voffka wrote:Hello Guenther,
guenther wrote:
And I just cannot believe my eyes, 3.0.0 seems three! times faster now for old hardware? (ok before it was really slow with nn)
I doubt Igel became significantly faster. It is true that I removed some extensions (no capture extensions, less history extensions, etc) and Igel should be able to reach bigger depth faster, but this would not be a huge difference compared to 2.9.0 or dev-0.
I think you need to have a bigger sample size of positions to see more results :)
voffka wrote: ↑Mon Apr 05, 2021 8:16 pm
...
Thanks! During last releases it became significantly easier to compile Igel using gcc with the existing makefile. Please give it and try and if it does not work - I will help you to compile it.
This way you will not depend on me and will be able to compile future of versions of Igel :)
Well you were right about the speed diff to 3.00, actually it was the other way round!
All my compilations from 2.70 to 2.90 simply were three times too slow, because I used a simplified makefile in the beginning
and didn't check your improvements in the makefile later and always compiled with what was 'working' for 2.70.
Now with 3.00 I adapted/modified the newer makefile and finally got the speed I should have had already before ;-)
(short story: with -march=native I thought the usual instruction sets would be automatically taken already until I now saw
that they should be added additionally for triggerering some definitions in e.g. nnue.common.h for speeding nnue up)
I recompiled all my versions for my hardware from 2.70 to 2.90 and now they all run between 320kn/s to 330 kn/s from the
start position, before they were only around 110-120 kn/s!