Search found 2204 matches

by Michael Sherwin
Thu Sep 19, 2019 6:13 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Back To The Beginning
Replies: 35
Views: 6117

Re: Back To The Beginning

Just a note about overclocking. I finally got a 4.4GHz overclock (3.2GHz is stock) stable across all apps and stress test. The nice thing is that it was done with a lower cpu voltage because I finally got the vitt frequency correct. Because of the lower cpu voltage it runs cooler than before. SF run...
by Michael Sherwin
Wed Sep 11, 2019 9:50 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Back To The Beginning
Replies: 35
Views: 6117

Re: Back To The Beginning

When I had first written qperft (I was using a Pentium M at the time), I tried to speed it up by using assembler. To not waste time on non-time-ritical code, I started with the assembler output of the C compiler, and 'hand-optimized' the code of the time-critical loops. It turned out I could delete...
by Michael Sherwin
Wed Sep 11, 2019 4:35 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Back To The Beginning
Replies: 35
Views: 6117

Re: Back To The Beginning

thought I would reply out of interest. On my 2 year old MacBook, 2.2ghz i7 circa 2017 I see this number for perf 6 from start position: White(1): perft 6 total moves=119060324 time=3.11 White(1): Here's the next one: White(1): perft 7 total moves=3195901860 time=82.65 My perft was never optimized f...
by Michael Sherwin
Sun Sep 08, 2019 2:21 am
Forum: Computer Chess Club: General Topics
Topic: BrainLearn
Replies: 16
Views: 2832

Re: BrainLearn

Finally, again public! https://github.com/amchess/BrainLearn This is an article explaining our idea: https://github.com/amchess/BrainLearn/blob/master/doc/MASTERING%20THE%20GAME%20OF%20CHESS%20THROUGH%20A%20RE.pdf Have fun! Kelly and Andrea Forgive me, but it seems all you've done is add a "permane...
by Michael Sherwin
Sat Sep 07, 2019 8:28 pm
Forum: Computer Chess Club: General Topics
Topic: Irresistible Force Meets Immovable Object
Replies: 3
Views: 689

Irresistible Force Meets Immovable Object

by Michael Sherwin
Sat Sep 07, 2019 6:11 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Back To The Beginning
Replies: 35
Views: 6117

Re: Back To The Beginning

On my 6950X at 4GHz. I get on average 52 mnps without bulk counting and including updating/downdating hash, pawn-hash, material-hash, material-score and phase. The latest processors from Intel have a 15 to 20% higher IPC and a somewhat higher clock-frequency, so my guess is that I would get between...
by Michael Sherwin
Sat Sep 07, 2019 5:53 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Back To The Beginning
Replies: 35
Views: 6117

Re: Back To The Beginning

On my 6950X at 4GHz. I get on average 52 mnps without bulk counting and including updating/downdating hash, pawn-hash, material-hash, material-score and phase. The latest processors from Intel have a 15 to 20% higher IPC and a somewhat higher clock-frequency, so my guess is that I would get between...
by Michael Sherwin
Sat Sep 07, 2019 5:43 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Back To The Beginning
Replies: 35
Views: 6117

Re: Back To The Beginning

For those that might be interested in the difference in speed of the pure C version and the assembler version these are the numbers on a 3930k at stock 3.2 GHz. Bench 6 Pure C, 33 879 522 moves per second Bench code in C, move gen and MakeMove in assembler. 65 608 247 moves per second I don't believ...
by Michael Sherwin
Sat Sep 07, 2019 5:05 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Back To The Beginning
Replies: 35
Views: 6117

Re: Back To The Beginning

On an i7- 3930k overclocked to 4.4 GHz using only one thread in the opening position the assembler version generates 75 million positions per second. That includes all moves made and unmade and no hash counting. On an i9-9900k overclocked to 5GHz it would do double that. Why would going from 4.4 GH...
by Michael Sherwin
Sat Sep 07, 2019 10:42 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Back To The Beginning
Replies: 35
Views: 6117

Re: Back To The Beginning

My new chess engine has no evaluation function and therefore move generation speed becomes more important. Not having an evaluation function? So it just generates moves, searches through trees, and then doesn't evaluate them? Did you actually benchmark how much time the move generation takes? I don...