TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
User avatar
Don
Posts: 5106
Joined: Tue Apr 29, 2008 2:27 pm

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by Don » Tue Nov 30, 2010 1:13 pm

kasinp wrote:
Don wrote:
notyetagm wrote:
Albert Silver wrote:
Tom Barrister wrote:It's apparent to me that if Mr. Rajlich were to bring a lawsuit against the clones, he would need to
...identify who he was suing.
Wow, is this controversy *ever* going to end?

Is it not clear that Rybka 1 was heavily dependent on the Fruit source code? And that Houdini is partially based on the Ippolit family, as the author clearly states on his website?
I'm only going to comment on this one point, as I don't really want to get sucked into another Rybka discussion.

The question is whether Rybka is "heavily dependent" on the Fruit source code. Fruit is a 32 bit program and Rybka is a 64 bit program so I think the answer is no. However I think a more relevant question is whether Rybka derived it's strength by stealing the Fruit code.

What do you think the answer is? I looked on the IPON rating list and Fruit is 2626 ELO, and 32 bit Rybka on this same list is 2849 which is a difference of 223 ELO points which is massive. It's much more if you consider the 64 bit version.

So even if Vas did start with the open source fruit program, it has nothing to do with Rybka now - there is nobody who can say that Vas was some loser who only has a strong program because he stole "Fruit" and changed a couple of lines of code because he did not have any original ideas of his own.

This is a completely different issue from what the so called "cloners" are being accused of. Regardless of where you stand on this issue the accusation is that the so called "cloners" took some strong program, made a few changes without making any substantial improvements, and put their name on it.

If you want to accuse Vas of something, accusing him of adding several hundred ELO of strength without using some else's ideas.

As far as the "clones" are concerned, that is for you to decide but you really need to take this point that Rybka is a Fruit clone off the table - it's just so hollow sounding and it makes you look stupid (not directed to anyone in particular, but to all those that keep saying it over and over.)


Not to randomly change the topic, but can you share your plans for (hopefully) releasing a new version of Komodo?

Regards,
PK
We have a new version which appears to be about 30 ELO stronger than the last release - but we are holding out for more before we release. To me 30 ELO is barely worth a release. We seem to be able to still keep finding improvements but not as quickly as before. We want to quickly follow this up with an MP release, or perhaps simultaneously.

kasinp
Posts: 216
Joined: Sat Dec 02, 2006 9:47 pm
Location: Toronto
Full name: Peter Kasinski

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by kasinp » Tue Nov 30, 2010 1:23 pm

Thanks for the response.
MP version based on 30+ELO improvement to v.1.2 could shake up current rating lists. Can't wait to see it!

PK

Tom Barrister
Posts: 227
Joined: Tue Oct 05, 2010 3:29 pm

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by Tom Barrister » Tue Nov 30, 2010 1:45 pm

M ANSARI wrote:There we go again ... all Rybka's are clones of Fruit and Ippolit is not RE from R3. I think if you would put up a Poll and ask who believes that R3 is a clone of Fruit, very predictably you will get the same clueless group voting that R3 is a Fruit clone and that Ippolit has nothing to do with R3. You ask "where is the proof" that they are clones ... I think the best proof is the BB report.
It would be ludicrous to flat state that the Ip* clones aren't based on some strong chess engine. Several strong engines don't just pop up out of nowhere and become major players from scratch. Since Rybka 3 was at the top of the heap before they arrived, it's also logical to assume that this was the engine that was decompiled/source-stolen/whatever. Since it isn't proven either way, there are still supporters, to some degree, on each side.

Since these engines have improved over their evolution, it would be equally ludicrous to state that this improvement is also based on Rybka. At some point, they take on their own lives. Where the improvements came from is open to debate. However, if they WERE initially Rybka clones, their credibility takes a hit, no matter how far the final product evolves away from the original. It makes one wonder where the improvements are coming from. The authors' own creativity? Ripped bits and pieces from other engines? IF the originals are clones of Rybka/whoever, the credibility quotient would make it logical to assume that at least some of the improvements have also been obtained elsewhere.
M ANSARI wrote: Even with Rybka 2.3.2a, at LTC Zappa Mexico II was scoring very close or even equal to Rybka 2.3.2a. The big breakthrough in chess engine strength came with R3 ... and R3 is the original creation of Vas
It would be equally ludicrous to believe that Rybka 4 is still a full clone of Fruit. How did it improve over 3 years, when Fruit hasn't changed? Obviously, it's evolved into something else.

