| Community Blog Discussions - Using LC0's WDL to Analyze Games#30 @affstein said in #29: > As these graphs tend to look a bit twitchy, I was wondering, if it makes more sense to just scale draw to a constant (and discarding it) and have white's winning "chances" abo… | RwSF75  |
| Community Blog Discussions - Using LC0's WDL to Analyze Games#26 @ItsGam3Tyme said in #25: > *rawWinningChances(300) returns 0.5. You assume that this is a percentage but it is not. The scale is -1 to +1. If you multiply by 50 to get [-50 to + 50] and add 50 to get… | RwSF75 |
| Community Blog Discussions - Using LC0's WDL to Analyze Games#20 @dboing said in #19: > Both lichess and SF are using lichess or equivalent human databases to compute their WDL conversion curves with some statistical assumptions within. LC0 uses its own games, and … | RwSF75 |
| Community Blog Discussions - Using LC0's WDL to Analyze Games#17 @TotalNoob69 said in #14: > BTW, Stockfish 16 also has a parameter to output WDL values. I used it to create a visual bar in the Move Assistant feature of LiChess Tools. I actually find it more intere… | RwSF75 |
| Community Blog Discussions - Using LC0's WDL to Analyze Games#5 @Craftyawesome said in #4: > +3 doesn't sound right. IIRC it was 1.x measured with some version of SF. rawWinningChances(300) returns 50% The model they use is based on Lichess 2300+ Elo rated rapid g… | RwSF75 |
| Community Blog Discussions - Using LC0's WDL to Analyze Games#3 I don't think Lichess's graphs are centipawn, I think they are WDL too. They are most likely using the same model they use to determine blunders and how the eval bar should be filled. This means that … | RwSF75 |
| Lichess Feedback - Forced checkmate sequence can increase in length as analysis goes deeper#55 @schlawg said in #50: > Stockfish mate results are viewed as guarantees in practice, but are not necessarily proven. Numerous examples (such as this one) demonstrate they can be ephemeral even within … | RwSF75 |
| Lichess Feedback - Forced checkmate sequence can increase in length as analysis goes deeper#48 @schlawg said in #47: > Assuming that you understand how a mate in n increases at greater depth within the same search (having plenty of free hash to spare), can you explain it here? I do not know why… | RwSF75 |
| Lichess Feedback - Forced checkmate sequence can increase in length as analysis goes deeper#46 It is not uncommon for chess engines like Stockfish to "lose" a shorter mate. This does not mean that the shorter mate announced previously is wrong because Stockfish makes sure that they are proven t… | RwSF75 |
| Lichess Feedback - what is the difference btw SF16 7MB & SF16 40MB#11 Both the Stockfish 16 (7MB) and Stockfish 14 versions currently available at Lichess are using unofficial and smaller NNUEs designed to be faster to download. They are weaker than their official count… | RwSF75 |
| Next |
|---|