Closed anuejn closed 3 years ago
I added some further optimizations, these bring the time for a simple toplevel read design.something
down to about 33us
from about 410us
.
I also tried another variant here: https://github.com/apertus-open-source-cinema/naps/commit/1651f85401ef2c68cad5310ebdfeed3056c7198e but that one is actually slightly slower.
I also tried to replace BitwiseAccessibleInteger with bitarray but that is also actually slower
there is still a BitwiseAccessibleInteger Implementation that is not used in pydriver for performance reasons but seemed handy to have for other applications.