Leveling The Playing Feild

[George Tsavdaris]

Hydra v Deep Blue is of interest and would make a great match. However with my media hat on trying to explain all the differing hardware, at the World Championship, to a general audience is a problem. I almost got daily coverage of the tournament in Beijing on National Radio in the UK but when trying to explain to the editor that its not about just finding the best software all interest was lost.

I'm not sure i get you here. Do you mean that people lose interest to the tournament when they find out that it is not about finding who is the best via the tournament's winner?
So people believed WCCC's winner is the best program, WCCC's 2nd placed is the second best program, etc?
So what if they believed that? It's total nonsense so they should not believe that.

Is this a reason to change the tournament to something some ignorant people want? I don't think so.
For so many years conditions were different and interest was there from Chess fans. Who cares about ignorant general non-Chess fans?

With a limit of 8 cores i believe it will be Chess fans that will lose interest because it would just be just another tournament and a small one also. For that we have SSDF(own books), CEGT, CCRL, Sedat's tournament(own books) and some more.
In WCCC we want to see some diversity and this would come from having super hardware to see some great moves and then compare them with our hardware etc.

From a programmer's perspective also it's a clear violation of his freedom and to his ambitions to develop new algorithms and test and test to succeed in increasing more and more the speed efficiency for more and more cores.

While some programmers may feel that 1 or 2 cores vs 40 cores creates a pointless match, it's true, but it seems most programmers don't share this opinion.

[Nid Hogge]

Oh wow, this is just retarded.

I'm sure all those Hiarcs whiners are swimming in joy right now, they got they're wish. apparently it's good to have connections in the right places.
I didn't realize the ICGA would bend over so quickly though. two united on a desprate crusade.

But then again..... Precisely how embarrassing is it going to be for them, next year, when Rybka liquidates them, and everyone else once again, this time using equal hardware? This is going to be overly amusing to watch.

Oh well, I'm sure they'll have something else to blame next time.

[George Tsavdaris]

That is abundantly true. It is an ant-Rybka move quite clearly.

He certainly has been suggesting an upper hardware limit on the forums, and I agree with him.

Do you agree that we have to handicap Rybka for not winning?

[George Tsavdaris]

I'm sure all those Hiarcs whiners are swimming in joy right now, they got they're wish. apparently it's good to have connections in the right places.
I didn't realize the ICGA would bend over so quickly though. two united on a desprate crusade.

I guess Hiarcs team wanted that change and i guess Rybka team didn't want it, but to say that Hiarcs team is responsible even in an indirect way for that change is not correct without proof.
Perhaps other teams like Naum, Junior, Fritz etc put pressure in ICGA to make this change.

But then again..... Precisely how embarrassing is it going to be for them, next year, when Rybka liquidates them, and everyone else once again, this time using equal hardware? This is going to be overly amusing to watch.

Lukas(the owner of the 40-core cluster) said that the 40-core cluster is about 100 ELO stronger(i think in Rybka-Rybka match) from a Skulltrail 8-core machine.
So while Rybka on a 8-core machine will still be the favourite to win, even in a such low number of games tournament, perhaps the 100 ELO loss, will cost it the tournament.

[AdminX]

I'm sure all those Hiarcs whiners are swimming in joy right now, they got they're wish. apparently it's good to have connections in the right places.
I didn't realize the ICGA would bend over so quickly though. two united on a desprate crusade.

I guess Hiarcs team wanted that change and i guess Rybka team didn't want it, but to say that Hiarcs team is responsible even in an indirect way for that change is not correct without proof.
Perhaps other teams like Naum, Junior, Fritz etc put pressure in ICGA to make this change.

But then again..... Precisely how embarrassing is it going to be for them, next year, when Rybka liquidates them, and everyone else once again, this time using equal hardware? This is going to be overly amusing to watch.

Lukas(the owner of the 40-core cluster) said that the 40-core cluster is about 100 ELO stronger(i think in Rybka-Rybka match) from a Skulltrail 8-core machine.
So while Rybka on a 8-core machine will still be the favourite to win, even in a such low number of games tournament, perhaps the 100 ELO loss, will cost it the tournament.

Good Points George ....

[Nid Hogge]

I guess Hiarcs team wanted that change and i guess Rybka team didn't want it, but to say that Hiarcs team is responsible even in an indirect way for that change is not correct without proof.
Perhaps other teams like Naum, Junior, Fritz etc put pressure in ICGA to make this change.

Than I guess that means they're all crybabies then.

Lukas(the owner of the 40-core cluster) said that the 40-core cluster is about 100 ELO stronger(i think in Rybka-Rybka match) from a Skulltrail 8-core machine.
So while Rybka on a 8-core machine will still be the favourite to win, even in a such low number of games tournament, perhaps the 100 ELO loss, will cost it the tournament.

