New Houdini

[Leo]

Thanks. First time I have heard of that.

[APassionForCriminalJustic]

I've been watching some of Houdini's TCEC games. They are impressive. The engine seems to have a punchy agressive style too. I wouldn't be surprised if it isn't close to parity with Stockfish.

Steve

Your impressions seem to be confirmed as of now. Not only Houdini leads and has most wins, its wins are on average faster (less moves):

Code: Select all

         SF    Hou    Ko
         25    30     26    <-- number of wins  
Length            
21-30     1     0      2
31-40     5     7      1
41-50     8     9     10 
51-60     2     9      5
61-70     6     2      3
71-80     2     2      1
81-100    1     1      4
=========================
Won      52   50.5    55.4
Game
Length

I do not follow TCEC, can it not be that Houdini's opponents resign earlier?

No, simply Houdini plays better chess, a bit better than SF and Komodo, but nothing earth-shaking.

I bet after version 5 release H will top most rating lists 20-30 elo above SF 8 (which should be released about the same time) with contempt set on, about fully equal with no contempt.

I do not know about TCEC final though, everything will depend on the openings used, with shorter openings and more closed ones I would favour slightly Houdini, If the 8-move GM suite is used, which would be a major mistake to me, as will only bring more draws because of exhausted positions and less interesting play, especially if there are a lot of Sicilians and complex open positions, SF will undoubtedly be on top.

In any case we are heading into a very interesting final.

Houdini plays better chess?! Lmao. Holy crap man, you are something else. All of you so-called "experts" of computer chess think that the final will be close - or "interesting" (whatever that means)? Haha. That version of Houdini in Rapids right now lost two games versus Stockfish in stage three. They only played like six games or something to that effect in that stage. Rapids means nothing; head to head match ups are what matter. Furthermore, openings don't matter much either. Stockfish doesn't need unbalanced openings or tactical prowess to win. It's just a better engine. There is NO reason to suspect a close superfinal, and it won't be. I can't wait to see how Stockfish wins the superfinal with ease, so we can see what some of you have to say.

Even if Houdini was somehow close to Komodo 10.1 in strength and Elo, it would not even be close to enough in strength to even be that competitive. The real issue here is that we do not know what Houdart's private tests show in terms of Elo differential between Houdini 4 and the latest Houdini development version. We do know that he has gained approximately 25 Elo versus the development version participating presently in Rapids via TCEC. All that said, we know that Stockfish has also gained in Elo significantly since the version in stage three.

You think that Houdini 5 will be of equal strength to Stockfish 8? Lol. Houdini would need like 200+ Elo (since Houdini 4) for that - literally. The last time Houdini used to beat Stockfish on the rating lists was versus Stockfish 4. You can have an opinion; I have common sense. Houdini is definitely stronger than Houdini 4. But you are pushing things into a world of imagination, and fantastical thought. We know for a fact that Houdini is definitely the weaker engine. And we know that it is by a significant amount. 50 Elo seems a little bit low; something like 75 to 100 seems more logical and appropriate. I will say this: I predict that Stockfish will win TCEC with a score of approximately 20-5-75. This would be +53 Elo for Stockfish. Perhaps Dan's 50 Elo prediction has some merit. Nevertheless, the truth in all of this will come to light by December I would imagine.

Houdini 4 versus Stockfish 4 64-bit 4CPU (+18−12=38)
= = = = = 1 0 1 = = 0 = 0 = 0 = 1 0 = = = 1 = = 1 = = 0 = = 1 = 1 = 1 = 0 = = = = 1 = 1 1 1 1 = 1 1 = 0 1 = 1 = = 0 = = 0 = = 0 = 1 0 =

Houdini 4 versus Stockfish DD 64-bit 4CPU (+17−29=52)
= 0 = = = 1 0 = 0 0 0 0 = = = 1 = = = 1 1 1 0 = 0 0 = 0 = = 1 0 0 = 1 = = 0 1 1 = 0 0 = 1 = 0 0 = = = = 0 = 1 = 0 = = 0 1 = = = 0 1 = = 0 = = = = = 0 = = 1 1 = 0 = = 1 0 = = = = 0 0 0 0 1 = = = =

PS We should not forget that Stockfish had the quickest victory thus far in Rapids - and it is undefeated since stage three. Stockfish is absolutely dominating in all ways.

By the way, I am an avid supporter of Houdini. I will be purchasing Houdini 5; Houdart is doing an incredible job with his engine. He is a very talented author, and computer-chess programmer. I simply believe that this is Stockfish's time and year to dominate.

[Steve Maughan]

Adam,

Houdini plays better chess?! Lmao. Holy crap man, you are something else. All of you so-called "experts" of computer chess think...

You may be we right, you may be wrong; but I don't care for your manners,

Steve

[Laskos]

Houdini plays better chess?! Lmao. Holy crap man, you are something else. All of you so-called "experts" of computer chess think that the final will be close - or "interesting" (whatever that means)? Haha. That version of Houdini in Rapids right now lost two games versus Stockfish in stage three. They only played like six games or something to that effect in that stage. Rapids means nothing; head to head match ups are what matter. Furthermore, openings don't matter much either. Stockfish doesn't need unbalanced openings or tactical prowess to win. It's just a better engine. There is NO reason to suspect a close superfinal, and it won't be. I can't wait to see how Stockfish wins the superfinal with ease, so we can see what some of you have to say.

