how good is a GeForce GTX 1060 6GB for Leela ?

[Robert Flesher]

I got that message once. It was because I forgot to extract the weights file from the weights_###.txt.gz package.

Well, since quite a while that is not necessary anymore.
LCZero meanwhile reads the compressed file directly too.

I have no idea what I am doing wrong but I cannot get it to run. I get the same message over and over!

Can you describe exactly what you are doing and what files are there?

C:\users\robert\desktop\lczero\lczero.exe --tune-only --full-tuner -w network


the id file is
is in the LCzero folder and named network

Did you check that it is really renamed to network w/o any extension as I wrote already earlier? (file manager : display extensions for known file types)

Yes

However, I found a way to make it work. I renamed the network to weights.txt, then ran the bat file that the chessbase article had a link to, and it worked. Strange!

[Hai]

Titan V best by test

[AdminX]

Titan V best by test

Hey,

Don't forget to pick me up one while you are out! Thanks in advance for the gift.

[JohnS]

Titan V best by test

Hey,

Don't forget to pick me up one while you are out! Thanks in advance for the gift.

Actually you should be buying us one of these cards each. Otherwise we may be inclined to alert your better half to the real reason you bought your new card. Don’t think of this as a blackmail threat, just a friendly incentive.

[AdminX]

Titan V best by test

Hey,

Don't forget to pick me up one while you are out! Thanks in advance for the gift.

Actually you should be buying us one of these cards each. Otherwise we may be inclined to alert your better half to the real reason you bought your new card. Don’t think of this as a blackmail threat, just a friendly incentive.

LMAO, Touche! "It may pay to Discover" but I'm not Discovering that.

[Werewolf]

Titan V best by test

Is that true?

The processing power of the Titan V (in GFLOPS) isn't vastly higher than a 1080 ti, but the price is 4x higher!

[Dann Corbit]

Titan V has tensor cores.
But they are fiddly to use and you have to program especially for them.

Otherwise, it is not a bargain

[jpqy]

I would go for this one..only check prise when you sit down

https://www.nvidia.com/en-us/data-center/tesla-v100/

[Milos]

Titan V has tensor cores.
But they are fiddly to use and you have to program especially for them.

Otherwise, it is not a bargain

They are absolutely not worth the money.
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. But ppl with a lot of money are usually stupid. Why would computer chess be an exception.
Btw. on my GTX 770 that I bought second hand for 70$, I get 8000nps with cuDNN batch size 256 on windows. That is only 5 times worse than Titan V, and price-wise Titan V is like 40 times more expensive .

[Albert Silver]

Titan V has tensor cores.
But they are fiddly to use and you have to program especially for them.

Otherwise, it is not a bargain

They are absolutely not worth the money.
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. But ppl with a lot of money are usually stupid. Why would computer chess be an exception.
Btw. on my GTX 770 that I bought second hand for 70$, I get 8000nps with cuDNN batch size 256 on windows. That is only 5 times worse than Titan V, and price-wise Titan V is like 40 times more expensive .

You're saying there is away to accelerate LC0? How?