings.
"Version 12.3 will probably be the last one that combines regular and MCTS Komodo, they will soon become distinct programs, although subscribers will still get both. My personal view is that the future looks very bright for Komodo MCTS, as it is still in its infancy or perhaps early childhood".

THIS...might just convince me to subscribe once again!

As I use Komodo 11.3.2 (or other engines) ONLY for analyzing games or openings I am curious as to how you or others do this. Stockfish 10 really isn't that suitable for such things as far as I can see.

I've an ordinary Desktop i7-6700 @3.4GHz with 16 GB RAM...like to use this computer for some minor things while analyzing over the course of an evening, so I generally limit engines under Chessbase 15 to 2 threads, 2-3 Multi-PV.

I do not remember seeing on the Komodo website (have not visited it in a while) anything about how to get the most out of Komodo MCTS for game or opening analysis...do you or anyone else have some tips or could you point me somewhere which would be of assistance?

Thanks!

The main tip is that with MCTS you can now feel free to look at all the moves you like; as far as we can tell, there is no elo loss from setting MultiPV to 2, 3, or even 218 ! With normal Komodo, MultiPV 2 costs something like 100 elo, MultiPV 3 another fifty or so, etc. Also with normal Komodo you have to pay attention to trying to finish iterations; with MCTS it makes sense to quit any time that the two best moves are not nearly equal.

[Werewolf]

Also please note that although 12.3 MCTS is still nearly a hundred elo below normal 12.3

I thought the previous Komodo MCTS was 100 Elo below normal Komodo? I’m sure I read that somewhere.

Surely the improvements have closed the gap?

[lkaufman]

Also please note that although 12.3 MCTS is still nearly a hundred elo below normal 12.3

I thought the previous Komodo MCTS was 100 Elo below normal Komodo? I’m sure I read that somewhere.

Surely the improvements have closed the gap?

The last version was about 100 elo below normal Komodo at longer time limits (maybe 40/40' or so), but quite a bit more than 100 at blitz. Version 12.3 should come out about 100 below normal at blitz, quite a bit less at 40/40. The devil is in the details with statistics.

One curious fact that I just noticed is that in self-play (i.e. playing two versions with just a small change between them), the percentage of draws is MUCH lower with Komodo MCTS than with normal Komodo, even if the time limits are adjusted to make them about the same strength. I'm not sure of the reason, but perhaps it's an inherent property of MCTS to play more actively and boldly than alpha-beta chess engines.

[Nordlandia]

For fritz cloud matches i usually use 250 ms - 300 ms as move overhead.

What i can't comprehend is why can't ponder be available as an option in the chessbase engine room for machine vs machine matches?

Playchess - Computer Chess - Engine Room. Require subscription.

The way it is now -> Engine play each other over the internet without taking advantage of thinking during opponents time

[Ozymandias]

anything about how to get the most out of Komodo MCTS for game or opening analysis...do you or anyone else have some tips or could you point me somewhere which would be of assistance?

What Larry says. I've gone up all the way to 9 MPV lines and there's no indication of a slowdown, depth wise. Under the Fritz 16 GUI, kn/s stay between 1 and 2 on a regular computer, so that's not of much help when it comes to gauging speed.

In essence, you can have many lines updating simultaneously without any apparent penalty. That's very useful for analysis.

[Gabor Szots]

May I have a question. Let's say I'mwilling to allocate 256 MB of RAM altogether for Komodo MCTS, playing 40/4 games. Is that amount enough? What is the optimal distribution of RAM among the 3 areas Komodo uses (table momory, hash, MCTS hash). In general, what is the optimal ration between them using x MB total RAM?

[Joerg Oster]

Also please note that although 12.3 MCTS is still nearly a hundred elo below normal 12.3, it can already defeat 12.3 if both are in MultiPV = 2 (or more) mode. For this reason, I have switched to using the MCTS version myself when analyzing games or openings.
"Version 12.3 will probably be the last one that combines regular and MCTS Komodo, they will soon become distinct programs, although subscribers will still get both. My personal view is that the future looks very bright for Komodo MCTS, as it is still in its infancy or perhaps early childhood".

THIS...might just convince me to subscribe once again!

As I use Komodo 11.3.2 (or other engines) ONLY for analyzing games or openings I am curious as to how you or others do this. Stockfish 10 really isn't that suitable for such things as far as I can see.

I've an ordinary Desktop i7-6700 @3.4GHz with 16 GB RAM...like to use this computer for some minor things while analyzing over the course of an evening, so I generally limit engines under Chessbase 15 to 2 threads, 2-3 Multi-PV.

