Search found 2210 matches

by Michel
Sun Jul 11, 2021 9:34 am
Forum: Computer Chess Club: General Topics
Topic: Is Scorpio stronger than AlphaZero?
Replies: 9
Views: 1736

Re: Is Scorpio stronger than AlphaZero?

chrisw wrote:
Sun Jul 11, 2021 8:45 am
AndrewGrant wrote:
Sun Jul 11, 2021 4:44 am
connor_mcmonigle wrote:
Sun Jul 11, 2021 4:28 am
Daniel Shawul wrote:
Sun Jul 11, 2021 2:37 am
...
Also May I suggest you take a step back of from starting so many posts in talkchess -- I certainly didn't ask for Scorpio to be compared to A0.
+1
+2
+3
+4

What is the point of trying to stir up all these contentious issues...? It seems very close to trolling.
by Michel
Fri Jul 09, 2021 5:39 pm
Forum: Computer Chess Club: General Topics
Topic: Are hybrid approaches stronger than pure nnue?
Replies: 15
Views: 1601

Re: Are hybrid approaches stronger than pure nnue?

elo, especially outside the PV. Seer and Pedone/Revenge are two examples of engines that only use pure nnue. There are a handful, but most engines using nnue today do so alongside their HCE. It’s a free 20-40 elo. What?? In Stockfish it is 1-2 Elo https://tests.stockfishchess.org/html/live_elo.html...
by Michel
Fri Jul 09, 2021 3:11 pm
Forum: Computer Chess Club: General Topics
Topic: Are hybrid approaches stronger than pure nnue?
Replies: 15
Views: 1601

Re: Are hybrid approaches stronger than pure nnue?

Why hybrid earns elo? If it is only because of speed then I wonder if it holds at longer time control. If it is because nnue is relatively bad in positions when one side has a big advantage then maybe it is better to improve nnue and teach it based on scoring system that not all wins are the same a...
by Michel
Fri Jul 09, 2021 6:42 am
Forum: Computer Chess Club: General Topics
Topic: Are hybrid approaches stronger than pure nnue?
Replies: 15
Views: 1601

Re: Are hybrid approaches stronger than pure nnue?

Why hybrid earns elo? If it is only because of speed then I wonder if it holds at longer time control. If it is because nnue is relatively bad in positions when one side has a big advantage then maybe it is better to improve nnue and teach it based on scoring system that not all wins are the same a...
by Michel
Wed Jul 07, 2021 9:12 am
Forum: Computer Chess Club: General Topics
Topic: Time to rethink what Vasik Rajlich has done?
Replies: 125
Views: 15676

Re: Time to rethink what Vasik Rajlich has done?

IT WAS NEVER ASKED [!] How is that relevant? Well, it's you who said that we can't know. We can. Chris accused him of having no intention to give it, even when asked. That is a pretty damning accusation. As a commercial, someone who took the risk to make a living in the paper thin market of chess p...
by Michel
Tue Jul 06, 2021 7:31 pm
Forum: Computer Chess Club: General Topics
Topic: Time to rethink what Vasik Rajlich has done?
Replies: 125
Views: 15676

Re: Time to rethink what Vasik Rajlich has done?

IT WAS NEVER ASKED [!] How is that relevant? Well, it's you who said that we can't know. We can. Chris accused him of having no intention to give it, even when asked. That is a pretty damning accusation. As a commercial, someone who took the risk to make a living in the paper thin market of chess p...
by Michel
Tue Jul 06, 2021 8:17 am
Forum: Computer Chess Club: General Topics
Topic: Time to rethink what Vasik Rajlich has done?
Replies: 125
Views: 15676

Re: Time to rethink what Vasik Rajlich has done?

I really don't understand why people keep coming back to this. The ICGA clearly thought there was enough circumstantial evidence (decompiling, move similarity, variable names, Strelka, ...) for the claim that Rybka was written by modifying Fruit. In reaching this conclusion they no doubt also took V...
by Michel
Tue Jul 06, 2021 5:12 am
Forum: Computer Chess Club: General Topics
Topic: Should Vas be in the chess programmers hall of fame?
Replies: 5
Views: 973

Re: Should Vas be in the chess programmers hall of fame?

He also invented very aggressive LMR (far beyond the 1 or 2 plies that were the norm before).

When this became public knowledge through Ippolit, other engines (including SF) became hundreds of Elo stronger.
by Michel
Tue Jul 06, 2021 5:06 am
Forum: Computer Chess Club: General Topics
Topic: Time to rethink what Vasik Rajlich has done?
Replies: 125
Views: 15676

Re: Time to rethink what Vasik Rajlich has done?

Another piece of evidence (not based on decompiling) was the very high move similarity between Fruit 2.1 and Rybka 1.0b as reported here

https://www.sciencedirect.com/science/a ... 2113000177

(I am sure this can also be found by searching old talkchess posts).
by Michel
Tue Jul 06, 2021 4:13 am
Forum: Computer Chess Club: General Topics
Topic: Time to rethink what Vasik Rajlich has done?
Replies: 125
Views: 15676

Re: Time to rethink what Vasik Rajlich has done?

I always thought that the main smoking gun was Strelka which was based on Fruit and which Vas himself claimed to be a clone of Rybka.