Stockfish 13 crushed and other 1500 elo engines

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

lkaufman
Posts: 6279
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA
Full name: Larry Kaufman

Re: Stockfish 13 crushed and other 1500 elo engines

Post by lkaufman »

mvanthoor wrote: Sun Dec 19, 2021 5:44 pm
ydebilloez wrote: Mon Sep 20, 2021 11:03 am When I play against human players of around 1500, they don't get certain concepts such as i.e. passed pawn or weak vs strong bishops. They blunder pieces away from time to time but most often overlook moves in a complex situation or cannot handle attacks on different fronts at the same time... but not the kind of errors Stockfish makes.
Elo-comparisons are difficult. I've often played against engines in the 1200-1400 Elo CCRL range, and at around 1500 Elo, they become hard to beat for me. Even when playing against many of the 1200-1300 Elo engines I often have the feeling that they are MUCH stronger than a 1200 Elo human.

Someone once posted that the CCRL list is compressed, and calibrated at 2800 Elo. The formula to convert from CCRL Elo to FIDE Elo would be (according to that post) FIDE = (CCRL * 0.7) + 840.

That would indeed set the calibration to 2800 Elo, because (2800 CCRL * 0.7) + 840 = 2800 FIDE.

If that is (still) correct, a 1200 CCRL engine would be 1680 Elo. As UCI_LimitStrength is intended to simulate human / FIDE Elo, it would explain why your 1200 CCRL defeats Stockfish set to UCI_LimitStrength = 1500. Try it at 1680 - 1700.
The problem with this explanation is that Stockfish based their UCI_Elo levels on CCRL ratings, not on any human tests. Komodo Dragon does try to match Elo levels with humans, but he is talking about Stockfish so that is not relevant.
Komodo rules!
purechess
Posts: 83
Joined: Wed Nov 28, 2018 1:28 pm
Full name: Heinrich Pulliter

Re: Stockfish 13 crushed and other 1500 elo engines

Post by purechess »

Also consider that Stockfish 14.1 has an improved UCI_Elo mechanism.
lkaufman
Posts: 6279
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA
Full name: Larry Kaufman

Re: Stockfish 13 crushed and other 1500 elo engines

Post by lkaufman »

purechess wrote: Sun Dec 19, 2021 7:09 pm Also consider that Stockfish 14.1 has an improved UCI_Elo mechanism.
Can you explain how UCI_Elo has been improved in SF14.1, or point to a post that explains it?
Komodo rules!
purechess
Posts: 83
Joined: Wed Nov 28, 2018 1:28 pm
Full name: Heinrich Pulliter

Re: Stockfish 13 crushed and other 1500 elo engines

Post by purechess »

lkaufman wrote: Sun Dec 19, 2021 7:20 pm
purechess wrote: Sun Dec 19, 2021 7:09 pm Also consider that Stockfish 14.1 has an improved UCI_Elo mechanism.
Can you explain how UCI_Elo has been improved in SF14.1, or point to a post that explains it?
https://github.com/official-stockfish/S ... /pull/3769
User avatar
mvanthoor
Posts: 1784
Joined: Wed Jul 03, 2019 4:42 pm
Location: Netherlands
Full name: Marcel Vanthoor

Re: Stockfish 13 crushed and other 1500 elo engines

Post by mvanthoor »

lkaufman wrote: Sun Dec 19, 2021 6:35 pm The problem with this explanation is that Stockfish based their UCI_Elo levels on CCRL ratings, not on any human tests. Komodo Dragon does try to match Elo levels with humans, but he is talking about Stockfish so that is not relevant.
Also, Stockfish 11 has been recalibrated. At the same skill level, SF11 and newer are much stronger than SF10 was. Therefore I prefer SF10 to play against (on the Raspberry Pi) because it has more usable skill levels for me. (I'm going to try and see if I can find and compile some open source engines in the 1200-2000 CCRL range to play against.)
Author of Rustic, an engine written in Rust.
Releases | Code | Docs | Progress | CCRL