Houdini routs Rybka to start, routs Rybka to end match

[BB+]

I've never understood this whole "no human intervention" rule during tournament games. It seems kinda arbitrary, particularly as human intervention is allowed between games (opening book selection for instance). I mean, what if my engine was designed in such a way to expedite human intervention as much as possible (via output from "sampled search" or something)? Furthermore, there's reduced incentive to design and implement such centaur-esque capabilities if the tournaments don't allow them.

[Martin Thoresen]

The goal of TCEC is to wave Houdini in Vas's face. I grok that motive. I don't need your help which is not to say you all don't need some.

Thanks for telling me what the goal of my own project is. It's appreciated.


Best,
Martin

[orgfert]

I think your posts doesn't make sense at all. Dumbed-down Rybka? Your analogy is the same as saying that a car with 300 hp which is driven legally at 100 km/h on the highway is dumbed down because it can theoretically run at 250 km/h.

I suppose a better analogy would be putting the same motor in two different automobile designs and then racing them. It makes no sense. The goal is the best chess possible with an intended design. For a match to mean something, the programmers should put forward their best setup. This is probably why interest was minimal in a uniform platform world championship. After all, why bother to win a crippleware world title? Ergo, why bother to take interest in yet another crippleware chess match? Such matches are done ad infinitum in CCRL and other places, and the resulting Elos touted as illumination. Dingoes kidneys.

Perhaps you should keep looking for the proper analogy. Try putting two drivers in identical cars and then let them
race. Obviously, that is nowhere near as exciting to you. Yet, it may illuminate the drivers' ability better than
having them race in their own cars. In the latter scenario, there is the question of what was most important in
winning the race. Was it the car or the driver?

Perhaps you should keep looking for the proper analogy. Try limiting different car designs to the same horsepower of motor (like you limit chess AIs to the same "horsepower" of CPU).

Obviously, I am in the minority as far as the purpose of a rating list. I keep seeing rating lists and
competitions being lumped together. Perhaps the term rating list should be abandoned and everybody
start using the term ranking list. That would be more consistent with most people's idea of what the lists
are. Pity that some of the people involved with the lists don't see it that way. Of course, if the true purpose of
a list is to help show the level an engine has obtained rather than to show which engine is better than another
engines, then the current terminology is appropriate.

You are interested in the efficiency of the search and evaluation algorithm in isolation -- IQ/MIPS -- rather than performance levels in the intended application. This dispenses with the design perspective where compromises may have been made with respect to IQ/MIPS to obtain other advantages, such as perhaps ease of maintenance, portability, scalability, etc, where chess IQ is made up in other aspects of the design that your tests deliberately (or accidentally) ignore. The result can only be misleading -- possibly egregiously so. It cannot possibly "show the level that an engine has obtained", simply because your are looking at IQ/MIPS out of context -- the context of holistic design.

[HalloweenGambit]

While I'm a chess enthusiast, my level of knowledge about hardware or software is simply not on par with the expert personalities that have posted here - for that I apologize in advance.

I happened upon this site, however, because of my keen interest in this match - made known through my loyal following of the esteemed Kingscrusher on YouTube, who is currently posting his analysis of some key Houdini/Rybka games.

While I may be lacking in relative depth of I.T. knowlege, my skill set lies in technical writing and editing - in the logical presentation and substantiation of concepts and ideas. Stripping away the emotionalism and chest thumping resident in some of the posts in this thread - and granting that both sides of this debate have points of merit - it is my humble opinion that Hyatt and Orgfert make the best case for their assertions.

I am not qualified to judge which of these programs is stronger, nor do I dispute there's a rather powerful set of "bottom-line" / "at the end of the day" data that the Houdini apologists rest their cases on. And yet, there's no denying that a whole host of variables combines to determine performance of any software application. Hyatt and Orgfert's basic premise - that the relative value of any given variable to the performance of one application is not necessarily shared by another - is rooted in the very foundation of scientific method. Only through isolation of cause/effect dynamics to a single variable can one yield results from which concrete conclusions may be drawn. In this case, it would seem that the endeavor to make all else equal - thereby reducing the supposed variability in the Versus equation to the (seductive) simplicity of Win/Loss - in reality undermines the very equality being sought after.

The opportunity for analogies seems endless. I'd suggest that the skill of a knight in a joust is highly dependent on his steed, their level of experience working together, their symbiosis. If two knights on their favorite mounts were to square off, one or the other might win. Take them both off and put them on identical horses with which neither had any prior knowledge, the outcome might be entirely different. Leave one on his trusted horse and give the other an unfamiliar one, the outcome might be different still. Give the unknown horse to the other knight, another result entirely. Which of these scenarios speaks best to the essential strength of each jouster?

I'd submit that each knight on his favorite horse is the best indicator. For in a jousting match, the knight and horse act as one unit. Manipulating intrinsic elements of that unit in an effort to "even the playing field" and ostensibly isolate the knight's skill as the only remaining variable, in fact does no such thing. The knight's skill is largely dependent upon the interplay derived from that variable system. Such is the case with Houdini and Rybka.

I agree with Hyatt and Ogfert. Play on!

[T-Bone]

Just thought I'd ask an absurd question:

When Rybka was trumpeting its superiority on its own website and forum about how it was in the lead on the CCRL, CEGT, and SSDF websites a year ago, was that wrong, since by and large these sites test different engines on the same hardware?

I appreciate what Martin has done. He is an enthusiast as am I. I think the much more interesting thing is how the chess world is seeing some very interesting chess as a result. Some the games in the Elite match were quite good. I also think his testing is helping to break the political stranglehold that Rybka has, and is allowing more competition which can only be good for computer chess in the long run.

[orgfert]

Just thought I'd ask an absurd question:

When Rybka was trumpeting its superiority on its own website and forum about how it was in the lead on the CCRL, CEGT, and SSDF websites a year ago, was that wrong, since by and large these sites test different engines on the same hardware?

Yes.

I appreciate what Martin has done. He is an enthusiast as am I. I think the much more interesting thing is how the chess world is seeing some very interesting chess as a result. Some the games in the Elite match were quite good. I also think his testing is helping to break the political stranglehold that Rybka has, and is allowing more competition which can only be good for computer chess in the long run.

But that is politics, not fair competition.