WCCC 2009 - Limit of 8 cores

[chessfurby]

The ICGA has officially decided to degrade the WCCC to something that i can play in my living room. I wholeheartedly agree with everyone who thinks that WCCC should stay "no limits, no holds barred".

Nuff said.

[playjunior]

What is this "enormous scaling difference" you are talking about? On 8 cores my speedup averages about 6x. The best you can do is 8x. That is not "enormous" so I am not sure what you are referring to.

[\quote]

I mean the difference ranking of Zappa relative to other engines. The single-processor version is good, 2 is better, 4 is II best after Rybka. on 8 cores it won a match. It scales better, and the difference is significant. I argue that the example of Zappa shows that 8 cores is enough to give you significant boost if you have really better parallel search.

No. There are ways to access a cluster from directly contacting industry (which I did to become affiliated with Cray in the old Cray Blitz days) or AMD/Intel. Or one can apply for time on public-funded clusters. Etc.. Where there's a will, and some effort, there's a way...

OK, let's try. I write them a letter: "I am an amateur, I want to participate in computer chess championship. Please, can I use that new 500K server you just announced..." Are you serious?

Because that is the stated purpose of the WCCC. Again, for the Nth time, you are free to hold any sort of tournament you want. Uniform platform. Uniform operating system. Uniform compiler. Common GUI. Common book. No book. Whatever you want. But the WCCC's sole purpose is to encourage development by having "the best of the best" get together to measure up against each other, no holds barred...

Formula1 to me is about the best of the best, despite the unlimited number of regulations and "hardware limitations". But then again, that's a matter of taste.

[Dr.Wael Deeb]

What is this "enormous scaling difference" you are talking about? On 8 cores my speedup averages about 6x. The best you can do is 8x. That is not "enormous" so I am not sure what you are referring to.

[\quote]

I mean the difference ranking of Zappa relative to other engines. The single-processor version is good, 2 is better, 4 is II best after Rybka. on 8 cores it won a match. It scales better, and the difference is significant. I argue that the example of Zappa shows that 8 cores is enough to give you significant boost if you have really better parallel search.

No. There are ways to access a cluster from directly contacting industry (which I did to become affiliated with Cray in the old Cray Blitz days) or AMD/Intel. Or one can apply for time on public-funded clusters. Etc.. Where there's a will, and some effort, there's a way...

OK, let's try. I write them a letter: "I am an amateur, I want to participate in computer chess championship. Please, can I use that new 500K server you just announced..." Are you serious?

Because that is the stated purpose of the WCCC. Again, for the Nth time, you are free to hold any sort of tournament you want. Uniform platform. Uniform operating system. Uniform compiler. Common GUI. Common book. No book. Whatever you want. But the WCCC's sole purpose is to encourage development by having "the best of the best" get together to measure up against each other, no holds barred...

Formula1 to me is about the best of the best, despite the unlimited number of regulations and "hardware limitations". But then again, that's a matter of taste.

Hell true,no doubt whatever

[bob]

What is this "enormous scaling difference" you are talking about? On 8 cores my speedup averages about 6x. The best you can do is 8x. That is not "enormous" so I am not sure what you are referring to.

[\quote]

I mean the difference ranking of Zappa relative to other engines. The single-processor version is good, 2 is better, 4 is II best after Rybka. on 8 cores it won a match. It scales better, and the difference is significant. I argue that the example of Zappa shows that 8 cores is enough to give you significant boost if you have really better parallel search.

No. There are ways to access a cluster from directly contacting industry (which I did to become affiliated with Cray in the old Cray Blitz days) or AMD/Intel. Or one can apply for time on public-funded clusters. Etc.. Where there's a will, and some effort, there's a way...

OK, let's try. I write them a letter: "I am an amateur, I want to participate in computer chess championship. Please, can I use that new 500K server you just announced..." Are you serious?

Yes, Vincent did it. Cozzie did it. How do you think _I_ got started with Cray Research??? So yes, I am dead serious. It helps if you have a decently strong engine for starters, but amateurs have success with good hardware all the time.

Because that is the stated purpose of the WCCC. Again, for the Nth time, you are free to hold any sort of tournament you want. Uniform platform. Uniform operating system. Uniform compiler. Common GUI. Common book. No book. Whatever you want. But the WCCC's sole purpose is to encourage development by having "the best of the best" get together to measure up against each other, no holds barred...

Formula1 to me is about the best of the best, despite the unlimited number of regulations and "hardware limitations". But then again, that's a matter of taste.

Formula 1 is supposed to be about the _driver_. Not so much the car. Here the human has no place in the competition, so it is _all_ about the car itself, which includes hardware and software... We are doing NHRA top fuel... Not pro stock. Some tracks call this "outlaw class"...

[playjunior]

What is this "enormous scaling difference" you are talking about? On 8 cores my speedup averages about 6x. The best you can do is 8x. That is not "enormous" so I am not sure what you are referring to.

[\quote]

I mean the difference ranking of Zappa relative to other engines. The single-processor version is good, 2 is better, 4 is II best after Rybka. on 8 cores it won a match. It scales better, and the difference is significant. I argue that the example of Zappa shows that 8 cores is enough to give you significant boost if you have really better parallel search.

No. There are ways to access a cluster from directly contacting industry (which I did to become affiliated with Cray in the old Cray Blitz days) or AMD/Intel. Or one can apply for time on public-funded clusters. Etc.. Where there's a will, and some effort, there's a way...

OK, let's try. I write them a letter: "I am an amateur, I want to participate in computer chess championship. Please, can I use that new 500K server you just announced..." Are you serious?

