TalkChess.com

Here's another comparison...

Leela used 32.2 billion transistors while SF used 14.4 billion

The difficult part is that how does one say the hardware is equal - and one cannot. Essentially ,we're comparing two different breeds of cats and they both have a lot of plusses for what they can do. Thanks, that's an excellent explanation... But people have found ways to do it, using formulas base...

It's comical to see people rate chess engines by their performance in TCEC. TCEC is pure entertainment. Many specialized opening suites already exist to do proper testing. Yes, I can get one opening suite, maybe even two. Then all what remains is to play a dual RTX GPU against 43 CPU cores 100 game...

It's comical to see people rate chess engines by their performance in TCEC.
TCEC is pure entertainment.
Many specialized opening suites already exist to do proper testing.

Nay Lin Tun wrote: ↑

Wed May 15, 2019 12:49 pm

You will be surprised that Leela actually already beat 44 cores (SF with 10x TC bonus ) in one pair of opening! 3 mins vs 30 mins.

No point having a million cores searching the wrong lines.

corres wrote: ↑

Sat Apr 20, 2019 5:34 pm

So we have different aim: I (and others) should like to know the power of own card relative to others.

Check benchmarks of various configurations with the actual settings here...

https://docs.google.com/spreadsheets/d/ ... 1508569046

Leo wrote: ↑

Mon Apr 15, 2019 2:50 pm

It sounds interesting. Good thinking. Let lczero run day and night calculating a book for SF.

I'm sorry to say that the status of the Sirius GUI has not changed, in the last two years neither I or my brother had enough time to finish this project, because it takes an enormous amount of time to test everything, fix all bugs and make it user friendly. I recall offering my help on 3rd Dec, 201...

The 950M for laptops outputs at 1kn.
You should get much more than 1 kn with a desktop card.

Raphexon wrote: ↑

Sat Apr 13, 2019 1:31 pm

Would it be possible to distill the latest network to 16x2?

The distillation process doesn't work on test 40 and above networks, yet.