abulmo2 wrote:What do you put on the FPGA, just the eval? the eval + qs search? the upper part of the search? with the hash table ? How would you make it parallel? Maybe you could reach 30 Bnps, but I doubt it will be with the same efficiency than in current SF search.
Move gen, eval and qs, probably even small sub-tree, local hash per FPGA, root positions copied to global hash regularly.
Ultrascale+ is gen4 PCIe and supports OpenCAPI. If haven't heard of it Google it. It is far memory performance comparable to DIMMs so no bottleneck there.
Evert wrote:This is irrelevant, and besises the point. People arguing about hardware or number of cores are equally missing the point.
Even discussions of which is stronger under what conditions, or whether Stockfish dev would be stronger or not, miss the point.
The astounding thing here is that Alpha Zero is in the same ballpark as Stockfish. Especially given the claim of how little time was spent training Alpha Zero.
How could 1850 and 3200 be in the same league?
Those 350 points are the most difficult to achieve.
You really believe the 4 hours stunt?
Then, 48 hours later, it is at least, 4000 elo. You really believe it?
Why did not they make a new statement then?
Evert wrote:This is irrelevant, and besises the point. People arguing about hardware or number of cores are equally missing the point.
Even discussions of which is stronger under what conditions, or whether Stockfish dev would be stronger or not, miss the point.
The astounding thing here is that Alpha Zero is in the same ballpark as Stockfish. Especially given the claim of how little time was spent training Alpha Zero.
whats does that means practicaly? even a supercomputer before aplha zero can make
a 80% score vs st8
its not about chess its about NN revolution with special chips(TPU-tensor cpu),thats the missed point.
No it can't. Hydra died because Rybka running on a puny computer was able to outplay it despite Hydra running on expensive hardware.
Johnny runs on these crazy 2000+ core servers and it's no match for Stockfish. There's only so much that hardware can do for software.
Stockfish 8 probably doesn't get much of an elo increase after 32 cores. You can give Stockfish 8 two billion cores and it wouldn't beat AlphaZero.
But it would beat it on a TPU.
Why would we like to claim a 1800 engine is stronger than that?
That is what they achieved with their software: a 1800-elo engine.
The rest is simply hardware.
stavros wrote:to make it more practical: i want to know the price of 64core st hardware and the price of alpha zero hardware pls anyone? to make a more fair comparison
+ 10, most probably larger than 20/1, just as in the case with hardware speed.
And it used significantly more memory.
And an opening book, which actually decided the whole match.
stavros wrote:to make it more practical: i want to know the price of 64core st hardware and the price of alpha zero hardware pls anyone? to make a more fair comparison
+ 10, most probably larger than 20/1, just as in the case with hardware speed.
And it used significantly more memory.
And an opening book, which actually decided the whole match.
stavros wrote:to make it more practical: i want to know the price of 64core st hardware and the price of alpha zero hardware pls anyone? to make a more fair comparison
+ 10, most probably larger than 20/1, just as in the case with hardware speed.
And it used significantly more memory.
And an opening book, which actually decided the whole match.
shrapnel wrote:The Age of Chess Engines based on alpha-beta Search is over.
This is just nit-picking/quibbling by old men who are too set in their ways of thinking to fully grasp or accept the sudden change in the World of Computer chess.
No point arguing with them, because if the change is for real, they will have to accept the Truth sometime or the other or risk making themselves a laughing stock.
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)
stavros wrote:to make it more practical: i want to know the price of 64core st hardware and the price of alpha zero hardware pls anyone? to make a more fair comparison
+ 10, most probably larger than 20/1, just as in the case with hardware speed.
And it used significantly more memory.
And an opening book, which actually decided the whole match.
We don't know the price of a TPU so your comparision doesn't mean anything
Price is infinite because no one can buy it. Comparing with equivalent V100 system is certainly much cheaper and certainly makes sense. What obviously doesn't make sense is your comment.
stavros wrote:to make it more practical: i want to know the price of 64core st hardware and the price of alpha zero hardware pls anyone? to make a more fair comparison
+ 10, most probably larger than 20/1, just as in the case with hardware speed.
And it used significantly more memory.
And an opening book, which actually decided the whole match.
We don't know the price of a TPU so your comparision doesn't mean anything
Price is infinite because no one can buy it. Comparing with equivalent V100 system is certainly much cheaper and certainly makes sense. What obviously doesn't make sense is your comment.
You shouldnt make any price comparisions then. Additionally, The V100 chips are too general purpose compared to TPU's....
stavros wrote:to make it more practical: i want to know the price of 64core st hardware and the price of alpha zero hardware pls anyone? to make a more fair comparison
+ 10, most probably larger than 20/1, just as in the case with hardware speed.
And it used significantly more memory.
And an opening book, which actually decided the whole match.
We don't know the price of a TPU so your comparision doesn't mean anything
Price is infinite because no one can buy it. Comparing with equivalent V100 system is certainly much cheaper and certainly makes sense. What obviously doesn't make sense is your comment.
You shouldnt make any price comparisions then. Additionally, The V100 chips are too general purpose compared to TPU's....
Price comparison makes perfect sense coz Google used something that would cost at least a quarter of million bucks to run on (not yet existing) top of the range GPUs to beat old, handicapped version of chess program on computer that is under 3'000$ and ppl call it fair???
I'm just offering a different view on it.
I understand if you don't like it, but again it is not meant for you.
Additionally, The V100 chips are too general purpose compared to TPU's....