Something goes wrong with lc0 since yesterday?

[crem]

So... the main server nets are gone, never to return?

The old server is still running under the url http://oldmain.lczero.org/
We'll have all networks conveniently available for download in one place eventually.

[Uri Blass]

I do not know what is wrong with lc0 but I read at the lc0 forum that lc0 is good only for bullet.

I read the following in this thread:
https://groups.google.com/forum/#!topic ... 1iEenDOpVM

"it seems above 400,000 nodes ELO growth diminishes heavily (1-7 ELO per doubling NPM) "

"From my tests Leela is far stronger than Stockfish at very short time controls, however given longer thinking time and more nodes, SF outscales a 20 block Leela ID 480 "

I know that A0 scaled better than stockfish8 and certainly earned elo from more time and it seems that the lc0 team has a serious bug or do something clearly different than A0 or both of them.

[tmokonen]

So... the main server nets are gone, never to return?

The old server is still running under the url http://oldmain.lczero.org/
We'll have all networks conveniently available for download in one place eventually.

Thanks!

[Milos]

I know that A0 scaled better than stockfish8 and certainly earned elo from more time and it seems that the lc0 team has a serious bug or do something clearly different than A0 or both of them.

A0 scaling vs SF from that crappy paper is totally bogus. It is a fact! I wonder when are ppl gonna stop citing that BS graph...

[jp]

I know that A0 scaled better than stockfish8 and certainly earned elo from more time

A0 scaling vs SF from that crappy paper is totally bogus. It is a fact! I wonder when are ppl gonna stop citing that BS graph...

I wonder why Tord or some other SF guys don't just test & put up their own graph for SF scaling. I think they should.

[Dann Corbit]

I know that A0 scaled better than stockfish8 and certainly earned elo from more time

A0 scaling vs SF from that crappy paper is totally bogus. It is a fact! I wonder when are ppl gonna stop citing that BS graph...

I wonder why Tord or some other SF guys don't just test & put up their own graph for SF scaling. I think they should.

Because Pohl does that for him.
https://www.sp-cc.de/

[jp]

Because Pohl does that for him.
https://www.sp-cc.de/

Thanks, Dann, but I don't see a Pohl graph of (fixed version of) SF, elo vs. time/move (really nodes/move). Is one there?

[Uri Blass]

I know that A0 scaled better than stockfish8 and certainly earned elo from more time and it seems that the lc0 team has a serious bug or do something clearly different than A0 or both of them.

A0 scaling vs SF from that crappy paper is totally bogus. It is a fact! I wonder when are ppl gonna stop citing that BS graph...

I prefer to assume they are not lying but if they are lying then what is the reason for them not to tell the public that A0 is good at bullet time control.

It is better to be strong both at bullet time control and at long time control and not only at long time control.

[jp]

A0 scaling vs SF from that crappy paper is totally bogus. It is a fact! I wonder when are ppl gonna stop citing that BS graph...

I prefer to assume they are not lying

Shouldn't assume anything. "Lying" is also an assumption about why it's wrong.

I think Milos is saying the SF line is totally wrong. SF devs or testers (maybe Pohl) should have a graph somewhere to see.

[jkiliani]

I know that A0 scaled better than stockfish8 and certainly earned elo from more time and it seems that the lc0 team has a serious bug or do something clearly different than A0 or both of them.

A0 scaling vs SF from that crappy paper is totally bogus. It is a fact! I wonder when are ppl gonna stop citing that BS graph...

I prefer to assume they are not lying but if they are lying then what is the reason for them not to tell the public that A0 is good at bullet time control.

It is better to be strong both at bullet time control and at long time control and not only at long time control.

I also think it's more plausible that we either have some sort of bug that hurts scaling, or simply haven't figured out some part of the AlphaZero approach properly yet. One experiment being tried now is changing the search parameters, i.e. PUCT, FPU reduction and futile move aversion constant. I'm confident that this will eventually be solved, it will just take some time now that this issue has come to light.