Rybka Rental program is a nonsense.

[F. Bluemers]

Any wealthy chess enthusiast would rather invest $30000 in a Cray Cx1
(Windows HPC Server 2008) with Houdini, rather than renting R4 cluster
for an equivalent amount. By the way, CX1 is only the first step in Cray' s
plan to take advantage from Windows hegemony.

What will the R4 cluster be worth when Cray 256-cores Windows power computers will be available ? Can amateur clustering compare with supercomputing experience ? Harsh times ahead for Rybka connection.

Qapla, Robert !

Your "wealthy chess enthusiast" would do both.

[bob]

SedatChess gives a node rate of 18205 kN/s for 200 CPUs:

http://sedatchess.110mb.com/index.php?p=1_72

Does that mean that the speedup per doubling the number of CPUs remains constant for Cluster-R4, even with that many CPUs? I was trying to make a rough estimation in comparison to the fastest i7 in the list, and the factor would be ~1.78 if I got the math right. Which is not sure.

That's not what he asked for. NPS is not important here. The _SPEEDUP_ is the critical number.

that 18M nps is garbage anyway, so who would use that to measure anything???

[Tom Barrister]

Mr. Rajlich has obfuscated (if that's the right word) his nps count for regular Rybka in the past, so how can we assume it's correct now?

I don't know how much Mr. Rajlich wants per-hour for the use of a cluster, but I can't imagine it being worthwhile even to a grandmaster. I suppose that he has a market, though; otherwise he wouldn't have spent the time.

[Dr.Wael Deeb]

The speedup of a parallel program is unfortunately is not the same as the NPS. Unless that's accounted for in the NPS number, in which case you still have to take Vasik's word for it. And then you *still* only have a raw number for 200 CPUs, which doesn't tell you much about how much cluster nodes that actually is. (25 nodes x 8 CPU/node?)

My impression too....
Dr.D

[K I Hyams]

I don't know how much Mr. Rajlich wants per-hour for the use of a cluster, but I can't imagine it being worthwhile even to a grandmaster. I suppose that he has a market, though; otherwise he wouldn't have spent the time.

http://rybkachess.com/Cluster/rent.html
There is a disclaimer at the bottom.

[bob]

Mr. Rajlich has obfuscated (if that's the right word) his nps count for regular Rybka in the past, so how can we assume it's correct now?

I don't know how much Mr. Rajlich wants per-hour for the use of a cluster, but I can't imagine it being worthwhile even to a grandmaster. I suppose that he has a market, though; otherwise he wouldn't have spent the time.

This is about hyperbole and marketing. Not about logic and understanding. Lots of GM players are completely computer-ignorant when talking about this kind of issue. Faster is better, and if the claim is that this is faster than anything else around, they will likely believe it...

[Mike S.]

I understand, and I would prefer e.g. a "time to depth x"(*) comparison. But I didn't find any other performance figures from the cluster yet, than these nps which are from a Rybka 4 benchmark page. I think that page was initially intended to compare "normal" comps running retail Rybka 4.

*) I guess time to depth x is not an optimal comparison either, because if I get things right the search tree on a bigger number of cpus will be bigger, for the same depths? - Maybe solving times for some good test positions are more relevant, in terms of "effective performance" so to speak.

Anyway we don't have such figures and I doubt that we will get them. Although, anyone willing to pay for it could rent Rybe cluster time just for that, too.

[Dr.Wael Deeb]

Mr. Rajlich has obfuscated (if that's the right word) his nps count for regular Rybka in the past, so how can we assume it's correct now?

I don't know how much Mr. Rajlich wants per-hour for the use of a cluster, but I can't imagine it being worthwhile even to a grandmaster. I suppose that he has a market, though; otherwise he wouldn't have spent the time.

This is about hyperbole and marketing. Not about logic and understanding. Lots of GM players are completely computer-ignorant when talking about this kind of issue. Faster is better, and if the claim is that this is faster than anything else around, they will likely believe it...

An excellent point you made there Bob and that where Mr. is trying to get some pesos as long as his trick is not discovered yet....

As I wrote before,this project will die in the embrion stages of its development....
Dr.D

[De Vos W]

Mr. Rajlich has obfuscated (if that's the right word) his nps count for regular Rybka in the past, so how can we assume it's correct now?

I don't know how much Mr. Rajlich wants per-hour for the use of a cluster, but I can't imagine it being worthwhile even to a grandmaster. I suppose that he has a market, though; otherwise he wouldn't have spent the time.

This is about hyperbole and marketing. Not about logic and understanding. Lots of GM players are completely computer-ignorant when talking about this kind of issue. Faster is better, and if the claim is that this is faster than anything else around, they will likely believe it...

An excellent point you made there Bob and that where Mr. is trying to get some pesos as long as his trick is not discovered yet....

As I wrote before,this project will die in the embrion stages of its development....
Dr.D

This Rybka Rental baby is a miscarriage.
Lets stop the breating and start the bleeding!

[rbarreira]

I understand, and I would prefer e.g. a "time to depth x"(*) comparison. But I didn't find any other performance figures from the cluster yet, than these nps which are from a Rybka 4 benchmark page. I think that page was initially intended to compare "normal" comps running retail Rybka 4.

*) I guess time to depth x is not an optimal comparison either, because if I get things right the search tree on a bigger number of cpus will be bigger, for the same depths? - Maybe solving times for some good test positions are more relevant, in terms of "effective performance" so to speak.

Anyway we don't have such figures and I doubt that we will get them. Although, anyone willing to pay for it could rent Rybe cluster time just for that, too.

If many people are curious, you can set up a collection to rent some Rybka-cluster time for those tests. Then make a page with the results (which is probably what Vasik should do as well, if the cluster actually performs well).