-
Notifications
You must be signed in to change notification settings - Fork 2.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
PV-Nodes likely to fail low #3349
Conversation
congrats!!! |
Congrats too, way to go!! :-) |
Merged via 76daa88, thanks :-) |
According to the comment, this should be
|
Yeah, I've already noticed this. |
It becomes riskier at very high depths. Probably nothing fishtest can detect. Even though it passed SPRT, I'm not happy it was committed as is. Sorry. |
At best what it does is making reductions change by 2 - at higher depths it effect becomes more and more subtle. What it means is that it will have no effect as soon as we increase depth by 2. |
I apologize for conflating matters, but I have a modified engine which without NNUE supports exotic (more than 32 pieces) material combinations; and while that in itself is not supported, the other day I observed an infinite recursion in qsearch. I'm not asking for support but am curious whether there are legal positions which cause a regression. |
qsearch has nothing to do with this patch, write an issue. |
Do not decrease reduction at pv-nodes which are likely to fail low. The idea of this patch can be described as following: during the search, if a node on the principal variation was re-searched in non-pv search and this re-search got a value which was much lower than alpha, then we can assume that this pv-node is likely to fail low again, and we can reduce more aggressively at this node. Passed STC https://tests.stockfishchess.org/tests/view/6023a5fa7f517a561bc49638 LLR: 2.95 (-2.94,2.94) {-0.25,1.25} Total: 70288 W: 6443 L: 6223 D: 57622 Ptnml(0-2): 239, 5022, 24436, 5174, 273 Passed LTC https://tests.stockfishchess.org/tests/view/6023f2617f517a561bc49661 LLR: 2.94 (-2.94,2.94) {0.25,1.25} Total: 105656 W: 4048 L: 3748 D: 97860 Ptnml(0-2): 67, 3312, 45761, 3630, 58 Closes official-stockfish/Stockfish#3349 Bench: 3766422
Do not decrease reduction at pv-nodes which are likely to fail low.
Passed STC
https://tests.stockfishchess.org/tests/view/6023a5fa7f517a561bc49638
LLR: 2.95 (-2.94,2.94) {-0.25,1.25}
Total: 70288 W: 6443 L: 6223 D: 57622
Ptnml(0-2): 239, 5022, 24436, 5174, 273
passed LTC
https://tests.stockfishchess.org/tests/view/6023f2617f517a561bc49661
LLR: 2.94 (-2.94,2.94) {0.25,1.25}
Total: 105656 W: 4048 L: 3748 D: 97860
Ptnml(0-2): 67, 3312, 45761, 3630, 58
Idea of this patch can be described as following - if PvNode was researched in non-pv search and value this search got was much lower than alpha we can assume that this PvNode is likely gonna fail low so we can reduce more at it.
bench 3766422