New engine: Perception

[LazySMP]

Hello there,

My name is Daniel Pierce, and I’m a computer science graduate from Stanford University. Some time ago I developed LazySMP engine which was not well received and not even tested by some CCRL members and CEGT team. So I decided to develop a stronger engine derived from LazySMP. I'm not familiar with computer ratings, but I'd guess my engine is currently at 2500 Elo. Also, my engine speed is about 5 million NPS on a single thread.

Code: Select all

info score cp 19 depth 19 seldepth 22 nodes 49758651 time 11952 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8c5 e1g1 e8g8 f3e5 c6e5 d2d4 c5d4 d1d4 e5c6 d4e3 d7d6 b5c6 b7c6
info score cp 12 depth 20 seldepth 23 nodes 111173328 time 27000 pv e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 f1b5 f8c5 e1g1 e8g8 d2d3 d7d6 c1g5 c8e6 b5c6 b7c6 d3d4 e5d4 f3d4 c5d4 d1d4
info score cp 18 depth 21 seldepth 24 nodes 221694662 time 54171 pv e2e4 e7e5 g1f3 b8c6 f1c4 g8f6 d2d3 f8e7 b1c3 d7d6 e1g1 e8g8 a2a4 c8g4 c1e3 a7a6 h2h3 g4f3 d1f3 c6d4 e3d4 e5d4

Although my engine does not support Transposition Table, it is strong enough. Any feedback would be really appreciated!

Download link: https://drive.google.com/file/d/1AHnwET ... 2UeVg/view

[Jjaw]

[chesskobra]

As I had suggested earlier, if you make a linux executable, I am willing to run a bunch of tests in February. Are you using some specific technologies or frameworks or a language that you cannot make it more portable?

Do some testing on your own as well. It is not that hard to take 7-8 engines in the 2200-2600 range and run a gaunlet with sufficiently many rounds so that you have 300-400 games. You can do this easily with cutechess. That will give you a rough idea of the rating. Then upload the games somewhere so that people see that you are not just making claims about the rating.