Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
give a little bonus for moving pieces to squares where they give check STC: https://tests.stockfishchess.org/tests/view/631da742162491686d2e40b5 LLR: 2.95 (-2.94,2.94) <0.00,2.00> Total: 80072 W: 21753 L: 21368 D: 36951 Ptnml(0-2): 421, 8876, 21075, 9225, 439 LTC: https://tests.stockfishchess.org/tests/view/631dd9e6b85daa436625de1d LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 263480 W: 70916 L: 70158 D: 122406 Ptnml(0-2): 322, 26156, 78029, 26908, 325 similar ideas have been tested by Viz and Guenther closes #4165 bench: 4326572
- Loading branch information
dc0c441
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@Vizvezdenec
Thank you for approving my test. But why did you lower the throughput from 100 to 50? I'm at this moment running only one test, unlike you. Your 5 tests together are taking 500 cores, while my test only 30.
dc0c441
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
because it's master vs master test and not an attempt for immediate elo gain, this tests are usually ran at 50% tp.
dc0c441
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why do you think tests not for immediate elo gain are less important?
There are also SPSA tests, simplification SPRT tests - they don't immediately gain elo either and are usually run with 100 throughput.
dc0c441
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sorry for posting this in the official repository. I wanted to post these comments in my own repository, but conflated them.