|Jun-10-10|| ||Jim Bartle: With this name, I'm surprised it doesn't play the dragon against 1. e4.|
|Jul-08-10|| ||Albertan: This is an amazing engine which you can obtain for free by going to:
where you will find a link to a page you can download the program from for free.In addition at the same link you will find a link to an opening book which
is referred to as the "official opening book" for Komodo.
Why do I say that the new Komodo 1.2 JA
program is amazing? Well as I type this this program is playing Deep Hiarcs 13 on my quad core computer and in the first game of a regulation time control match (40 moves/120 minutes+2o moves/hr+ g/30.
match between these two engines Komodo has an advantage over Deep Hiarcs 13, (despite the fact that Deep Hiarcs 13 is using four cores of my computer to one core being used by Komodo!)
|Dec-08-11|| ||invas0rX: I am from Dominican Repablic, I love Komoto|
|May-05-13|| ||AylerKupp: After what seemed to me an inordinate amount of time, I received an email that MP Komodo 5.1 will be available on May-28-13, free to those who had previously purchased Komodo 4.0 and Komodo 5.0.|
Given Komodo's good results when playing other multi-core engines even though it has only been able to use a single core, I'm looking forward to see how the multi-core Komodo 5.1 does in computer vs. computer competition.
|May-05-13|| ||cro777: A longtime researcher in computer chess, GM Larry Kaufman is a co-creator (with programmer Don Dailey) of Komodo. He was also responsible for the evaluation function of Rybka 3. |
According to Kaufman, while most programs are optimized to beat other programs, he and Dailey made an effort with Komodo not to sacrifice reasonable evaluation in pursuit of rating points.
In 2008 Kaufman surprised almost everyone by winning the World Senior Championship, thus becoming a Grandmaster.
|May-12-13|| ||AylerKupp: According to the Komodo website (http://komodochess.com/komodoCCT-31a) there is an enhanced single-core version of Komodo 5.0 available, Komodo CCT, which the Komodo developers estimate is about 30 Elo points stronger than Komodo 5.0.|
And Komodo CCT is apparently the end of the line for the Komodo single-core engine versions, written in C. Komodo 5.1 is apparently a rewrite written in C++. So that's a good reason why its release took what I thought was an inordinate amount of time. The new code base should be more easily maintained and enhanced than the previous versions so I applaud the Komodo developers for biting the bullet and making an investment in the future. It's not easy managing concurrent development of multiple versions of the same program. As a former software manager, believe me, I know.
Given the extensive rewrite of the multi-core version an of course it's multi-threading capability, I'm puzzled by their decision to label it Komodo 5.1. More usual would have been to label it Komodo 6.0 and label the current Komodo CCT release as Komodo 5.1. I suppose that was a marketing decision and that they had good reasons to label them as they have. But Komodo 5.1 MP has not yet been released so maybe they will reconsider.
One word of advise to all. Given that Komodo 5.1 MP is a major rewrite, <expect> some bugs. You can test multi-threaded software as best you can but there <will> be errors upon its release. Again, believe me, I know. So keep the faith and support the developers (both verbally and financially) and this will soon be an outstanding product, if previous versions are any indication.
However, being a cautious person, I'm now really looking forward to Komodo 5.2! :-)
|May-13-13|| ||pbercker: @ <AylerKupp> I don't know if this relevant, but Komodo 4.5 came in 4rth place at the TCEC tournament that completed just a few weeks ago. Houdini 3 took first, followed closely by Stockfish 2.5, a free open source engine. |
Stockfish is multi-core, and Stockfish 3 just came out last week.
|May-13-13|| ||AylerKupp: <pbercker> Of course it's relevant. Komodo has been doing very well in the CCRL tournaments ever since at least Feb-04-11, when Komodo 1.2 was ranked 10th even though it was, and still is, only a single-core version. And if you look at the single-core version rankings, Komodo is invariably ranked #2, behind Houdini. So I would suspect that Komodo 5.1, when released, will rank #2, again behind Houdini 3. And I suspect that Komodo 4534 might be the same as the Komodo CCT that they advertise in the Komodo site.|
Another reason to follow CCRL, CEGT, and other tournaments is that this is a way to find out about new engines and old engine updates. So, yes, I downloaded Stockfish 3.0 about a week ago but I haven't tried it yet.
And, being an engine junkie, I collect various engines and whenever possible run the analyses of a position using multiple engines, since engines very often give different results and certainly different evaluations, plus some engines may find good moves that other engines miss.
As far as the nTEC tournament I have not followed it as much as I intended to. So thanks for the results link. No big surprises here. Houdini, Stockfish, Rybka, Komodo, and Critter as the top 5. And I was aware of both Vitruvius and Equinox, two engines we seldom hear about. Unfortunately Vitruvius is not free and Equinox is private. Too bad that the tournament didn't include Bouquet, a fairly strong engine released relatively recently. It was ranked #5 in the latest (Apr-27-13) CCRL 40/40 tournament.
|May-14-13|| ||pbercker: I only followed the final slugfest between Houdini and Stockfish, and then only some of the 48 games, most of which were draws. Many were somewhat boring games, but there were some exiciting ones including one that went on to almost 200 moves I think. |
What was particularly interesting is the chat window since many of the engine programmers, including those for H3 and Stockfish were there chatting, and answering questions from others.
I've been occasionally using Stockfish on some of the Norwegian games (Carlsen vs. Anand, Carlsen vs. Karjakin, Anand vs. Topalov ...) to compare to what Houdini was evaluating. There were some differences, but on the whole there were mostly similarities especially on crucial moves.
I'm actually quite new to all this chess engine business and very much still feeling my way around. I'm using Arena 3 which seems fairly functional.
What I still can't figure out is how to limit the output of the analysis on the "automatic analysis" mode. I'm interested only in the final best line analysis, and not with what is computed for every single level, but I haven't figured that out yet.
Is there anything better than Arena 3 in your opinion?
|May-14-13|| ||AylerKupp: <pbercker> I use Arena myself but I certainly wouldn't call myself an expert on it. When I first started using chess engines I selected Arena because it was the most comprehensive of the free engines (I'm cheap). I have Rybka 4 so I also have the Rybka/Fritz interface, but I haven't used it much. Arena serves my needs reasonably well and I'm used to it by now so I haven't bothered to learn the others.|
As far as limiting the output of the analysis on the "automatic analysis" mode, well, I've never used it and I don't really know what it does. I looked at my set up and some time ago (2 years?) I specified an explicit protocol file but it doesn't seem to exist on my hard drive. So maybe I found the way to <really> limit its output!