MYG

[Houdini]

That's a good progress Robert. Will you try to get more elo before an official release ?

The Houdini 6 release will be soon.
I'm happy to see that you're disappointed with Houdini leading the rating list by only 30 points .

[Uri Blass]

Kai Laskos: Is new Houdini on par or superior to SF9 Dev?

At this time control on one core Houdini seems stronger than the best SF dev. But TCEC is a bit different, hard to say.

On TCEC Houdini 5 was beaten by 54% against Stockfish 8( short sample )
On fast.gm Houdini was beaten 51% on 1 core and 1 hour time control by Stockfish 8

Maybe we could guess Stockfish 8 is slighty better than Houdini 5 in TCEC condition.

So, this Houdini win barely Stockfish 8, probably loose against Stockfish dev and has to play in TCEC condition where he might be slighty below as well.

So the goal to be the absolute best is not totally achevied here. Still a good update, because it seems closer than Stockfish than last time.

I do not see the basis for it.

I think that Claiming that stockfish is relatively better than houdini6 at long time control has no basis.

Stockish is optimized for bullet and not for long time control in the stockfish framework and if in the past they at least used (-4.5,1.5) at short time control to give a better chance to changes that are not so good at faster time control than bullet then today they even do not do it.

[Laskos]

Thank you very much Ingo. Impressive results, and my ELO predictions were pretty accurate (aside the last, where I missed a couple of ELO points). Congratulations to Robert!

You made some pretty astute predictions.
But maybe not sufficiently taking into account the error margins on the individual results - with 200 games it must be like ±30 Elo. With only 2 or 3 data points on the right side of the diagram, the slope of your line is pretty uncertain.

A slope like Komodo has due to contempt would be detectable even with this sparse data:



I am pretty confident that if Houdini 6 used here Contempt, it was very mild compared to that of Komodo.

[JJJ]

Houdini is leading but will loose in direct encounter against Stockfish dev, and I want your program to win

[JJJ]

Kai Laskos: Is new Houdini on par or superior to SF9 Dev?

At this time control on one core Houdini seems stronger than the best SF dev. But TCEC is a bit different, hard to say.

On TCEC Houdini 5 was beaten by 54% against Stockfish 8( short sample )
On fast.gm Houdini was beaten 51% on 1 core and 1 hour time control by Stockfish 8

Maybe we could guess Stockfish 8 is slighty better than Houdini 5 in TCEC condition.

So, this Houdini win barely Stockfish 8, probably loose against Stockfish dev and has to play in TCEC condition where he might be slighty below as well.

So the goal to be the absolute best is not totally achevied here. Still a good update, because it seems closer than Stockfish than last time.

I do not see the basis for it.

I think that Claiming that stockfish is relatively better than houdini6 at long time control has no basis.

Stockish is optimized for bullet and not for long time control in the stockfish framework and if in the past they at least used (-4.5,1.5) at short time control to give a better chance to changes that are not so good at faster time control than bullet then today they even do not do it.

Still, Houdini won only by 52% against Stockfish 8 at this time control. Maybe Stockfish dev would do the same result instead his 54% in bullet. That's still very close

[Uri Blass]

Houdini is leading but will loose in direct encounter against Stockfish dev, and I want your program to win

We do not know if houdini will lose a direct encounter against stockfish dev.

[Nordlandia]

Robert: one question -> did you look upon Nalimov crash?

[Uri Blass]

Houdini is leading but will loose in direct encounter against Stockfish dev, and I want your program to win

We do not know if houdini will lose a direct encounter against stockfish dev.

I am not sure even if stockfish developement is expected to do better than 116-104 against stockfish8 in IPON conditions and 116-104 may be a lucky run for stockfish8.

Note that I expect stockfish developement to do worse in longer time control against stockfish8 because of getting more draws and IPON is significantly longer time control than the time that people test stockfish in the stockfish framework.

[Houdini]

Robert: one question -> did you look upon Nalimov crash?

The Nalimov code is legacy (meaning that I'm not going to tinker too much with it) and in my opinion quite obsolete since Syzygy is available.
If you encounter crashes, my best advice would be to reduce the number of threads or increase the probing depth.

[Modern Times]

Houdini is leading but will loose in direct encounter against Stockfish dev, and I want your program to win

We do not know if houdini will lose a direct encounter against stockfish dev.

Indeed. I ran some test games recently with Stockfish Dev and I wasn't convinced that there was much improvement since SF8. OK it wasn't thousands of games, but I'd seen enough to tell me that I wouldn't bother running more.