Clone engine list

[Ron Murawski]

The Wincraft==> crafty cone was a conviction by shouting, and nobody ever gave any sort of fact to indicate a sniff of a reason for guilt.

Wincraft was absolutely, positively, not a crafty clone.

The only evidence I know of is the sound of the names.
The guy's name was Craft.

It was about 5% as strong as crafty, which would instantly blow Wincraft out of the water.

The Wincraft binary:
11/01/1999 02:51 AM 264,704 WinCraft.exe

A contemporary Crafty binary:
02/14/1999 12:51 PM 466,944 Crafty1605.exe

Wincraft lacks opening books, tablebase files, and any other sophistication found in crafty.

I don't know where the 'Wincraft is a crafty clone' rumor started, but it has been perpeturated and is clearly, utterly, absurdly wrong.

Thanks, Dann!

For now Wincraft has been deleted from my list.

I found Wincraft mentioned as a clone in these two places:
http://209.85.165.104/search?q=cache:Bp ... cd=4&gl=us
http://www.exactachess.com/modules.php? ... gine_Links
Those were my sources of reference.

Perhaps it was a clone, but not a Crafty clone?

Ron

[Ron Murawski]

There are valid clones and invalid clones.
Valid clones include:
1. GPL sources that have been forked (e.g. Toga)
2. Projects that have a simple chess program start (used with permission)
3. Other projects that have no legal encumbrances

Invalid clones include:
1. GPL projects where the source is kept private (a clear violation)
2. Projects where large blocks of code are cut and paste from other programs (without permission).
3. Masquerade projects, where a program claims to be original but is just a tweaked version of some other program (sometimes commercial programs)
4. Code used without permission for any other reason.
5. Various other violations

Learning from someone else's code and writing the same algorithm yourself is not a violation, unless the original author has a patent. However, in this case, it is very unlikely that the two programs would behave identically (or even extremely similarly).

Are you saying that Gridchess *is* a clone? Or that it is *not* a clone? Or is this a general comment to the discussion? I'm not sure what you intended.

One of the clone engines in my list is a 1:1 copy of the original, but it conforms to this 'Valid clone' rule:
"2. Projects that have a simple chess program start (used with permission)"

Trying to make rules is a difficult thing to do. I'm trying to evaluate the circumstances of each engine to make a judgement. It's not always easy making a clone/no-clone decision.

Ron

[Tony Thomas]

Gridchess was/is a multiprocessor fruit according to many who competed at the tournament. However, he did not try to cover it up, and it was part of some sort of experiment, in my humble opinion you shouldnt add it to the list.

[Tony Thomas]

I posted about an engine, after looking at your list I saw that it was already listed.

[Claude Dubois]

Hello,

El Turco from Filipe Maina n the old Arena's list was a clone (a big patchwork).

Now the list is on Extachess but I don't find the new list

But, in WBEC and RWBC, El Turco is mentionned without problem.

Günther write from which engine's basis is El Turco written (Faile and Crafty)

I retired El Turco but it seems that I could let it play.

What can I do ?

Thanks for your remarks.

Best wishes

Claude

[Ron Murawski]

Hello,

El Turco from Filipe Maina n the old Arena's list was a clone (a big patchwork).

Now the list is on Extachess but I don't find the new list

But, in WBEC and RWBC, El Turco is mentionned without problem.

Günther write from which engine's basis is El Turco written (Faile and Crafty)

I retired El Turco but it seems that I could let it play.

What can I do ?

Thanks for your remarks.

Best wishes

Claude

Hi Claude,

I had El Turco on the clone list briefly. But then I noticed that it had its own website.
http://xray.bmc.uu.se/~filipe/elturco/
And then I noticed that it was an open source program. You can be fairly certain a program is not a clone when source code is supplied. (But be aware I did not look at the source code to verify if it was not a clone.) Of course, it's possible it was a clone a long time ago, but I have no information on that.

Ron

[Claude Dubois]

Hello Ron,

Thanks for your list.

El Turco will play in my next tournament.

