Page 1 of 2
128 CPU Stockfish Cluster Comments.
Posted: Sun Feb 03, 2019 6:56 pm
by Leo
Sorry if I missed a thread on this. I was taken aback by the 4 billion nps of this machine. Its found here:
http://www.ipmanchess.yolasite.com/amd- ... -bench.php Of course deep blue was at 200 million nps in the 1990s.
What would this entity be capable of? I think chess engines have become better at using more cores in the last few years but still need improvement using all these cores. It seems like it could take the chess world where no man has gone before.
Re: 128 CPU Stockfish Cluster Comments.
Posted: Sun Feb 03, 2019 8:07 pm
by pferd
Here are some numbers for the cluster branch against a single 32 thread machine. Would love to see this monster in action
https://github.com/official-stockfish/S ... /pull/1931
Re: 128 CPU Stockfish Cluster Comments.
Posted: Sun Feb 03, 2019 8:17 pm
by M ANSARI
With all these threads available maybe some of those cores would be better utilized for Monte Carlo search. You reach a point of diminishing returns and maybe those resources are better used via a different form factor.
Re: 128 CPU Stockfish Cluster Comments.
Posted: Sun Feb 03, 2019 11:15 pm
by henk2
If some billionaire was willing to waste all his money on dedicated Stockfish 10 ASICs.
Then the cluster could easily be blown away.
4 billion nps is still very impressive though.
Re: 128 CPU Stockfish Cluster Comments.
Posted: Mon Feb 04, 2019 11:18 pm
by Leo
In those tests the cluster did whomp on the 32 core SF. It looks like it maxed out at 130 Elo better.
Re: 128 CPU Stockfish Cluster Comments.
Posted: Mon Feb 04, 2019 11:19 pm
by Leo
M ANSARI wrote: ↑Sun Feb 03, 2019 8:17 pm
With all these threads available maybe some of those cores would be better utilized for Monte Carlo search. You reach a point of diminishing returns and maybe those resources are better used via a different form factor.
There are diminishing returns. Yes, I would like to see Monte Carlo tried out on it.
Re: 128 CPU Stockfish Cluster Comments.
Posted: Tue Feb 05, 2019 9:09 pm
by Dann Corbit
henk2 wrote: ↑Sun Feb 03, 2019 11:15 pm
If some billionaire was willing to waste all his money on dedicated Stockfish 10 ASICs.
Then the cluster could easily be blown away.
4 billion nps is still very impressive though.
That billionaire would learn the lesson of Hydra.
A dump truck full of money will make an engine that will win for 4 years.
Ten years later, a commodity PC will give it a spanking.
Re: 128 CPU Stockfish Cluster Comments.
Posted: Tue Feb 05, 2019 10:08 pm
by Jouni
This +130 gives nice 3700 ELO! Maybe the upper limit for chess?
Re: 128 CPU Stockfish Cluster Comments.
Posted: Tue Feb 05, 2019 10:47 pm
by Dann Corbit
Jouni wrote: ↑Tue Feb 05, 2019 10:08 pm
This +130 gives nice 3700 ELO! Maybe the upper limit for chess?
I doubt it, but possibly an upper limit for shared hash table alpha-beta.
On the other hand, the evaluation of LC0 is clearly vastly superior to the evaluation of Stockfish (due to less than 0.1% as many nodes during search giving an answer that is about equal in quality).
So suppose that the evaluation of Stockfish were tuned to equal the evaluation of LC0.
Perhaps the limit is 5000 Elo.
Re: 128 CPU Stockfish Cluster Comments.
Posted: Tue Feb 05, 2019 10:51 pm
by Werewolf
Dann Corbit wrote: ↑Tue Feb 05, 2019 10:47 pm
Jouni wrote: ↑Tue Feb 05, 2019 10:08 pm
This +130 gives nice 3700 ELO! Maybe the upper limit for chess?
I doubt it, but possibly an upper limit for shared hash table alpha-beta.
On the other hand, the evaluation of LC0 is clearly vastly superior to the evaluation of Stockfish (due to less than 0.1% as many nodes during search giving an answer that is about equal in quality).
So suppose that the evaluation of Stockfish were tuned to equal the evaluation of LC0.
Perhaps the limit is 5000 Elo.
No, I doubt any entity will get over 4000 elo. I think we'll start to see the draw rate dramatically rise as this level of play is reached. But I'd love to be proved wrong.