Yes, on my phone, just tried to reply to that a bit strange results and some not very correct methodology. Just saw now that the second test is bogus. I have doubts about the first one too, theu should check against some regular engines.Guenther wrote: ↑Thu Dec 13, 2018 2:19 pmIt seems you did not read the thread? Test B is discredited - games for test A are not available...Laskos wrote: ↑Thu Dec 13, 2018 2:16 pmHave they been tried aganst regular engines? Last time I checked, 2 days ago, test30 was still some 40 Elo points behind test10 and advancing very, very slowly, on the average of less than 0.2 Elo points per net on 700 nets span. Against SF8.mehmet karaman wrote: ↑Thu Dec 13, 2018 12:49 pm For now 31970 is the best
31970- 11248: +9 -5 =186
Hardware used: 2080Ti and 6men TBs
Time control: 2min + 1sec
Book: Private (about 30 = positions)
31970- 11248: +39 - 27 =139
GPU 2080Ti and 6men TBs
5000 nodes/move
https://groups.google.com/forum/#!forum/lczero
Also, don't use fixed nodes when testing strength.
LCzero network
Moderators: hgm, Rebel, chrisw
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: LCzero network
-
- Posts: 142
- Joined: Tue Jan 28, 2014 8:37 am
- Location: TURKEY
Re: LCzero network
But how about the other 98 games. I don't think the games are scam
-
- Posts: 142
- Joined: Tue Jan 28, 2014 8:37 am
- Location: TURKEY
Re: LCzero network
31954 vs 11248: +5 =93 -2
GPU 2080Ti and used 6men TBs. Time control was 2min + 1sec
Private opening book.
And this is his old test result
GPU 2080Ti and used 6men TBs. Time control was 2min + 1sec
Private opening book.
And this is his old test result
-
- Posts: 142
- Joined: Tue Jan 28, 2014 8:37 am
- Location: TURKEY
Re: LCzero network
"There is a debate in discord about these results. On long TC (or high nodes count) test30 seems to be better. However, most of the tests are done at very short TC (or low nodes count) where it seems test10 is superior. So it seems test30 scales better… However, most of the elo tests are performed in bullet conditions."
From the Lc0 forum
From the Lc0 forum
-
- Posts: 4611
- Joined: Wed Oct 01, 2008 6:33 am
- Location: Regensburg, Germany
- Full name: Guenther Simon
Re: LCzero network
1. the rest is now 96mehmet karaman wrote: ↑Thu Dec 13, 2018 2:30 pm But how about the other 98 games. I don't think the games are scam
2. there is no doubt about the rest, otherwise he had noticed his error(?) and restarted the whole match
a quick check of the next 4 game pairs confirms this...
you should believe facts
https://rwbc-chess.de
trollwatch:
Talkchess nowadays is a joke - it is full of trolls/idiots/people stuck in the pleistocene > 80% of the posts fall into this category...
trollwatch:
Talkchess nowadays is a joke - it is full of trolls/idiots/people stuck in the pleistocene > 80% of the posts fall into this category...
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: LCzero network
How long TC on what hardware?mehmet karaman wrote: ↑Thu Dec 13, 2018 2:38 pm "There is a debate in discord about these results. On long TC (or high nodes count) test30 seems to be better. However, most of the tests are done at very short TC (or low nodes count) where it seems test10 is superior. So it seems test30 scales better… However, most of the elo tests are performed in bullet conditions."
From the Lc0 forum
-
- Posts: 1470
- Joined: Mon Apr 23, 2018 7:54 am
-
- Posts: 142
- Joined: Tue Jan 28, 2014 8:37 am
- Location: TURKEY
Re: LCzero network
Unfortunately there isn't much information about thisLaskos wrote: ↑Thu Dec 13, 2018 3:00 pmHow long TC on what hardware?mehmet karaman wrote: ↑Thu Dec 13, 2018 2:38 pm "There is a debate in discord about these results. On long TC (or high nodes count) test30 seems to be better. However, most of the tests are done at very short TC (or low nodes count) where it seems test10 is superior. So it seems test30 scales better… However, most of the elo tests are performed in bullet conditions."
From the Lc0 forum
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: LCzero network
Nets vary in speed, even among test30 nets there is quite a variarion. Strength is measured at same time control, not same number of nodes. Same time control could mean different number of nodes.
-
- Posts: 1470
- Joined: Mon Apr 23, 2018 7:54 am
Re: LCzero network
Interesting. Do you know why? I guess most assume same-size NN = same speed.