Best wishes

Claude

[Dann Corbit]

There are valid clones and invalid clones.
Valid clones include:
1. GPL sources that have been forked (e.g. Toga)
2. Projects that have a simple chess program start (used with permission)
3. Other projects that have no legal encumbrances

Invalid clones include:
1. GPL projects where the source is kept private (a clear violation)
2. Projects where large blocks of code are cut and paste from other programs (without permission).
3. Masquerade projects, where a program claims to be original but is just a tweaked version of some other program (sometimes commercial programs)
4. Code used without permission for any other reason.
5. Various other violations

Learning from someone else's code and writing the same algorithm yourself is not a violation, unless the original author has a patent. However, in this case, it is very unlikely that the two programs would behave identically (or even extremely similarly).

Are you saying that Gridchess *is* a clone? Or that it is *not* a clone? Or is this a general comment to the discussion? I'm not sure what you intended.

One of the clone engines in my list is a 1:1 copy of the original, but it conforms to this 'Valid clone' rule:
"2. Projects that have a simple chess program start (used with permission)"

Trying to make rules is a difficult thing to do. I'm trying to evaluate the circumstances of each engine to make a judgement. It's not always easy making a clone/no-clone decision.

Ron

The simplicity of the start has no connection to validity of the clone. Toga II is every bit as valid as an authorized clone of Faile. Now, as to whether these sorts of clones are allowed to participate in contests is totally another matter (separated in my view from whether or not the clone is legitimate).

GridChess is a clone, but it is a *valid* clone (if source code is provided somewhere) because it is clearly explained in the project paper that Fruit is used as the baseline code. They did not pretend to have some other origin (gave credit for the original). Fruit is GPL and so Toga and GridChess are not violations of the spirit of the GPL agreement so long as source is provided.

It is possible that the Paderborn controversy could be a violation of *contest* rules, whether or not the clone itself is valid.

It is possible that the GridChess source has not yet been made avialable (I have not looked).

In the case of a GPL project, there are two possible sorts of violations:
1. The origin of the project is hidden. This is (essentially) plagiarism. It is also a violation of the second point (below). This is a deliberate step and causes the perpetrator to lose all credibility.
2. The modified source is hidden. (This is a violation of the GPL agreement itself). This sometimes happens due to ignorance. Credibility can be restored if the source is made public.

IMO-YMMV

[Tony van Roon-Werten]

There are valid clones and invalid clones.
Valid clones include:
1. GPL sources that have been forked (e.g. Toga)
2. Projects that have a simple chess program start (used with permission)
3. Other projects that have no legal encumbrances

Invalid clones include:
1. GPL projects where the source is kept private (a clear violation)
2. Projects where large blocks of code are cut and paste from other programs (without permission).
3. Masquerade projects, where a program claims to be original but is just a tweaked version of some other program (sometimes commercial programs)
4. Code used without permission for any other reason.
5. Various other violations

Learning from someone else's code and writing the same algorithm yourself is not a violation, unless the original author has a patent. However, in this case, it is very unlikely that the two programs would behave identically (or even extremely similarly).

Are you saying that Gridchess *is* a clone? Or that it is *not* a clone? Or is this a general comment to the discussion? I'm not sure what you intended.

One of the clone engines in my list is a 1:1 copy of the original, but it conforms to this 'Valid clone' rule:
"2. Projects that have a simple chess program start (used with permission)"

Trying to make rules is a difficult thing to do. I'm trying to evaluate the circumstances of each engine to make a judgement. It's not always easy making a clone/no-clone decision.

Ron

The simplicity of the start has no connection to validity of the clone. Toga II is every bit as valid as an authorized clone of Faile. Now, as to whether these sorts of clones are allowed to participate in contests is totally another matter (separated in my view from whether or not the clone is legitimate).

GridChess is a clone, but it is a *valid* clone (if source code is provided somewhere) because it is clearly explained in the project paper that Fruit is used as the baseline code. They did not pretend to have some other origin (gave credit for the original). Fruit is GPL and so Toga and GridChess are not violations of the spirit of the GPL agreement so long as source is provided.

It is possible that the Paderborn controversy could be a violation of *contest* rules, whether or not the clone itself is valid.

It is possible that the GridChess source has not yet been made avialable (I have not looked).

