Computer Chess Club
https://talkchess.com/
The Titan V also "only" has 640 of them. I suspect its successor will cram in many more.Dann Corbit wrote:Titan V has tensor cores.
But they are fiddly to use and you have to program especially for them.
How do you work that out? The Titan V isn't 2.2x faster than a 1080 ti according to either the GFLOPS (12300 vs 10600 respectively) or the benchmarks hereMilos wrote:
Judging be the current proper benchmark (LC0 on cuDNN), Titan V is only 3.5 times faster than GTX 960, and 2.2 times 1080ti. And GTX 960 is at least 15 times cheaper than Titan V.
They can multiply a small matrix in a single cycle (the tensor cores).Werewolf wrote:The Titan V also "only" has 640 of them. I suspect its successor will cram in many more.Dann Corbit wrote:Titan V has tensor cores.
But they are fiddly to use and you have to program especially for them.
You should look at windows version of LC0-cudnn on Titan V since CUDA libs for windows are obviously better for tensor cores (Titan V) than linux ones.Werewolf wrote:How do you work that out? The Titan V isn't 2.2x faster than a 1080 ti according to either the GFLOPS (12300 vs 10600 respectively) or the benchmarks hereMilos wrote:
Judging be the current proper benchmark (LC0 on cuDNN), Titan V is only 3.5 times faster than GTX 960, and 2.2 times 1080ti. And GTX 960 is at least 15 times cheaper than Titan V.
https://docs.google.com/spreadsheets/d/ ... edit#gid=0
Unless I'm misreading them.
4x4 one to be precise. Since LC0 kernal is 3x3 there is roughly only 1/3 efficiency ((27+9*2)/(81+16*3) operations) when running LC0 only on tensor cores, assuming ofc that they are fully loaded and that cuDNN libs are efficient for them, which is a big question mark at the moment.Dann Corbit wrote:They can multiply a small matrix in a single cycle (the tensor cores).Werewolf wrote:The Titan V also "only" has 640 of them. I suspect its successor will cram in many more.Dann Corbit wrote:Titan V has tensor cores.
But they are fiddly to use and you have to program especially for them.
Is it enough to download and install CUDA or do I need something else such as special commandline or executable? If so, can you point the way?Milos wrote:You should look at windows version of LC0-cudnn on Titan V since CUDA libs for windows are obviously better for tensor cores (Titan V) than linux ones.Werewolf wrote:How do you work that out? The Titan V isn't 2.2x faster than a 1080 ti according to either the GFLOPS (12300 vs 10600 respectively) or the benchmarks hereMilos wrote:
Judging be the current proper benchmark (LC0 on cuDNN), Titan V is only 3.5 times faster than GTX 960, and 2.2 times 1080ti. And GTX 960 is at least 15 times cheaper than Titan V.
https://docs.google.com/spreadsheets/d/ ... edit#gid=0
Unless I'm misreading them.
You need to compile proper executable, however there are already precompiles available.Albert Silver wrote:Is it enough to download and install CUDA or do I need something else such as special commandline or executable? If so, can you point the way?Milos wrote:You should look at windows version of LC0-cudnn on Titan V since CUDA libs for windows are obviously better for tensor cores (Titan V) than linux ones.Werewolf wrote:How do you work that out? The Titan V isn't 2.2x faster than a 1080 ti according to either the GFLOPS (12300 vs 10600 respectively) or the benchmarks hereMilos wrote:
Judging be the current proper benchmark (LC0 on cuDNN), Titan V is only 3.5 times faster than GTX 960, and 2.2 times 1080ti. And GTX 960 is at least 15 times cheaper than Titan V.
https://docs.google.com/spreadsheets/d/ ... edit#gid=0
Unless I'm misreading them.
Ok, that worked, thanks. I had downloaded CUDA 9.1, so renamed the files to 90 where needed. How do I do a full-tune? Or is that no longer done?Milos wrote:You need to compile proper executable, however there are already precompiles available.Albert Silver wrote:Is it enough to download and install CUDA or do I need something else such as special commandline or executable? If so, can you point the way?Milos wrote:You should look at windows version of LC0-cudnn on Titan V since CUDA libs for windows are obviously better for tensor cores (Titan V) than linux ones.Werewolf wrote:How do you work that out? The Titan V isn't 2.2x faster than a 1080 ti according to either the GFLOPS (12300 vs 10600 respectively) or the benchmarks hereMilos wrote:
Judging be the current proper benchmark (LC0 on cuDNN), Titan V is only 3.5 times faster than GTX 960, and 2.2 times 1080ti. And GTX 960 is at least 15 times cheaper than Titan V.
https://docs.google.com/spreadsheets/d/ ... edit#gid=0
Unless I'm misreading them.
Look here:
https://github.com/mooskagh/leela-chess/tree/master/lc0
Download 9.0 directly it is available from NVIDIA just search other releases.Albert Silver wrote:Ok, that worked, thanks. I had downloaded CUDA 9.1, so renamed the files to 90 where needed. How do I do a full-tune? Or is that no longer done?Milos wrote:You need to compile proper executable, however there are already precompiles available.Albert Silver wrote:Is it enough to download and install CUDA or do I need something else such as special commandline or executable? If so, can you point the way?Milos wrote:You should look at windows version of LC0-cudnn on Titan V since CUDA libs for windows are obviously better for tensor cores (Titan V) than linux ones.Werewolf wrote:How do you work that out? The Titan V isn't 2.2x faster than a 1080 ti according to either the GFLOPS (12300 vs 10600 respectively) or the benchmarks hereMilos wrote:
Judging be the current proper benchmark (LC0 on cuDNN), Titan V is only 3.5 times faster than GTX 960, and 2.2 times 1080ti. And GTX 960 is at least 15 times cheaper than Titan V.
https://docs.google.com/spreadsheets/d/ ... edit#gid=0
Unless I'm misreading them.
Look here:
https://github.com/mooskagh/leela-chess/tree/master/lc0