Not to be argumentative, but regarding these 20 games, you also said “With only 20 games, the CI is massive and any result barring an extreme result (my ephasis), is meaningless.”connor_mcmonigle wrote: ↑Sat Sep 18, 2021 5:25 pm
Modern chess engines behave like chaotic systems as slight perturbations to initial conditions can evolve into very different behavior. Many factors can lead to these perturbations: OS using time for some background process, thermals affecting clock speed and nps, etc.
The reason I asked about the results is precisely because I think 4 wins vs 1 loss in a short 20 game match where the engines are so very similar otherwise...IS an 'extreme result'.
Ultimately, I think you just disagree with amchess on the merits of STC vs LTC and his “characteristics concept” approach vs what I refer to as ‘the wisdom of big number’ (LOTS of data pointing towards a mean) found in normal Stockfish testing....or Komodo / Dragon, etc.
In many things in life I believe in the 'wisdom' of big numbers. Classical chess however – not talking about the variants, which I do not care about - where similar tabias are normally played to and 'essentially' where the game starts, I can see might well lean more towards the ‘wisdom’ found in the ShashChess results, if indeed these positions at LTC are as representative of normal '10 center types/characteristcs' as amchess claims.