I do not remember seeing on the Komodo website (have not visited it in a while) anything about how to get the most out of Komodo MCTS for game or opening analysis...do you or anyone else have some tips or could you point me somewhere which would be of assistance?

Thanks!

The main tip is that with MCTS you can now feel free to look at all the moves you like; as far as we can tell, there is no elo loss from setting MultiPV to 2, 3, or even 218 ! With normal Komodo, MultiPV 2 costs something like 100 elo, MultiPV 3 another fifty or so, etc. Also with normal Komodo you have to pay attention to trying to finish iterations; with MCTS it makes sense to quit any time that the two best moves are not nearly equal.

One might argue this is no true MultiPV mode, as long as the PV lines don't have about the same number of visits.

OTOH, this could lead to reconsider the way how MultiPV is done with Alpha-Beta search.
It doesn't seem to make much sense to put/waste the same amount of time/calculations into inferior lines.

Thank you for this new version,
and good luck with the further development.

[Gary Internet]

I don't think that we can continue to gain five elo per month with regular Komodo. But we have been averaging close to fifty elo per month with MCTS, so even if this drops to something reasonable like twenty per month, after a year everyone should be happy!

Do you think you can sustain 20 Elo per month once Komodo MCTS reaches the current strength of regular Komodo?

If so that would make Komodo incredibly strong in very little time. The Stockfish team took basically took 10 months to gain 50 Elo going from SF9 to SF10. That's 5 Elo per month. 20 Elo per month is 4 times the rate of improvement. It would be unheard of. It would be insane.

[S.Taylor]

Yes, we released 12.3 this morning at Komodochess.com, which is free to subscribers and to purchasers of 12.2 since there were still some bugs unfixed by 12.2.2. It is virtually the same version now playing in division one of TCEC, and is an improved version of the one looking likely to qualify for the next stage of CCC. We have added support for seven man Syzygy Tablebases to both regular and MCTS Komodo 12.3. Although the full set of seven man bases is too large for most people to download and store, a majority of the benefit can be had from just the two most common endings, which are easily downloaded and stored on SSD.
Normal Komodo has been improved by just a few elo (aside from 7 man Syzygy) by revised contempt, revised time management, and speedups. But Komodo MCTS has been improved quite a bit since 12.2.2, about thirty elo in blitz games on single thread, from improved time management (especially for repeating time controls) and improved move selection criteria. We have also fixed bugs and found improvements to MCTS MP, so the gains on four threads exceed fifty elo and on machines averaging about 15 threads exceed seventy elo! This is all on top of the enormous gains made between 12.1.1 and 12.2.2 MCTS. Also please note that although 12.3 MCTS is still nearly a hundred elo below normal 12.3, it can already defeat 12.3 if both are in MultiPV = 2 (or more) mode. For this reason, I have switched to using the MCTS version myself when analyzing games or openings.
In blitz matches on four threads Komodo 12.3 MCTS came out ahead against Ethereal 11.00, Stockfish 6, and Komodo 9.2, which average nearly 3400 on CCRL four cpu 40/4 list, while losing by about 15 elo to the higher-rated Fire 7.1, so we expect 12.3 MCTS to top 3400 on CCRL 40/4 and to earn whatever ratings are equivalent on other lists. There are no known bugs in 12.3 so we encourage rating lists to test the MCTS version at least, especially on multiple threads.
Regarding the future, given that Komodo MCTS has been progressing at about ten times the rate of normal Komodo, it does not take a computer scientist to predict that it will become our flagship version in 2019, although this is not guaranteed. Version 12.3 will probably be the last one that combines regular and MCTS Komodo, they will soon become distinct programs, although subscribers will still get both. My personal view is that the future looks very bright for Komodo MCTS, as it is still in its infancy or perhaps early childhood.

This is good and exciting news.
But i hope the elo increase will still be +5 elo per month, for the past 3 years, so that it would be 240 elo more in 2019 than it was in 2015.

Now thay i/we have waited so long, for the monthly 5 elo to add up, i will be very satisfied with that, if it means that 240 is increased since 2015, by 2019.

I don't think that we can continue to gain five elo per month with regular Komodo. But we have been averaging close to fifty elo per month with MCTS, so even if this drops to something reasonable like twenty per month, after a year everyone should be happy!

I would like to see it trounce any Stockfish.

[ChiefPushesWood]

Larry or Mark,
I'm getting an error when trying to download 12.3. I purchased 12.2.2 so I'm eligible. My Account is, indeed, showing me the 12.3 folder. But when I click on download 12.3, I get this: Not Found

The requested URL /play/komodo-12.3-ee4a9c9a.zip was not found on this server.
Apache/2.4.18 (Ubuntu) Server at u.komodochess.com Port 80

Advise please?

CPW

PS.
I sent an email via the website as well.