Applying the same logic and arguments to Rybka as we did to the clones, Rybka appeared out of nowhere and rose quickly to or near the top of the computer world. It would be ludicrous to believe that it did so as fresh code created from scratch. Since it has always (to my knowledge) been closed source (e.g. commercial, private, beta-commercial, etc.), it isn't proven that some/most/all of that code would have had to come from either another commercial engine or from open-source. However, "Not proven" does not necessarily equate to "unlikely". Using code from a commercial engine without the author's permission is piracy; my understanding is that using the open-source in a commercial product (or at least without stating this fact) is a violation of the GPL. Most/all fingers who point to the first Rybka or two as being a clone, point to Fruit as the target that was cloned. They provide convincing arguments. Arguments also exist on the other side. In any case, it's not illogical to assume the original Rybka was ripped, to some degree, from somewhere.

IF Rybka one was largely a clone of Fruit, then its (or the author's) credibility is shot the same as the Ip* clones would be IF they are clones of Rybka. It doesn't really matter if Rybka 3 is entirely programmed from scratch and has no code whatever from Fruit/wherever. You can't have it both ways. If they're cloned, the authors were thieves.

IF Rybka one is a clone of Fruit/whatever-engine(s), then the next question is: where did the improvements for Rybka 2, 3, and 4 come from? Not from the last version of Fruit, as it hasn't been worked on in the interim. Did Mr. Rajlich provide all of the improvements himself? Did he take/steal/gather some/most/all of them from open-source projects or other commercial engines? Did he gleam the best bits and pieces of the ideas of others and put them together, with or without his own ideas? The old cliche comes to mind: "Once a thief, always a thief." You can't have it both ways. Assuming for the nonce that Ip* engines are clones of Rybka, and Rybka One is a clone of Fruit, then both (sets of) authors are at fault. One can't be a wrong without the other also being wrong.

And the next question is the one that I can't fathom: If Rybka 3 and beyond is/are, from a programming standpoint, distanced far enough away from the original Rybka to make said original trivial by comparison and its source harmless if viewed (meaning that seeing Rybka One's source won't allow others to make Rybka 4 from it), and IF Rybka One is an original work (or at least original enough to not be largely a line-for-line clone of something else or in violation of the GPL), then WHY won't Mr. Rajlich release the source of Rybka One and settle the argument once and for all? He would have much to gain if he's innocent and if Rybka One is his own legitimate work. Certainly a brace of Rybka One clones, as-is or improved upon, won't threaten the current top engines. Of course, he'd have much to lose if the source were released, and Rybka One was almost a line-for-line copy of, say, Fruit 2.1.

Since the source of Rybka One hasn't (to my knowledge) been released, then naturally people are going to be skeptical of its origin(s). Can you blame people for wondering what Mr. Rajlich has to hide?

My conclusions might or might not be correct, but to flat-state that they can't possibly be right would seem to be blindly obstinate, in light of current circumstances. Call me all the names you want, but I don't see a full refutation to the questions raised, nor do I see a full verification to same.
This production is being brought to you by Rybka: "The engine made from scratch.™"

User avatar
M ANSARI
Posts: 3499
Joined: Thu Mar 16, 2006 6:10 pm

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by M ANSARI » Tue Nov 30, 2010 2:31 pm

What you seem to not understand is that the issue between Rybka 1.0 beta and Fruit was not if there was any code "stolen", as you really cannot steal from what is open source ... but rather that if code was used then that would constitute a GPL violation. From what I have read so far, the only thing that could have possibly constituted a GPL violation was the UCI parsing code and not the chess engine code. When you add to that the fact that the very author of Fruit also seemed to think that although some ideas were used that code was re-written so therefore ok. I would tend to take his word over others. That is a dramatically different thing than RE engineering a closed source commercial program and offering it off as free source to the world. It is always so very easy to give away something that is not yours for free. Also I have not seen any of the Ippolit derivatives give one iota of credit to the true person behind these engines, which is Vas. Most seem to be thinking that if anything unethical was done it was done with Ippolit so legally and ethically they are in the clear. I would think that the least these authors should do, after all the data from BB has been produced, is to give credit to Vas for creating the incredible code that gives their engine its strength. So far I have yet to see any one of the clones give Vas any credit.

