Rybka 4 just around the corner it seems

[Gian-Carlo Pascutto]

What's the gain in running 4 unrelated instances?

If the instances are related in any way (such as being part of a bigger search tree, as in IDeA), there will be SMP loss, because you won't analyze the tree 4 times faster.

Wait you got it wrong. IDEA is not a chess search engine. Nothing close to that. It is just a chess crawler. It will use the engine as a compass to build a tree of positions.

You already contradicted yourself there. What do you think a "chess search engine" is except for something that builds a tree of positions?

IDeA is just a search algorithm and you could implement it in an engine. IDeA + Rybka is in fact an engine, and people are playing games and making books with it.

[Gian-Carlo Pascutto]

Either I don't understand you or you don't understand how IDeA works. It's true that IDeA decides about the next positions to analyze based on the results of previous analysis. However, IDeA only does that at the end of each stage when it finds the next batch of positions to analyze.

In this case, the scalability would have a hard limit on the number of batched up positions and the efficiency of the method you describe is way lower than that of a search algorithm that can switch away from poor branches as soon as one of the analysis threads finds it.

So you confirmed my point exactly: either IDeA is a very poor algorithm, and it scales well (whereby you should take into account, that scaling well starting from something crappy is useless), or it is a good algorithm, and it scales not perfectly.

Nobody has managed to make a good tree search algorithm scale perfectly, and claiming IDeA does it is preposterous.

If IDeA could play games, you could play a match with Rybka parallelism (IDeA + Rybka 4 core) versus IDeA parallelism (IDeA + 4 x Rybka 1 core) and check the results.
I would bet that Vasik has got his parallelism working more efficiently than the Convekta guys.

So my claim is specifically: if Convekta claims IDeA scales perfectly, this means that either IDeA is crap or they're wrong. There's decades of research in parallelism to support that conclusion. TANSTAAFL

[Gian-Carlo Pascutto]

I am shocked from the users blindness GCP,that they can't accept the fact that Vasik is trying to squeeze every single $ from their pockets by any means necessary

The claim about IDeA is made by Convekta. I'm sure Vasik isn't making it, because it would be equivalent to saying that you shouldn't buy Deep Rybka, because IDeA parallelizes better anyway.

But it's just not true.

_The online "rent" issue is an ugly as it can be....if you are so despered to use your promised Rybka 3+ or the current version of Rybka 4,come on,buy our Aquarium 2010 for 124 $ and prepare your money for the rent

Renting seems like a good solution to having the program hacked and rereleased for free as Ippolit. Very understandable decision by Vasik in order to safeguard his livelihood. And more evidence to users that pirating engines just will come back to hurt them in the long run.

[Leto]

Either I don't understand you or you don't understand how IDeA works. It's true that IDeA decides about the next positions to analyze based on the results of previous analysis. However, IDeA only does that at the end of each stage when it finds the next batch of positions to analyze.

In this case, the scalability would have a hard limit on the number of batched up positions and the efficiency of the method you describe is way lower than that of a search algorithm that can switch away from poor branches as soon as one of the analysis threads finds it.

So you confirmed my point exactly: either IDeA is a very poor algorithm, and it scales well (whereby you should take into account, that scaling well starting from something crappy is useless), or it is a good algorithm, and it scales not perfectly.

Nobody has managed to make a good tree search algorithm scale perfectly, and claiming IDeA does it is preposterous.

If IDeA could play games, you could play a match with Rybka parallelism (IDeA + Rybka 4 core) versus IDeA parallelism (IDeA + 4 x Rybka 1 core) and check the results.
I would bet that Vasik has got his parallelism working more efficiently than the Convekta guys.

So my claim is specifically: if Convekta claims IDeA scales perfectly, this means that either IDeA is crap or they're wrong. There's decades of research in parallelism to support that conclusion. TANSTAAFL

If Dadi is correct about IDeA analysing positions more quickly in an hour with 4 instances of Rybka then there's no need to do a match.

[Gian-Carlo Pascutto]

If Dadi is correct about IDeA analysing positions more quickly in an hour with 4 instances of Rybka then there's no need to do a match.

