TalkChess.com

Been streaming against lc0 for about 12 hours now I guess and Ceres is ahead even with just 1 GPU. With 2 or 4 GPUs the speedup should help the bad TM even more. It doesn't use as much time in tricky spots as lc0, or ever pretty much. https://www.twitch.tv/jjoshua2

For now just use latest release at https://github.com/nodchip/Stockfish/releases 384 architecture and avx2 or bmi2 etc

Some of the recent patches gain 30-60 elo so matters more than a little speed. I will do better compiles again when I have time, probably next weekend, but might not be any faster.

Someone told me there was an error when I posted, but its resolved now.

https://www.patreon.com/posts/stockfinn-0-2-38923625 The net is one of the best, but the binary being made more than a few days ago is already out of date now with those massive bugfixes for TM and fixed nodes. I've seen other recent compiles floating around. Some of them have said they support 30mb...

People are understandably jaded from many bad SF clones that pretended to be stronger and very different from SF even with alphazero technology, but were in fact almost identical to SF... Makes it hard to convince people when in fact the entire evaluation function has no shared essence at all.

The main point is that it produces evals and moves more unique as a 2nd opinion to Stockfish than any stockfish clone or even Komodo, and yet manages to be stronger be much stronger than Komodo. Whether it is 30 elo above SF or below is largely irrelevant here. But there are some circumstances in wh...

It looks like SPCC uses a skylake CPU so that build should be good. Sadly I do not have an intel machine to test and compile with, but it should be just a little bit faster than the AVX2 builds I would think. It looks like SPCC also uses Hertz book which I did some tests with and found it was simila...

The bench test didn't work when I tried it in SFNN2, Simply type isready and it will load net. Then bench command will work. So you cannot just run sf.exe bench outside, but rather load the exe, then the net, then run bench. Single threaded bench I got 82% of SF-dev speed, but 16 thread it is down ...

Normal users have struggled using lc0 for a very long time. Most chess players are not engine enthusiasts, and even struggle with commercial products like chessbase software, which is a reason why they are so big, and why they found it worth doing fat fritz. I don't like it all, but anyone with an a...