Komodo 9.0

[APassionForCriminalJustic]

I do expect that these two will be the dominant engines unless the Gull author finds time to improve it.

He seems to have dropped steam on the project for some reason.

Gull or Fire would first need to deal with catching Houdini before even thinking about Stockfish and Komodo.

[shrapnel]

One thing most people are forgetting.
Any advance/discovery/idea Stockfish team makes/discovers is available to all. BUT any advance/idea of Komodo Team ( or Robert Houdart for that matter) is NOT available to all !
If by chance, said idea/advance advances the Engine's ELO by a large margin, Stockfish will be left floundering in the shallows !
Remote possibility, I know, but the possibility is always there !

[Dann Corbit]

Another thing to think about, due to the open source nature of Stockfish:
All of Stockfish's innovations are instantly available for public scrutiny as soon as they have proven to have benefit.
None of Komodo's innovations are ever available for public scrutiny.

Hence, if the Stockfish team comes up with something brilliant, everyone can do the same thing in a week or so.

If the Komodo team comes up with something brilliant, nobody can do the same thing unless the Komodo teams explains their idea publicly.

A clear competitive advantage for Komodo.

[Dann Corbit]

I do expect that these two will be the dominant engines unless the Gull author finds time to improve it.

He seems to have dropped steam on the project for some reason.

Gull or Fire would first need to deal with catching Houdini before even thinking about Stockfish and Komodo.

Gull is lean an mean. Look at the code size. One person has made big improvement, when he did put the effort in.

Houdini does not seem to get any attention lately either.

Fire is interesting, but I think that Gull could go farther, since it has all of the important ideas from Ivanhoe like Fire does, but none of the cruft.

[TShackel]

So Komodo 9 still the strongest at very LTC. It's nice, can't wait for TCEC.

Stockfish is a moving target.
It will be stronger tomorrow that it is today.
And stronger the day after that.
A commercial software really can't have daily releases.

So the Komodo team will have a difficult time to keep up.

However, I must admit that they have performed admirably so far.

Well Komodo is also a moving target. Do you remember the subscription service that Larry and Mark offered? They will publish regular updates free for purchasers of komodo 9.0 and komodo 10.0 (ie you buy both komodo 9.0 and 10.0 at a discounted price and then updates in between are free). I personally don't want an update until it's 10-20 elo improvement. Because to be more consistent it would be good to stick with Komodo 9.0.

Stockfish is a fantastic engine though, but in my LTC testing Komodo 9.0 is leading 5 wins to 2 with many draws. Komodo just ruthlessly proved stockfish's zero score wrong several times proving that there were wins in the position. So Komodo 9.0 definitely adds something to the computer chess world that stockfish doesn't offer. Komodo's style of play has definitely gotten more attractive than 8.0 as well. I've seen it sacrificexchanges, pieces, and pawns at different times because it had a lead in development and stockfish's king was in the center.

To give you an idea about how excited I am about Komodo 9.0, is that Komodo 8.0 never beat stockfish on my machine at LTC, and Stockfish never beat Komdo 8.0 at LTC, but with Komodo 9.0 there are already many decisive games and they were spectacular games.

I always go through my "piece shuffle test", and I like to play over many komodo games and see if it ever shuffles it's pieces with no plan. There have been a time or two when it has made some dumb moves movingn the pieces back and forth, and stockfish also has times like that where it doesn't know what to do in the position, but these two engines do it a lot less than other engines in my opinion.

Sincerely,

Tim.

[TShackel]

So Komodo 9 still the strongest at very LTC. It's nice, can't wait for TCEC.

For me it's stronger than the April 12th development version of stockfish at LTC through 92 long time control games. In my opinion you need fewer LTC games to get a good rating picture than STC games. My goal is to get to 200-300 games before I consider the ratings real. We'll see if I get to 1000+ someday, but that's not quite as realistic for LTC games.

Sincerely,

Tim.

[Dr.Wael Deeb]

So Komodo 9 still the strongest at very LTC. It's nice, can't wait for TCEC.

Stockfish is a moving target.
It will be stronger tomorrow that it is today.
And stronger the day after that.
A commercial software really can't have daily releases.

So the Komodo team will have a difficult time to keep up.

However, I must admit that they have performed admirably so far.

Don't forget that they can always provide beta versions of their current development chess engine....
Dr.D