As for Ippolit being based on R3 ... well at least you seem to admit that it has to be based on R3. I think several hundred posts on this very forum were based on tooth and nail denial of this very fact not so long ago. Only after BB posted his report, and Zach, Chris and the author of DS as well as many other established and respected engine authors opinions came out, has this silly denial been laid to rest. Only a total idiot would today claim that Ippolit and its derivatives did not come from R3.

Milos
Posts: 4012
Joined: Wed Nov 25, 2009 12:47 am

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by Milos » Tue Nov 30, 2010 3:15 pm

M ANSARI wrote:What you seem to not understand is that the issue between Rybka 1.0 beta and Fruit was not if there was any code "stolen", as you really cannot steal from what is open source ... but rather that if code was used then that would constitute a GPL violation. From what I have read so far, the only thing that could have possibly constituted a GPL violation was the UCI parsing code and not the chess engine code. When you add to that the fact that the very author of Fruit also seemed to think that although some ideas were used that code was re-written so therefore ok. I would tend to take his word over others. That is a dramatically different thing than RE engineering a closed source commercial program and offering it off as free source to the world. It is always so very easy to give away something that is not yours for free. Also I have not seen any of the Ippolit derivatives give one iota of credit to the true person behind these engines, which is Vas. Most seem to be thinking that if anything unethical was done it was done with Ippolit so legally and ethically they are in the clear. I would think that the least these authors should do, after all the data from BB has been produced, is to give credit to Vas for creating the incredible code that gives their engine its strength. So far I have yet to see any one of the clones give Vas any credit.

As for Ippolit being based on R3 ... well at least you seem to admit that it has to be based on R3. I think several hundred posts on this very forum were based on tooth and nail denial of this very fact not so long ago. Only after BB posted his report, and Zach, Chris and the author of DS as well as many other established and respected engine authors opinions came out, has this silly denial been laid to rest. Only a total idiot would today claim that Ippolit and its derivatives did not come from R3.
Rybka 1 contained at least 30% of the code (verbatim copied) of Fruit and at least 80% of ideas. In total and practical sense Rybka 1 was bitboard Fruit, no more no less.
Before stealing a lot of Fruit code and most of its ideas Rybka bF (before Fruit) was a mediocre and totally amateur engine obviously written by someone with very little if any chess knowledge. Rybka aF (after Fruit) suddenly gained over 600 elos. Anyone with at least half a brain understands this. The rest have moral issues (since they obviously have the whole brain).
Further versions of chess engine are never written from scratch and probability that all further Rybka versions still contain Fruit code is extremely high.

Rybka 3 even today still contains at 30% of Fruit ideas (a well know fact obvious from BB report and to anyone that actually did disassemble it).

Ippolit might be based on ideas from disassembled Rybka 3. However, the percentage is certainly less than 50%. Ippolit also contains the same 30% of ideas from Fruit that Rybka 3 has.
Ippolit contains literary 0% of code from Rybka 3.
So it's safe to claim that Rybka 3 is closer to Fruit than is Ippo to Rybka 3.

Moreover, reverse engineering is a completely legal form of research and who knows how many programs Vas dissembled in order to "steal" ideas. Zappa Mexico is one for example.

So the most morally compromised author and engine in the world of all the established engines is Vas and Rybka.

Judging any other engine or author while this shameful engine and author are justified is hypocritical, immoral and dishonest.

Ralph Stoesser
Posts: 408
Joined: Sat Mar 06, 2010 8:28 am

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by Ralph Stoesser » Tue Nov 30, 2010 3:46 pm

Milos wrote: Rybka aF (after Fruit) suddenly gained over 600 elos. Anyone with at least half a brain understands this.
How exactly did that incapable guy gained that much ELO?

