Hier auf github:
https://github.com/LeelaChessZero/lc0/releases/tag/v0.28.0-rc1Multigather is now made the default (and also improved). Some search settings
have changed meaning, so if you have modified values please discard them.
Specifically, max-collision-events, max-collision-visits and
max-out-of-order-evals-factor have changed default values, but other options
also affect the search. Similarly, check that your gui is not caching the old
values.
Performance improvements for the cuda/cudnn backends.
Support for policy focus during training.
Larger/stronger 15b default net for all packages except android, blas and dnnl
that get a new 10b network.
The distributed binaries come with the mimalloc memory allocator for better
performance when a large tree has to be destroyed (e.g. after an unexpected
move).
The legacy time manager will use more time for the first move after a long
book line.
The --preload command line flag will initialize the backend and load the
network during startup.
A 'fen' command was added as a UCI extension to print the current position.
Experimental onednn backend for recent intel cpus and gpus.
Builds will show up at the above link shortly.
Note that this release will **not** work for contributing to training.
Feel free to discuss the release on the new GitHub discussion page:
<
https://github.com/LeelaChessZero/lc0/discussions/1579>
Viele Verbesserungen!
Viel Spaß!
Lothar
PS: Ich würde erstmal die Steigerung der Performance aus der Grundstellung gegenüber Lc0 v0.27.0 testen.
Danach könnte man Vergleiche zwischen den üblichen Testsuites anstellen.