TCEC: After 75 games of the 100 game match, Stockfish leads LCZero by +11-7=57. • CCC: Still no news on the technical problems affecting the GPU engines, which I first reported four weeks ago.
Once again, the title of this report gives the most important info on the current situation. Should I stop writing the reports and just post the titles? Fortunately, there's still a bit more to be said this time.
TCEC: In the previous post, I calculated,
Extrapolating to 100 games gives a final score of +15-9=76. The most recent 25 games went +6-3=16 for Stockfish.
The last 25 games finished +7-2=16 for a final score of +18-9=73 for Stockfish. The match was technically decided after 93 games. What's next? The TCEC '!commands' announce,
!next • Now GPU (Lc0, Allie, Stoof, Scorpio) !VSOB ~4d , Cup7 testing, Cup7 ~1week, StockfishClassical-LCZero Sufi Bonus 3 ~4d (might be earlier or later), FRC2 testing and FRC2 ~1.5 weeks. Then S20 QL+L4+L3, L2+L1, DivP, Sufi
VSOB stands for 'Viewer Submitted Openings Bonus', last seen on this blog in VSOB PGN (January 2020). The current event is titled 'GPU VSOB 1' and still has a couple of days to run.
CCC: The previous post, 'CCC15 Still Halted', gave a *running* commentary:-
The report from two weeks ago mentioned the site was running matches for 'CCRL Ratings Standardization'. Now it is running 'CCRL Calibration'.
And now the site is running 'Bullet Chess Is Fun'. What happens next? Who knows. A recent comment on the site's Discord resource informed, 'Since no one has said anything, the GPUs are back', but that was a week ago and has not been confirmed elsewhere.
[For further information from the various stakeholders in the engine-to-engine events, see the tab 'TCEC/CCC Links' at the top of this page. • NB: Leela = LC0 = LCzero]
No comments:
Post a Comment