Facts please ...

User avatar
Don
Posts: 5106
Joined: Tue Apr 29, 2008 2:27 pm

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by Don » Tue Nov 30, 2010 4:11 pm

Tom Barrister wrote:
Don wrote: [snip]
I'm not interested in rehashing who or what new program is or is not a clone of Rybka [snip]
Yeah, you are. You came back again after stating the same thing earlier.

I'm afraid that I don't see the point to your last post, unless calling me names and flaming me is the goal here. I suppose you've had a lot of practice at that, but I really don't care to read all 1065 of your posts to find out. You're certainly in good company here; a lot of others do the same thing. I try to stay out of that, but it's difficult sometimes.
I don't mean this as a personal insult to anyone and I'm sorry about the strong language. I used the term idiot unfairly and harshly, a more appropriate word is "fool" because a fool does not lack intelligence, he just tends to be easily swayed by emotional arguments and (in my opinion) generally is a participant in his own willingness to be fooled.

But the arguments that I keep seeing posted are just not reasonable or logical. I especially cringe at the references to Fruit and Rybka which is almost always presented to deflect attention away from the clones, not to present evidence that some other program is, or is not a clone. (But it works on an emotional level.)

Don't you see how foolish this is? Someone says that they think some program is a clone of Rybka, and then the Vas bashing begins. ????

It is clear that all programs are in some sense a product of all the programs that came before them. But I think Rybka pretty much blew away everyone else in this regard. You can try to go negative to refute this (since facts won't work for you in this regard), but the simple truth is that almost all of the last 200 ELO of progress came from Rybka. The new Rybka wannabees may have SOME original ideas in them, but they have clearly failed to put any distance between themselves and Rybka. Whether Rybka is a clone of Fruit hardly matters to me, she put hundreds of ELO between herself and the rest of the pack. You don't do this with clonesmanship.

There are people on this forum who squeal with delight if a clone gets close to Rybka on some rating list. They do not report this with any objectivity but treat it as if it's some kind of deliverance from evil. Such behavior makes it seem like there is an anti-Rybka movement composed of emotional people who have an axe to grind.

I also sense vigilantism because when the argument is presented that Rybka might be a Fruit clone it tends to be to criticize the morality of Vas and at the same time used as a justification for the clones. It is basically saying that it is only fair that Vas is "punished" and that he gets what he deserves, etc. It's like a pat on the back of the cloners for doing what they seem to think Vas did to them. Of course it's completely ignored that Vas advanced the state of the art by at least 200 ELO, but the cloners so far have not been able to increase it even by 1 ELO.


In a decade that saw massive improvement in chess engines as a whole, wouldn't you expect a program that's not been developed in over three years to lag behind those which have been actively worked on? It would be absurd to think otherwise.
There are many programs that have been in development for years and some of them are near the top. Rybka came and blew them all away. This has nothing to do with Fruit.

Of course you want to make another connection to Fruit by implying that Rybka only followed the path that Fruit would have taken (and thus Rybka is not a leader but just another follower.)

The "cause" is that Rybka blazed new trails. The "effect" is that everyone else had to catch up by copying Rybka. When you blaze a trail people follow your path and you are treating us like idiots by trying to imply that it was actually the other way around. This is why I keep saying this argument about fruit is so emotional and intellectually dishonest and appeals only to fools who do not see to want to think this through in any kind of objective way.


I stated nowhere (yet) that Mr. Rajlich stole code from Fruit. I did offer it as a hypothesis. There's a long line of others ahead of me who support that hypothesis, and contrary to your beliefs, not all of them are "fools" or "idiot(s)" (your words) for believing it.
This is yet another diversion. It's NOT important. I don't care if Vas lied or didn't. I don't care if ten years ago Vas started with Fruit to save having to write a move generator. This is just not important and in no ways helps to determine if some suspected clone is based on Rybka or not.

What is important is that Vas blazed new trials, and so far none of the many brand new programs have done anything to compare to this. No amount of sleight of hand, misdirection or emotionally appealing arguments will change this fact.

