Lyudmil Tsvetkov wrote:did they cancel TCEC final?
No, merely postponed until Stockfish Development Version gets a few more ELO.
Top Stockfish Programmer, Stéphane Nicolet, is hard at work churning out new Patches.
Powers-That-Be ( who DON'T want Houdini to win) are scared that if the Super-Final is played right now, Houdini will simply THRASH Stockfish, thereby causing them severe Acidity, Indigestion and Dyspepsia ! So, I hereby declare HOUDINI to be the Super-Final Winner by Default, as opponent didn't turn up !
Jouni wrote:So after TCEC rapid Houdini is improved by 214 ELO from H4!! Stage 3 says "only" by 153 ELO. And RH said latest version is +30 better. Before superfinal another 30? Considering this rapid I agree 100% this chat comment. "This rapid tournament, with this formula, was a mistake and a useless waste of time... And it hurted Sf chances of winning sf (no important patches in last two months), it helped Houdart catching up SF... What a shame."
Elo doesn't grow on trees. Have you not followed the development of Houdini over the years? Robert was averaging 50 Elo per year. Does it seem realistic, seriously, to gain 30 Elo versus the development version in TCEC - and then another 30 Elo just before the superfinal? Sorry but that's completely ludicrous. And you forgot to mention that Stockfish has gained WELL in excess of 20 Elo since the last version that had played in stage three. In fact two new big functional patches had just been committed like 10 hours ago. I don't see Houdart as a threat to Stockfish's dominance frankly. If anything it was Stockfish development that took Houdini's first-place dominance - and now you insinuate that Stockfish's chances are hurt because of Rapids? Come on. Ha-ha.
And your claim about 200+ Elo gained over Houdini 4, where is the proof of this? Pulling rabbits out of hats are you? I will believe it when I see it. Maybe something like 100 to 125 feels more appropriate. Even something like that would be record shattering for Houdini frankly versus say Houdini 2 to 3, or Houdini 3 to 4.
Hey Adam bro, what will you do/say if by chance Houdini wins the Super-Final ?
Do you promise to have an Apoplectic Fit, or something else ?
Just curious !
He could gain 200 Elo in a week just by studying what SF has done.
It's open source, after all.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
FriedmannC wrote:I don't think Houdini 5 will be as strong as Stockfish 8, maybe 25 ELO points above Komodo 10.2 but not even close to the latest Stockfish version
I think it will be weaker than Stockfish, the same strength as Stockfish, or stronger than Stockfish. Trichotomy, and all that. Somewhere there is a mathematician who thinks this is funny.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
FriedmannC wrote:just an estimation, no need for you to come across with these kind of ,,assertions''
I can assure you with 100% accuracy that I was correct. That is the beauty of math. My real opinion is that nobody knows how strong Houdini is. It seems likely that it is stronger than Komodo which is within 30 Elo of Stockfish. So I guess it will take a long time to know the real strength of Houdini, but in the end it will be pretty close one way or the other.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
I guess you're right, some kind of Stockfish 8 + Komodo 10.2 / 2,
arithmetic mean And I have this feeling that Houdini 5 ( the official, not the TCEC final version ) will be delayed quite much....
And I have this feeling that Houdini 5 ( the official, not the TCEC final version ) will be delayed quite much....