And Bob says it's 0. So? It doesn't matter, the whole purpose is for them handicap Rybka(or any other program out there that is going beat them silly and make the WCCC completely irrelevant) in any possible way, so when they do win the tourney, they'll have something big and shiny to stick to they're product boxes and websites. Just like the overly lying messege on hiarcs.com website. "HIARCS wins only major tournament of 2008 with ALL top chess software competing.." Yes.. Right!

These guys are about selling, more more more and more. It's not about "evening the playing field for real competition" or c*ap like that. it's nothing but good ole' marketing.

Get real..

[bob]

Oh wow, this is just retarded.

I'm sure all those Hiarcs whiners are swimming in joy right now, they got they're wish. apparently it's good to have connections in the right places.
I didn't realize the ICGA would bend over so quickly though. two united on a desprate crusade.

But then again..... Precisely how embarrassing is it going to be for them, next year, when Rybka liquidates them, and everyone else once again, this time using equal hardware? This is going to be overly amusing to watch.

Oh well, I'm sure they'll have something else to blame next time.

Let me state this again:

(1) the concept is bad, and nonsensical as well. The very idea of the original ICCA (now ICGA) was to foster research in computer chess. And parallel/distributed algorithms are an important part of this. Now this will be curtailed, which makes no sense.

(2) if this was directed at Rybka, it is doubly stupid, because the "cluster Rybka" was no faster or stronger than the normal SMP Rybka. So it won't have any effect on Rybka's performance level, but even worse, this directly discourages distributed search development.

The way for everyone to voice their disproval is to simply no longer enter the ICGA WCCC and move on to the CCT/ACCA events where such rules are not being discussed, much less cast in stone with no discussion...

I guess a "well-done" should be sent to these idiots...

[bob]

It seems that 8 cores would be a good number for this, at the current level of technology: most CPUs come as quads nowadays, and soon they will come as octals.

Oh, I found the e-mail from ICGA in my inbox, and I see this is exactly what they proposed as well, for the 2009 event.

And there will be separate classes, of some sort: The games olympiad will have a Chess tournement as well, where there is no hardware limitation. So programs that feel the need to surpass the 8-core limitation can compete there.

I think this is a very sensible solution.

No it isn't. We had this in years past, where we had two divisions. It was uniformly disliked by everyone and was discontinued several years ago as a result.

[glorfindel]

They are not proposing a level playing field just an upper limit. Opening books also will still be used.

It would be interesting to know what is your suggestion about this "upper limit". You have written in the past in a related discussion about the 2008 tournament here that

I used to believe that = hardware was the way to go, now I am not so sure. However I think an upper limit should be set. Hiarcs played on a Skulltrail and was only the 5th fastest hardware in the tournament. However it is hardware that anyone can buy. So perhaps the upper limit this year should have been an 8 core Skulltrail. Next year the upper limit could be whatever is easily commercially available then.

So for last year, the upper limit should have been Hiarcs's hardware. Perhaps next year it should also be the hardware Hiarcs will have. But this makes me ask myself some questions.
-How can somebody seriously trust that your intention is for the WCCC to be more interesting to the masses?
-The fact that Hiarcs will have the strongest hardware is to be considered a simple coincidence, and not a way to gain an advantage over other programs, that perhaps are able to use stronger hardware more efficiently?

[bob]

I'm sure all those Hiarcs whiners are swimming in joy right now, they got they're wish. apparently it's good to have connections in the right places.
I didn't realize the ICGA would bend over so quickly though. two united on a desprate crusade.

I guess Hiarcs team wanted that change and i guess Rybka team didn't want it, but to say that Hiarcs team is responsible even in an indirect way for that change is not correct without proof.
Perhaps other teams like Naum, Junior, Fritz etc put pressure in ICGA to make this change.

But then again..... Precisely how embarrassing is it going to be for them, next year, when Rybka liquidates them, and everyone else once again, this time using equal hardware? This is going to be overly amusing to watch.

Lukas(the owner of the 40-core cluster) said that the 40-core cluster is about 100 ELO stronger(i think in Rybka-Rybka match) from a Skulltrail 8-core machine.
So while Rybka on a 8-core machine will still be the favourite to win, even in a such low number of games tournament, perhaps the 100 ELO loss, will cost it the tournament.

First, the 100 Elo claim is nonsense. Second, Rybka will still be the favorite since no one will have faster hardware to try to equalize the gap through technology. IMHO, the ones wanting this restriction are basically saying "I am not intelligent enough to develop a parallel/distributed search that works, and since I can't do it, I don't want anyone else to be able to use their fancy stuff that I don't know how to develop to be able to compete with them..."