Hey, I know what we can do - let's talk about Fruit again since it doesn't have anything to do with anything.

As far as Rybka itself goes, it's arguably the strongest program available to the public, although its lead is tenuous, and the competition is closing in. Since Mr. Rajlich has stated that Rybka 5 won't be out until the fall of 2011, it would be logical to believe that at least one of the competition will pass Rybka in the interim. Some believe that it's happened already. Stockfish, Houdini, Ivanhoe, Firebird, and probably a few others could make a good case to stake a claim to the top spot.
What's the point of this?

By reverse engineering Rybka we have for the most part caught Rybka. We have taken her secrets and now we are bragging about it.

Whether this is right or wrong is another discussion, but it turns my stomach to carry on as if we have done something great and not give Rybka most of the credit for this.

I want everyone to know that I am not especially a Rybka lover. I can be pretty competitive and I am jealous of her accomplishments. If Rybka had never existed neither would any of the suddenly appearing new programs with authors nobody ever heard of and Stockfish would be 100 ELO weaker as would Komodo and many other programs.

But I believe in credit where credit is deserved, even if it has to be grudgingly admitted.

I also believe that if Fabien Letouzey started seriously working on chess engines again, that his engine would, in time, be a major player, possibly the best out there. in my opinion, he's that good.
It's not likely he would improved Fruit 300 ELO in a few months. What he did was impressive but we will never know.

If he is really that good he could take what he has and incorporate what is in the open source clones and then add 100 ELO to that. And THAT is my point, if you can take the best out there and add 300 ELO then you have my respect and that is what Vas did. I have not seen anyone else do anything like that since then, have you?
The more I see the Rybka community spouting about their almighty engine and Deity of a programmer, and the more I see almost every one of them flame anybody who dares attack either, the more disenchanted I get with Rybka as a whole. It's been my experience that when multitudes of people all proclaim something is so (in this case, that Rybka stole nothing/little from Fruit and that Mr. Rajlich is innocent of any chicanery) and flame anybody who disagrees, the opposite is generally true. I've tried to be objective about the whole clone issue, but I'm inclined to believe that the folks who say that Rybka was largely derived from Fruit are correct.
This is just not an important issue.

I agree with you about the hero worship given Vas and people do get carried away - but on the other hand it's incredibly difficult to buy into the propaganda that this was not an impressive accomplishment. You are trying to imply that what he did is not impressive at all by drawing attention to programs that only now are just catching up after several years (and it's almost certainly because they have picked apart what is in Rybka to do this.)

Then there's the matter that I haven't touched on (but many others have) that Rybka just happened along, out of thin air, and rose to the top in a very short time.....just about the same time Fruit's last release, and about the time it went open-source. It's almost as though xann (Mr. Letouzey) knew or at least believed that Mr. Rajlich ripped Fruit almost line-for-line, got fed up, and left in disgust. According to what I've read, Mr. Letouzey did make a comment, at about that time, inferring that several people were using Fruit's source (possibly even before it was open-source) for their engines. That may be relevant, or it may not, but one is certainly not an "idiot" or "fool" for wondering.
I can only speculate about what might have happened, I don't really know.

I think it's very possible that Vas initially started with Fruit but he did not end up with anything remotely like Fruit. I don't know if you are programmer but going to 64 bits is no trivial modification. It completely changes everything about a program. Vas added over 300 ELO and he didn't do it by making a few tweaks here and there - he changed virtually everything about the program including the very guts of it.

Rybka rose to the top in a very short time but you cannot claim that was because he copied someone. Every chess programmer knows that the last 100 ELO are by far the hardest. Those 300 ELO are not from Fruit, they are from Vas (and Larry.) And those 300 ELO represent 99 percent of the program and the work that went into it.

This is not to mention the fact that if he did start with Fruit, it's a completely new program now as you cannot cut and paste 32 bit source code to form a 64 bit program. This is just part of the reason that I find this argument so hollow and irrelevant.

