TalkChess.com

NN-testrun of Lc0 0.26.3 67336 finished.

https://www.sp-cc.de

(Perhaps you have to clear your browsercache or reload the website)

AB-testrun of Nemorino 6.05 (net 16) avx2 finished.

https://www.sp-cc.de

(Perhaps you have to clear your browsercache or reload the website)

AB-testrun of RubiChess 2.0 finished.

https://www.sp-cc.de

(Perhaps you have to clear your browsercache or reload the website)

NN-testrun of Lc0 0.26.3 67438 finished

https://www.sp-cc.de

(Perhaps you have to clear your browsercache or reload the website)

Gabor Szots wrote: ↑

Sat Jan 30, 2021 8:33 pm

I've updated the uploaded file using Günther's suggestion.

Yes, this binary runs on my Ryzen CPU and it is measureable faster than the avx2-compile on the Nemorino-website!

Testrun of Nemorino 6.05 avx2 started. Result on Tuesday...

https://www.sp-cc.de/

someone please can make a windows compile? I have uploaded an AVX2 compile here: https://drive.google.com/drive/folders/1AuVJ15IzldglF6kvxzdhwxIreoNOB2xX Does not run on my AMD Ryzen. But as far as I understand, what is written on the Nemorino Website, the only change from Nemorino 6.00 to 6.05 is ...

Intersesting. I see you underclock your GPU more now that lc0 26.3 is faster. Did you test that it was faster with the size of LS or PS? I think the speedup mostly comes from nets with lots of filters, especially the biggest 384 filters. Probably speedup is half for the small nets, although the nex...

RubiChess 2.0 was released !! Seems to be clearly stronger than V1.9 in selfplay (tests by the author of RubiChess): 20+0.2 on a BMI2+AVX machine: Score of 2.0 vs 1.9: 295 - 165 - 740 [0.554] 1200 Elo difference: 37.79 +/- 12.11 60+0.6 on a POPCNT/SSSE3 machine: Score of 2.0 vs 1.9: 521 - 202 - 127...

NN-testrun of Lc0 0.26.3 PhoenixStein 1.2 net finished.

https://www.sp-cc.de

(Perhaps you have to clear your browsercache or reload the website)

Why is the last green NNUE already 2 months old ? After that, about 100 new networks have already appeared, but none is recommended (i.e. green). Are they all worse than nn-62ef826d1a6d.nnue ? The series of “green” nets were trained via RL (reinforcement learning) and possibly are very close to the...