Even if Houdini was somehow close to Komodo 10.1 in strength and Elo, it would not even be close to enough in strength to even be that competitive. The real issue here is that we do not know what Houdart's private tests show in terms of Elo differential between Houdini 4 and the latest Houdini development version. We do know that he has gained approximately 25 Elo versus the development version participating presently in Rapids via TCEC. All that said, we know that Stockfish has also gained in Elo significantly since the version in stage three.

You think that Houdini 5 will be of equal strength to Stockfish 8? Lol. Houdini would need like 200+ Elo (since Houdini 4) for that - literally. The last time Houdini used to beat Stockfish on the rating lists was versus Stockfish 4. You can have an opinion; I have common sense. Houdini is definitely stronger than Houdini 4. But you are pushing things into a world of imagination, and fantastical thought. We know for a fact that Houdini is definitely the weaker engine. And we know that it is by a significant amount. 50 Elo seems a little bit low; something like 75 to 100 seems more logical and appropriate. I will say this: I predict that Stockfish will win TCEC with a score of approximately 20-5-75. This would be +53 Elo for Stockfish. Perhaps Dan's 50 Elo prediction has some merit. Nevertheless, the truth in all of this will come to light by December I would imagine.

Houdini 4 versus Stockfish 4 64-bit 4CPU (+18−12=38)
= = = = = 1 0 1 = = 0 = 0 = 0 = 1 0 = = = 1 = = 1 = = 0 = = 1 = 1 = 1 = 0 = = = = 1 = 1 1 1 1 = 1 1 = 0 1 = 1 = = 0 = = 0 = = 0 = 1 0 =

Houdini 4 versus Stockfish DD 64-bit 4CPU (+17−29=52)
= 0 = = = 1 0 = 0 0 0 0 = = = 1 = = = 1 1 1 0 = 0 0 = 0 = = 1 0 0 = 1 = = 0 1 1 = 0 0 = 1 = 0 0 = = = = 0 = 1 = 0 = = 0 1 = = = 0 1 = = 0 = = = = = 0 = = 1 1 = 0 = = 1 0 = = = = 0 0 0 0 1 = = = =

PS We should not forget that Stockfish had the quickest victory thus far in Rapids - and it is undefeated since stage three. Stockfish is absolutely dominating in all ways.

By the way, I am an avid supporter of Houdini. I will be purchasing Houdini 5; Houdart is doing an incredible job with his engine. He is a very talented author, and computer-chess programmer. I simply believe that this is Stockfish's time and year to dominate.

Calm down man. The evidence is too sparse to have very reliable predictions about superfinal. If Robert H improved Houdini by 100 points, it is not ruled out he improved it by 200 ELO points.

Better take Ordo, run TCEC_Stage_3 PGN file, with the -J switch which shows CFS (confidence for superiority) between successive entries. I also included the first, discarded in TCEC game between Houdini and Komodo (where Houdini won).

Code: Select all

   # PLAYER              : RATING  ERROR    POINTS  PLAYED     (%)   CFS(next)
   1 Stockfish 160716    : 3332.9   47.8      39.0      56    69.6      89    
   2 Houdini 200716      : 3291.0   43.1      36.5      57    64.0      75    
   3 Komodo 10.1         : 3269.2   41.5      34.5      57    60.5      99    
   4 Fire 5              : 3199.3   37.6      28.5      56    50.9      98    
   5 Andscacs 0.872b     : 3137.9   40.0      23.5      56    42.0      72    
   6 Jonny 8             : 3119.1   41.4      22.0      56    39.3      73    
   7 Gull 3              : 3100.0   42.4      20.5      56    36.6      50    
   8 Rybka 4.1           : 3100.0   45.6      20.5      56    36.6     ---    

White advantage = 85.92 +/- 13.10
Draw rate (equal opponents) = 92.63 % +/- 4.81

We see that it is 89% likely SF is stronger than Houdini, and 75% likely Houdini is stronger than Komodo in TCEC LTC conditions.

Better now?

[shrapnel]

heh heh............I was waiting for your Bombs to rain, Adam !
Only thing is if by chance Houdini beats Stockfish in the Superfinal, those bombshells will turn into Eggs...and rain in the opposite direction.
Of course, we all know that Stockfish is the Favorite, but many, like me, like to support the underdog ; though not as vehemently as you support Stockfish.
Anyway, it should be interesting.

[Alexander Schmidt]

It's not easy to reach the same level as the Tactical mode of Houdini 3 and Houdini 4.

New codebase?

[JJJ]

It will be so fun to see Houdini win and Stockfish not while it is at his peak. The bonus of course would be to see the hate of Adam fade aways

[corres]

We know nothing about the opening book for TCEC final.
A good opening book enhances the power of an engine ~100 Elo, whilst between Stockfish and Houdini is some tens Elo difference.
So the result of the final depends not only Stockfish and Houdini but the opening book will be used.
It is very pity that there is no standard for opening books to test.
Because of this the book used is a toy (or a weapon?) in the hands of the testers.
Note: A tests without opening book is worthless. Engines are made for middle game of chess with some endgame knowledge.

[stavros]

i think the point is why houdart doesnt says the h5-h4 difference as he did it in the past with houdini versions before releases whats wrong? looks weird isn't it?
chess cold war espionage:P

[corres]

Please ask Mr.Houdart about it.
He is the only man who knows the answer.
May be the organizer of TCEC does it also.