Yes, Vincent did it. Cozzie did it. How do you think _I_ got started with Cray Research??? So yes, I am dead serious. It helps if you have a decently strong engine for starters, but amateurs have success with good hardware all the time.

Because that is the stated purpose of the WCCC. Again, for the Nth time, you are free to hold any sort of tournament you want. Uniform platform. Uniform operating system. Uniform compiler. Common GUI. Common book. No book. Whatever you want. But the WCCC's sole purpose is to encourage development by having "the best of the best" get together to measure up against each other, no holds barred...

Formula1 to me is about the best of the best, despite the unlimited number of regulations and "hardware limitations". But then again, that's a matter of taste.

Formula 1 is supposed to be about the _driver_. Not so much the car. Here the human has no place in the competition, so it is _all_ about the car itself, which includes hardware and software... We are doing NHRA top fuel... Not pro stock. Some tracks call this "outlaw class"...

Didn't know that, sorry. Thought WCCC is the Formula1 of computer chess. Turns out it is some random truck race with those flashes, dirt tracks and so on.

And it's strange to hear from Robert Hyatt that humans have no place in engine competition. Who gets the job done then?

On a serious note I would suggest we agree that this is purely a matter of taste. I can see your point, but didn't change my opinion.

[bob]

What is this "enormous scaling difference" you are talking about? On 8 cores my speedup averages about 6x. The best you can do is 8x. That is not "enormous" so I am not sure what you are referring to.

[\quote]

I mean the difference ranking of Zappa relative to other engines. The single-processor version is good, 2 is better, 4 is II best after Rybka. on 8 cores it won a match. It scales better, and the difference is significant. I argue that the example of Zappa shows that 8 cores is enough to give you significant boost if you have really better parallel search.

No. There are ways to access a cluster from directly contacting industry (which I did to become affiliated with Cray in the old Cray Blitz days) or AMD/Intel. Or one can apply for time on public-funded clusters. Etc.. Where there's a will, and some effort, there's a way...

OK, let's try. I write them a letter: "I am an amateur, I want to participate in computer chess championship. Please, can I use that new 500K server you just announced..." Are you serious?

Yes, Vincent did it. Cozzie did it. How do you think _I_ got started with Cray Research??? So yes, I am dead serious. It helps if you have a decently strong engine for starters, but amateurs have success with good hardware all the time.

Because that is the stated purpose of the WCCC. Again, for the Nth time, you are free to hold any sort of tournament you want. Uniform platform. Uniform operating system. Uniform compiler. Common GUI. Common book. No book. Whatever you want. But the WCCC's sole purpose is to encourage development by having "the best of the best" get together to measure up against each other, no holds barred...

Formula1 to me is about the best of the best, despite the unlimited number of regulations and "hardware limitations". But then again, that's a matter of taste.

Formula 1 is supposed to be about the _driver_. Not so much the car. Here the human has no place in the competition, so it is _all_ about the car itself, which includes hardware and software... We are doing NHRA top fuel... Not pro stock. Some tracks call this "outlaw class"...

Didn't know that, sorry. Thought WCCC is the Formula1 of computer chess. Turns out it is some random truck race with those flashes, dirt tracks and so on.

And it's strange to hear from Robert Hyatt that humans have no place in engine competition. Who gets the job done then?

Hopefully the "AI" in the engine.

On a serious note I would suggest we agree that this is purely a matter of taste. I can see your point, but didn't change my opinion.

And there are plenty of supporters. The various rating lists use uniform platforms to play games, for example. But eventually, some of us want to see the Godzillas, king kongs, megalons, rodans, and other "monsters" duke it out to see which is best on that particular day...

[CThinker]

I wonder who would provide all the engine authors who participate a 40-core machine?

This is not about removing incentives for implementing good parallel search, it just gives a chance to amateurs who don't have 40-core clusters or University chair and funding with 500 core machines to compete. There are 1 billion such regulations in sports like Formula1, so that the rich teams cannot just outspend the poorer ones in a way that the drivers won't matter at all. Here, cars translate to hardware and pilots are the software. Well almost.


And, I think that if you have a really good parallel search compared to competitors, 8 cores would give quite some advantage. Look at scaling difference of Zappa vs engines of the same time.

Just because the hardware is the same, it does not automatically make the tournament "fair". Software development is not fair. I don't have 16 hours a day, every day, to spend on chess programming. I don't have the millions of test games that are run on massive clusters. I don't have a GM/IM to tune my code and prepare the opening books. I'm in a 100:1 disadvantage compared to the commercial developers.

This equal hardware competition is totally nonsense. I don't want to travel halfway around the globe just to see what CCRL/CEGT/SSDF already know.

[Dr.Wael Deeb]

I wonder who would provide all the engine authors who participate a 40-core machine?

This is not about removing incentives for implementing good parallel search, it just gives a chance to amateurs who don't have 40-core clusters or University chair and funding with 500 core machines to compete. There are 1 billion such regulations in sports like Formula1, so that the rich teams cannot just outspend the poorer ones in a way that the drivers won't matter at all. Here, cars translate to hardware and pilots are the software. Well almost.


And, I think that if you have a really good parallel search compared to competitors, 8 cores would give quite some advantage. Look at scaling difference of Zappa vs engines of the same time.

Just because the hardware is the same, it does not automatically make the tournament "fair". Software development is not fair. I don't have 16 hours a day, every day, to spend on chess programming. I don't have the millions of test games that are run on massive clusters. I don't have a GM/IM to tune my code and prepare the opening books. I'm in a 100:1 disadvantage compared to the commercial developers.

This equal hardware competition is totally nonsense. I don't want to travel halfway around the globe just to see what CCRL/CEGT/SSDF already know.

Thus you're doing a terrific job Lance