Your statement doesn't have all parameters so it's meaningless. You can put Rybka to 1 ply and it will also analyze quickly. But so what? The quality (=strength) of the analysis is what matters.

What is better?

a) Rybka analyzing on 4 cores each position for x seconds
b) (Rybka analyzing on 1 core each position for 4x seconds) * 4 instances
c) (Rybka analyzing on 1 core each position for 3x seconds) * 4 instances

Convekta claims c == a. I claim that is false, and think that probably not even b == a.

[oreopoulos]

What's the gain in running 4 unrelated instances?

If the instances are related in any way (such as being part of a bigger search tree, as in IDeA), there will be SMP loss, because you won't analyze the tree 4 times faster.

Wait you got it wrong. IDEA is not a chess search engine. Nothing close to that. It is just a chess crawler. It will use the engine as a compass to build a tree of positions.

LOL

So from the answer, you choose to quote something irrelevant!

Ok. i will do it with a question.

The task is. Analyze 4 positions till depth 17. You can do it with 1 4core engine, or with 4 1-core engines. You are a chess engine USER.

Which is faster?

[oreopoulos]

If Dadi is correct about IDeA analysing positions more quickly in an hour with 4 instances of Rybka then there's no need to do a match.

Your statement doesn't have all parameters so it's meaningless. You can put Rybka to 1 ply and it will also analyze quickly. But so what? The quality (=strength) of the analysis is what matters.

What is better?

a) Rybka analyzing on 4 cores each position for x seconds
b) (Rybka analyzing on 1 core each position for 4x seconds) * 4 instances
c) (Rybka analyzing on 1 core each position for 3x seconds) * 4 instances

Convekta claims c == a. I claim that is false, and think that probably not even b == a.

You are wrong. This is NOT the claim

Of course

Convekta claims c == a. I claim that is false, and think that probably not even b == a.

is FALSE

But this is not the claim. The claim is

a) Rybka analyzing 100 positions on 4 cores each position till depth 17 takes X seconds
b) Rybka analyzing 100 positions on 4x 1-cores (splitting work in 4) each position till depth 17 takes less than X seconds

and ofcourse as the number of cores increase the gap gets bigger. Much bigger.

Rybka is used as an evaluator. You are creating a tree, with fixed depth (or near constant quality nodes)

[Dr.Wael Deeb]

If Dadi is correct about IDeA analysing positions more quickly in an hour with 4 instances of Rybka then there's no need to do a match.

Your statement doesn't have all parameters so it's meaningless. You can put Rybka to 1 ply and it will also analyze quickly. But so what? The quality (=strength) of the analysis is what matters.

What is better?

a) Rybka analyzing on 4 cores each position for x seconds
b) (Rybka analyzing on 1 core each position for 4x seconds) * 4 instances
c) (Rybka analyzing on 1 core each position for 3x seconds) * 4 instances

Convekta claims c == a. I claim that is false, and think that probably not even b == a.

Clear now,thanks and I do agree....
Dr.D

[Gian-Carlo Pascutto]

Ok. i will do it with a question.

The task is. Analyze 4 positions till depth 17. You can do it with 1 4core engine, or with 4 1-core engines. You are a chess engine USER.

Which is faster?

If the positions are unrelated, you don't need IDeA to begin with, and can just run separate engines.

So rather hard to argue that IDeA gives you a free CPU. Better save your money and don't buy it.

People who use IDeA analyze related positions, and then my argument holds and yours is not relevant.

[Gian-Carlo Pascutto]

But this is not the claim. The claim is

a) Rybka analyzing 100 positions on 4 cores each position till depth 17 takes X seconds
b) Rybka analyzing 100 positions on 4x 1-cores (splitting work in 4) each position till depth 17 takes less than X seconds

You do not know the positions in advance. That's the point of IDeA, it builds the tree dynamically and picks the positions from that. It does much more than just analyze a list of positions, it produces new one based on the analysis already done.

Because of this, for what you claim you don't need IDeA, and obviously the people who want to sell IDeA don't make that claim in their marketing, because it hasn't got anything to do with IDeA in the first place.