When will we see HOUDINI in official tournaments?

[S.Taylor]

Because Critter is a huge real threat to knock down Houdini from the top of the rating lists....

Personally I'll be so happy when this happens
Dr.D

Why would this make you happy? Wouldn't it make you happy if Houdini got stronger and stronger, and so uttlerly burries Rybka?

You used to only want to see Rybka overtaken. Houdini has already done that for you!

[EmilV]

Thanks.
The Similarity test could be trustworthy for commercial engines.
But it is no problem for a special CSVN-tuned engine to recognize beeing tested…
Isn’t it, Ed?
So the CSVN rule seems to be doubtful.

[Dr.Wael Deeb]

Because Critter is a huge real threat to knock down Houdini from the top of the rating lists....

Personally I'll be so happy when this happens
Dr.D

Why would this make you happy? Wouldn't it make you happy if Houdini got stronger and stronger, and so uttlerly burries Rybka?

You used to only want to see Rybka overtaken. Houdini has already done that for you!

I don't like Houdini's author,simple as that....

After the release of Rybka 4.1 I've kinda lost that feeling of cheated customer and drew a line to that story forever....
Dr.D

[Dr.Wael Deeb]

Hi Ed,

If... Robbolito is a clone (ipolit) and not allowed, then have a look at my similarity test... what shall we do with Critter ??

sim version 3
------ RobboLito 0.09 x64 (time: 100 ms scale: 1.0) ----
65.16 Houdini 1.5 x64 (time: 100 ms scale: 1.0)
64.85 Strelka 5.1 (time: 100 ms scale: 1.0)
63.06 Critter 1.4 64-bit (time: 100 ms scale: 1.0)
61.71 Rybka 3 1-cpu (time: 100 ms scale: 1.0)

Oke, Houdini is well known as a Robbolito clone but Critter is very very close too !

Best
Johan

Yes, the course Critter has chosen is odd, see:

Code: Select all

 1) Critter 0.34 (time: 100 ms  scale: 1.0)
 2) Critter 0.36 (time: 100 ms  scale: 1.0)
 3) Critter 0.40 (time: 100 ms  scale: 1.0)
 4) Critter 0.42 (time: 100 ms  scale: 1.0)
 5) Critter 1.2  (time: 100 ms  scale: 1.0)
 6) Critter 1.4  (time: 100 ms  scale: 1.0)
 7) Houdini 1.0  (time: 100 ms  scale: 1.0)
 8) Houdini 1.5  (time: 100 ms  scale: 1.0)

        1     2     3     4     5     6     7     8
 1.  ----- 75.14 40.37 39.22 35.46 34.04 36.77 34.11
 2.  75.14 ----- 42.24 40.23 36.20 34.73 37.18 34.80
 3.  40.37 42.24 ----- 51.71 41.93 40.52 42.29 40.76
 4.  39.22 40.23 51.71 ----- 44.72 43.35 45.79 43.85
 5.  35.46 36.20 41.93 44.72 ----- 61.23 61.47 60.95
 6.  34.04 34.73 40.52 43.35 61.23 ----- 60.91 64.52
 7.  36.77 37.18 42.29 45.79 61.47 60.91 ----- 63.27
 8.  34.11 34.80 40.76 43.85 60.95 64.52 63.27 -----

It starts as a very orginal engine but since version 1.2 there is this link with Houdini. According Robert because Richard RE'ed Houdini (see a couple of pages up).

Most probably it is because Critter >=1.2 uses public piece-square tables from Ippolit.

Richard , i have a question for you.
As houdart accuses Critter 1.2 and later of REed Houdini 1.5a.
so what is reality ? As I think myself as one of the best wellwisher of critter , little confusion .

Will you pls clarify ?

Actually it's simple....

Robert is punching Richard under the belt whenever he has the opportunity....why

Even more simple....

Because Critter is a huge real threat to knock down Houdini from the top of the rating lists....

Personally I'll be so happy when this happens
Dr.D

[Sedat Canbaz]

[LudiBuda]

Exactly. It should be trivial to modify the engine to pass the test.
You can write the code that detects the test is being run and there are many other ways too.

[geots]

The more relevant issue is if your current engine stays below the required 60% similarity, which is already pretty tolerant.

LOL. What a joke.
I notice that the first CSVN tournament (Leiden) that is supposed to use this "60% similarity" rule has accepted Critter as participant.
Surely you're aware that Critter 1.2 and 1.4 are nearly indistinguishable from Houdini 1.5a in the similarity dendogram, and that this is not by accident, but by acknowledged RE-ing of Houdini 1.5a.

Robert, I have immense respect for you as a programmer ( houdini 1.5a)
But making serious accusation to a respected person and questioning the originality of critter is waste of time .
will anyone be agree in the whole world in the subject that critter is REed from Houdini 1.5a ? except you.

Poll " Critter is a REed engine from Houdini"

No.................1000000000000000000
Yes................1 (it is you)

The only advice I would give you since you are new here: pick your battles extremely carefully. Stay away from those you can't win. And calling Robert out- extremely bad choice. Besides being one of the 2 best programmers I have ever seen, he won't make a statement he can't back up. One of 2 things- you misunderstood what he was saying OR you understood it, and you are wrong. Maybe my advice and a buck will get you a coke, but it is still my advice.


Hope you enjoy the forum,

gts

[michiguel]

Exactly. It should be trivial to modify the engine to pass the test.
You can write the code that detects the test is being run and there are many other ways too.

I do not know what the organization is planning to do, but if it is done properly and the positions are changed, it is virtually impossible to cheat on this test. You can even pick 1000 positions at random from comp-comp games and the behavior is maintained.

Miguel

[hgm]

Exactly. It should be trivial to modify the engine to pass the test.
You can write the code that detects the test is being run and there are many other ways too.

Not really. Because the test can be run on positions from your own tourney games.

[LudiBuda]

I can just add code to play randomly picked good move when the new game starts. This is done just for the first move which should be enough to fool the similarity test.