MYG
Moderators: hgm, Rebel, chrisw
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: MYG
I agree, but first, it is too early to say (less than 400 games), second, I am divided between Houdini and Shredder. If it is Houdini 6, then the result against Houdini 5 should have been more drastic (but only 20 some games were played between the two).JJJ wrote:Timing and strenght fit more with Houdini 6 than any other engine. So to me it is clearly Houdini 6 and he is most likely stronger than Stockfish dev.
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: MYG
It at least doesn't seem to use Contempt, so it's probably not Komodo. The linear regression sees no significant negative slope (higher performance with weaker engines). The total performance as of now is at higher value than that shown by the Ingo's page.Laskos wrote:I agree, but first, it is too early to say (less than 400 games), second, I am divided between Houdini and Shredder. If it is Houdini 6, then the result against Houdini 5 should have been more drastic (but only 20 some games were played between the two).JJJ wrote:Timing and strenght fit more with Houdini 6 than any other engine. So to me it is clearly Houdini 6 and he is most likely stronger than Stockfish dev.
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: MYG
Add about 20 ELO points to the total performance (at the end of the list) shown on that page, if the final rating will be correctly done with Ordo. As of now, even better than Stockfish dev. Finally pulled heavily ahead of Houdini 5, so might be indeed Houdini 6. I am not discounting yet Shredder and new Fritz (look at the performance against Fritz as of now). Still too few games.JJJ wrote:So fast to do the stats Kai, as usual.
Now engine X is doing something equal to Stockfish dev. Maybe still a little better.
-
- Posts: 41455
- Joined: Sun Feb 26, 2006 10:52 am
- Location: Auckland, NZ
Re: MYG
Robert doesn't normally pre-release versions to anybody, so I'd be surprised if it's Houdini 6.JJJ wrote:Timing and strenght fit more with Houdini 6 than any other engine. So to me it is clearly Houdini 6 and he is most likely stronger than Stockfish dev.
gbanksnz at gmail.com
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: MYG
I am divided between Houdini and Shredder, would give them pretty similar chances. After some 670 games, it is too early, but a slight slope opposite to positive Contempt started to appear, which is similar more to Shredder and Stockfish than to Houdini, but things might have changed, and the statistical noise is too big. Komodo is ruled out.Werewolf wrote:My guess:
Houdini 6 - 80% likely
Shredder 14 - 10% likely
Fritz 16 - 5% likely
HIARCS 15 - 1% likely
All the rest - 4% likely
-
- Posts: 165
- Joined: Tue Dec 02, 2014 1:29 am
Re: MYG
Laskos wrote:Add about 20 ELO points to the total performance (at the end of the list) shown on that page, if the final rating will be correctly done with Ordo. As of now, even better than Stockfish dev. Finally pulled heavily ahead of Houdini 5, so might be indeed Houdini 6. I am not discounting yet Shredder and new Fritz (look at the performance against Fritz as of now). Still too few games.JJJ wrote:So fast to do the stats Kai, as usual.
Now engine X is doing something equal to Stockfish dev. Maybe still a little better.
no suprice here. houdini 6 will be above st and komodo
there is a logic formula:
houdart is a good programer+st is open source=houdini 6 can be above st and komodo
the main negative part of the open source st is that any good programer can improve it
and make it closed source, thats why NEVER stockfish will be ahead with big elo difference.
the only damage will be for komodo sales.