In the case of a GPL project, there are two possible sorts of violations:
1. The origin of the project is hidden. This is (essentially) plagiarism. It is also a violation of the second point (below). This is a deliberate step and causes the perpetrator to lose all credibility.
2. The modified source is hidden. (This is a violation of the GPL agreement itself). This sometimes happens due to ignorance. Credibility can be restored if the source is made public.

IMO-YMMV

iirc, Gridchess was also using an unauthorized book.

Tony

[Yannik Snoeckx]

Hello !

While setting up old engines for my Dinosaur League, I think that I've had run across a 20th century programm that looks like a clone to me, but has never been (to my perhaps lacking knowlege) discovered ... or accused and then innocented .

What would be the proper way to clear this case ? I am no programmer and can't be 100% sure of my allegations.

Is there some "expert", willing to take time to look at it, I can e-mail my suspicions to ?

Should I put my proofs here (or in another thread) without naming the suspected engine and the supposed model ?

Or should I go "tabloïd" and put it all here ?

Best wishes.
Y. Snoeckx

[Roger Brown]

Hello !

While setting up old engines for my Dinosaur League, I think that I've had run across a 20th century programm that looks like a clone to me, but has never been (to my perhaps lacking knowlege) discovered ... or accused and then innocented .

What would be the proper way to clear this case ? I am no programmer and can't be 100% sure of my allegations.

Is there some "expert", willing to take time to look at it, I can e-mail my suspicions to ?

Should I put my proofs here (or in another thread) without naming the suspected engine and the supposed model ?

Or should I go "tabloïd" and put it all here ?

Best wishes.
Y. Snoeckx

Hello Yannik,

Going tabloid is strictly prohibited.



Post the proofs here without naming names and perhaps one of the resident geniuses here - Dann or Guenther for instance - can have a look if they are sufficiently interested in the case.

Later.

[Yannik Snoeckx]

Hello !

Here are the reasons for my suscipicions. I hope I am clear enough, as I am neither a programmer, nor a good english writer. Supposed model is named M and presumed clone is called C.

1. they both use the same book. (not hidden, stated by C's author in his site).

2. they both use the same (a bit peculiar) time management, taking a lot of time for the first moves in a very fixed way. In 4m+2s, 11s for the first few moves, then 10, then 9 ...

3a they both "invoke" winboard with the term "xb" on the command line.
3b. and much more strangely, both can play without it !

4a the hashes are for both set on command line by the same term.
4b and both loads with 2mb and then once they have started to play go to 43mb on Arena 1.1 if you have set 64mb.

5. there is almost exactly the same display in console mode when starting the engine.
5a. After the name of the program (evidently different), hash and pawn size are indicated, with precisely the same value for hashes : "262144 entries, 5242880 bytes".
5b. Pawn size are differents : 65536 for M and the strange 900000 for C, meaning if I understand correctly that hashpawns are superior to standard hashes, which looks very peculiar, perhaps it is a "masquerade". (of course neither programm as any .ini file)

5c. Then it is both times written "type help for a list of commands"
5d. The programs waits in both cases for an ordrer this way : "White-To-Move[1]:"

6. the list of commands are a bit differents, and presented in a slightly different way, but ALL commands from C are present in M, including some specific ones, like "Hint", "Score" (same presentations and same criterions) and "nodisplay".

7. if you type "hash 64", you get exactly the same values displayed for both entries and bytes, for both pawns and hashes. It is in contradiction with the strange hashpawn value of C from point 5b.

8. after the command display, the board is shown exactly with the same "graphics", that is letters for the pieces and characters like "+" or "-" to draw the board. This appears in other programms too, but Black pieces are indicated with "<>" and dark squares with "...", which is not the case often. And what about the indications "white to move" and "computer time : 30 seconds" (29 seconds for C!). M gives some more informations, but as for point 6, all that it in C is in M.

Here is what I can say after investigating, but I insist I am no programmer, and I haven't got an absolute proof, if ever there exists one.

To explain points 6 and 8. I suspect that C could come from an earlier version of M than the ones that I own. It would also justify why M is stronger than C and plays not exactly the same way. I am not an 100% sure, but I think that M's source was avaible at the time C appeared. And C was "developped" very briefly, while M has been avalaible and improved for many years. To me, it looks like C for the least borrowed too much from M !

