Why don't strong engines enter WCCC?

[kranium]

One thing more,
Of course the originality plays a big role... !
But however 'strength' is more important than 'originality' in chess!

That's why:
- 99% of the users preferred Rybka, Houdini, Stockfish, Komodo, Gull etc...for their tests in the latest 5 years!!

So what is more important ??

agreed

it's well-known that Houdini started from Ippolit/Robbolito,
early versions of Gull were based on Strelka and newer ones started from Ippolit...
and there's plenty of 'evidence' to show it.

strong yes, but original?

[kranium]

Indeed, very unfortunate. I rejoiced for nothing.

I don't see your point, if indeed you have one. It d.

Actually I am meaning simply:
* If Rybka will be counted as a clone or derivative, then:
- Stockfish, Houdini, Gull, Komodo and plus many many engines should be counted in this group too !!

Otherwise,
It will be injustice...

there was definitely Fruit in Rybka 1.0 beta...
but as far as I know, nothing has been found in SF or Komodo yet?

[Sedat Canbaz]

One thing more,
Of course the originality plays a big role... !
But however 'strength' is more important than 'originality' in chess!

That's why:
- 99% of the users preferred Rybka, Houdini, Stockfish, Komodo, Gull etc...for their tests in the latest 5 years!!

So what is more important ??

agreed

it's well-known that Houdini started from Ippolit/Robbolito,
early versions of Gull were based on Strelka and newer ones started from Ippolit...
and there's plenty of 'evidence' to show it.

strong yes, but original?

Hello dear Norman


Before me too,
I had some suspicious about Gull's originality...

But I've changed my view...after the confirmation by some programmers:
http://www.talkchess.com/forum/viewtopi ... ight=sedat

And it seems, the programmer of Gull (Vadim Demichev) cleaned Ippolit code from his engine
Of course I am not 100 % sure about this is a right way ....

Btw, as I mentioned before (about SCCT new conditions),
Gull is as participant, due to Gull suits to SCCT rules

As we see,
Gull 3 is +100 Elo over Rybka 3

Code: Select all

CCRL (40/4)
Gull 3 64-bit 4CPU	 3277 Elo
Fire 3.0 64-bit 4CPU  3217 Elo
Rybka 3 64-bit 4CPU	3163 Elo 

For example,
If you add +50 Elo more then Fire can be as SCCT participant too


Best,
Sedat

[Sedat Canbaz]

Indeed, very unfortunate. I rejoiced for nothing.

I don't see your point, if indeed you have one. It d.

Actually I am meaning simply:
* If Rybka will be counted as a clone or derivative, then:
- Stockfish, Houdini, Gull, Komodo and plus many many engines should be counted in this group too !!

Otherwise,
It will be injustice...

there was definitely Fruit in Rybka 1.0 beta...
but as far as I know, nothing has been found in SF or Komodo yet?

It seems you missed to see....
Please check my published recent sim results (I posted especially on this thread )

http://www.talkchess.com/forum/viewtopi ... =&start=20

You can compare similarities and relationships...

As I mentioned before,
In my eyes all the mentioned engines (Rybka, Stockfish, Komado, Houdini, Gull etc...) are not based 100% on own ideas
But they should be counted as original work, because they managed to improve their engines with own original ideas!!

Otherwise...
It will be injustice if we say only Rybka is a clone
And Stockish or Komodo are 100 % original work!!

Btw,
You need to clean Ippoolit code too
The similarity percentage is too high )!

[kranium]

Sedat-

I find your approach to this matter well-founded and quite reasonable...
at least you have established sound and well-defined criteria (which have been woefully absent in this field for years).

I can assure you the Fire is getting stronger, and I work on rewriting it and improving it's uniqueness daily.
I hope I can meet or exceed your criteria within the next year or two.

Best Regards-
Norm

[Sedat Canbaz]

Sedat-

I find your approach to this matter well-founded and quite reasonable...
at least you have established sound and well-defined criteria (which have been woefully absent in this field for years).

I can assure you the Fire is getting stronger, and I work on rewriting it and improving it's uniqueness daily.
I hope I can meet or exceed your criteria within the next year or two.

Best Regards-
Norm

Dear Norman,

Thanks for your understanding and kind words!

I wish you good luck,
Sedat

[hgm]

Please give us some credit..... Of course we know the difference.

I don't doubt that a moment for you, but I was addressing Sedat, who shows by his remarks that he does not understand this at all.

If Stockfish has gained strength because of the release of Ippolit, it is certainly not by including Ippolit code. We can be sure about that, as both the sources of Stockfish and Ippolit are published. Yet Sedat keeps insisting that everyone does the same as what Vas is accused of...

* If Rybka will be counted as a clone or derivative, then:
- Stockfish, Houdini, Gull, Komodo and plus many many engines should be counted in this group too !!

[Sedat Canbaz]

Please give us some credit..... Of course we know the difference.

I don't doubt that a moment for you, but I was addressing Sedat, who shows by his remarks that he does not understand this at all.

If Stockfish has gained strength because of the release of Ippolit, it is certainly not by including Ippolit code. We can be sure about that, as both the sources of Stockfish and Ippolit are published. Yet Sedat keeps insisting that everyone does the same as what Vas is accused of...

* If Rybka will be counted as a clone or derivative, then:
- Stockfish, Houdini, Gull, Komodo and plus many many engines should be counted in this group too !!

Welcome again HGM ))

It seems you can't get sleep, right ??))

Ok...I don't understand this at all

But I understand one thing at all:
Ippolit = Rybka

In other words,
All top engines (Stockfish,Houdini,Komodo,Gull) include Rybka blood
And if Rybka is a clone or derivative, then all those top mentioned engines should be in that group too

There is no other explanation...!! do you understand me now ?))

And don't be more naive, ok )) ?

Have a good night )

[Sedat Canbaz]

Sometimes I am wondering too,

Exception own original ideas and both methods...
Is there other methods during creating engine programs?

Method 1:




Method 2:


Btw, (many engines include same/similar related opening code, when they play without books)
And those chess engines are created with method 1 or with method 2 ?

[jdart]

Stockfish has no close similarity to Ippolit or its derivatives.

You might think there is only one way to build a very strong, top-10 engine, and that Rybka and its known and suspected derivatives showed the way. It is reasonable to suppose that. But as Stockfish shows, that is not so. There are multiple ways to do it.

I think the major contribution from Rybka to Stockfish is actually the testing methodology, using a large cluster - that part is similar.

--Jon