- Multigather is now made the default (and also improved). Some search settings
have changed meaning, so if you have modified values please discard them.
max-out-of-order-evals-factorhave changed default values, but other options
also affect the search. Similarly, check that your gui is not caching the old
- 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
legacytime manager will use more time for the first move after a long
--preloadcommand 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.
- A better value for the backendbench Clippy threshold.
- Fix additional cases where 'invalid move' could be incorrectly reported.
- Replace WDL softmax in cudnn backend with same implementation as cuda
backend. This fixes some inaccuracy issues that were causing training
data to be rejected at a fairly low frequency.
- Ensure that training data Q/D pairs form valid WDL targets even if there
is accumulated drift in calculation.
- Fix for the calculation of the 'best q is proven' bit in training data.
- Multiple fixes for timelosses and infinite instamoving in smooth time
manager. Smooth time manager now made default after these fixes.
- Fix a bug which meant
position ... moves ...didn't work if the moves went off the end of the existing tree. (Which happens normally when playing from an opening book.)
Note: This version is very broken, do not attempt use it.
- Multigather search inspired by Ceres. (Default is off. Note that the meaning of max-collision-events changes considerably when enabled and max-collision-visits will need to be set to a value close to previous values of max-collision-events in order to have similar search behavior.)
- V6 training format with additional info for training experiments.
- Updated default search parameters.
- A better algorithm for the backendbench assistant.
- Terminate search early if only 1 move isn't a proven loss.
- Various build system changes.
Starting with this release, we are distributing two packages for windows with Nvidia GPUs: the cuda package and the cudnn package. The cudnn package is what we used to distribute so far (but we called it cuda), and comes with the same versions of cuda and cudnn dlls we were using for the last few months. The new cuda package comes with cuda 11.1 dlls and requires at least version 456.38 of the windows Nvidia drivers, and should give better performance on RTX cards and in particular the new RTX 30XX cards.
- The cudnn package will work as-is in existing setups, but for the cuda package you may have to replace
cuda-fp16) as a backend (if specified) - this will certainly be necessary for multi-gpu setups.
- Some testing indicates that cuda 11.1 may be slower for GTX 10XX cards, so owners of older cards may want to stay with the cudnn package. If your testing shows otherwise do let us know.
- Fix for uninitialized variable that led to crashes with the cudnn backend.
- Correct windows support for systems with more than 64 threads.
- A new package is built for the
cudabackend with cuda 11.1. The old
cudapackage is renamed to
Note: The cuda package requires nvidia driver 456.38 or newer.
- Residual block fusion optimization for cudnn backend, that depends on
custom_winograd=true. Enabled by default only for networks with up to 384 filters in fp16 mode and never in fp32 mode. Default can be overridden with
--backend-opts=res_block_fusing=falseto disable (or
- New experimental cuda backend without cudnn dependency (
No changes since rc1. Enjoy!
- Repetitions in the search tree are marked as draws, to explore more promising lines. Enabled by default (except in selfplay mode) use
- Syzygy tablebase files can now be used in selfplay. Still need to add adjudication support before we can consider using this for training.
- Default net updated to 703810.
- Fix for book with CR/LF line endings.
- Updated Eigen wrap to use new download link.
If you build from source, note that old versions of meson cannot download from the new Eigen download link. You will either have to update meson or build with