What do you think of all that ? Does my suspicions warrant further investigations ? Or should I let C play in my dinosaur league ?

Thanks in advance for your answers.

Best wishes.
Y. Snoeckx

[Guenther Simon]

Hello !

Here are the reasons for my suscipicions. I hope I am clear enough, as I am neither a programmer, nor a good english writer. Supposed model is named M and presumed clone is called C.

1. ...

snip...

What do you think of all that ? Does my suspicions warrant further investigations ? Or should I let C play in my dinosaur league ?

Thanks in advance for your answers.

Best wishes.
Y. Snoeckx

I would look at it if you mail me at r-w-b-c@g-m-x.d-e
(remove all '-' characters for a valid address.)

Regards,
Guenther

[Guenther Simon]

Hi Yannick,

You don't need to name me both programs anymore, I know
them already by now, but I need an older version
of the perhaps cloned program from 1999.
The oldest I have is from 2001 (version 2.55)

Regards,
Guenther

[Yannik Snoeckx]

Hi Yannick,

You don't need to name me both programs anymore, I know
them already by now, but I need an older version
of the perhaps cloned program from 1999.
The oldest I have is from 2001 (version 2.55)

Regards,
Guenther

Hi Guenther,

Thanks.

Both programms have been sent to the e-mail adress you indicated.

It looks like with my description ,the identities were not entirely protected
You have too much winboard engines knowledge

Best wishes.
Yannik

[Tony Thomas]

I think I know the mystery engines in question as well. I guess I cant say which one is the clone because that would be wrong, but I think the original engine is Exchess.

[Yannik Snoeckx]

I think I know the mystery engines in question as well. I guess I cant say which one is the clone because that would be wrong, but I think the original engine is Exchess.

Hi Tony,

If you have an opinion, would you tell about it ?
Clone or not clone ?

Best wishes.
Y. Snoeckx

[Tony Thomas]

I think I know the mystery engines in question as well. I guess I cant say which one is the clone because that would be wrong, but I think the original engine is Exchess.

Hi Tony,

If you have an opinion, would you tell about it ?
Clone or not clone ?

Best wishes.
Y. Snoeckx

Problem is that I no longer have the old winboard versions of Exchess (I was able to find the chessbase versions on one of my old computers but that doesnt really help). Even if I did, I wont know more than to compare the size of the executables and run few matches. Since Dann Corbit was able to identify Strelka as a clone, may be we should contact him. I think it might be a clone, but since I am not 100% sure I would rather say I dont know if it is a clone or not. I am also going to contact Ron.

<YO>

[Yannik Snoeckx]

Problem is that I no longer have the old winboard versions of Exchess (I was able to find the chessbase versions on one of my old computers but that doesnt really help). Even if I did, I wont know more than to compare the size of the executables and run few matches. Since Dann Corbit was able to identify Strelka as a clone, may be we should contact him. I think it might be a clone, but since I am not 100% sure I would rather say I dont know if it is a clone or not. I am also going to contact Ron.

Hi Tony,

Thanks for your answer and thanks for telling knowledgeable people.
Guenther, as he posted, in on the case too.

I am being too impatient.
One is innocent until proved guilty and as the accused has been here for years, why not wait a few more days ?

Best wishes.
Y. Snoeckx

[Tony Thomas]

Problem is that I no longer have the old winboard versions of Exchess (I was able to find the chessbase versions on one of my old computers but that doesnt really help). Even if I did, I wont know more than to compare the size of the executables and run few matches. Since Dann Corbit was able to identify Strelka as a clone, may be we should contact him. I think it might be a clone, but since I am not 100% sure I would rather say I dont know if it is a clone or not. I am also going to contact Ron.

Hi Tony,

Thanks for your answer and thanks for telling knowledgeable people.
Guenther, as he posted, in on the case too.

I am being too impatient.
One is innocent until proved guilty and as the accused has been here for years, why not wait a few more days ?

Best wishes.
Y. Snoeckx

Ron said that he took a look at both engines and that there are considerable differences among them.