I know this was not the main thrust of your argument and I appreciate that. But most of the posts on this subject of Rybka clones immediately ends up with an assertion that Rybka is a clone of Fruit and by implication that Vas didn't do anything original.

Milos
Posts: 4012
Joined: Wed Nov 25, 2009 12:47 am

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by Milos » Tue Nov 30, 2010 4:11 pm

Ralph Stoesser wrote:
Milos wrote: Rybka aF (after Fruit) suddenly gained over 600 elos. Anyone with at least half a brain understands this.
How exactly did that incapable guy gained that much ELO?

Facts please ...
Divine intervention, beatific inspiration, alien technology. Sure, why not, anything but Fruit becoming open source, right? ;)

Milos
Posts: 4012
Joined: Wed Nov 25, 2009 12:47 am

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by Milos » Tue Nov 30, 2010 4:39 pm

Don wrote:There are many programs that have been in development for years and some of them are near the top. Rybka came and blew them all away. This has nothing to do with Fruit.
It has everything to do with Fruit. If there was no Fruit, there would be no Rybka 1, no Rybka 2, etc. Period.
Rybka would not exist.
It's a simple causality principle. I don't understand how you don't see it?
In 5 years of development of Rybka Vasik improved it for 300 elo. In last 2 years the improvement was a shameful 40 elo.
In few days with stolen Fruit code Rybka 1 improved 600 elo. That is a fact.

The "cause" is that Rybka blazed new trails. The "effect" is that everyone else had to catch up by copying Rybka. When you blaze a trail people follow your path and you are treating us like idiots by trying to imply that it was actually the other way around. This is why I keep saying this argument about fruit is so emotional and intellectually dishonest and appeals only to fools who do not see to want to think this through in any kind of objective way.
Could you elaborate more on blazing trails. How the hell closed source program helped you improve your own program???
There is one, and only one way, dissasembling. So did you use information from disassembled Rybka (in case you disassembled it) before Ippolit appeared?
If not, how the hell did Rybka help you and the state-of-the-art???
Rybka only helped Vas pockets. That the only truth!
I want everyone to know that I am not especially a Rybka lover. I can be pretty competitive and I am jealous of her accomplishments. If Rybka had never existed neither would any of the suddenly appearing new programs with authors nobody ever heard of and Stockfish would be 100 ELO weaker as would Komodo and many other programs.
That is complete logic fallacy. The truth is completely other way around.
If there were no Ippolit, Komodo and SF would be 100 elo weaker, there would be no strong new engines and Rybka 4 would be Rybka 3 + buxfixes and not Rybka 3 + 40elo - bugfixes.
I really don't understand how intelligent man cannot see this.
This is not to mention the fact that if he did start with Fruit, it's a completely new program now as you cannot cut and paste 32 bit source code to form a 64 bit program. This is just part of the reason that I find this argument so hollow and irrelevant.
Going from 32 to 64 bit program is absolutely trivial. The only real difference is BSF, BSR and POPCNT implementation. I really don't understand what you are saying???

User avatar
michiguel
Posts: 6401
Joined: Thu Mar 09, 2006 7:30 pm
Location: Chicago, Illinois, USA
Contact:

Re: TEST position TCEC5- Houdini 1.03a-DRybka4 1-0

Post by michiguel » Tue Nov 30, 2010 4:55 pm

Milos wrote:
M ANSARI wrote:What you seem to not understand is that the issue between Rybka 1.0 beta and Fruit was not if there was any code "stolen", as you really cannot steal from what is open source ... but rather that if code was used then that would constitute a GPL violation. From what I have read so far, the only thing that could have possibly constituted a GPL violation was the UCI parsing code and not the chess engine code. When you add to that the fact that the very author of Fruit also seemed to think that although some ideas were used that code was re-written so therefore ok. I would tend to take his word over others. That is a dramatically different thing than RE engineering a closed source commercial program and offering it off as free source to the world. It is always so very easy to give away something that is not yours for free. Also I have not seen any of the Ippolit derivatives give one iota of credit to the true person behind these engines, which is Vas. Most seem to be thinking that if anything unethical was done it was done with Ippolit so legally and ethically they are in the clear. I would think that the least these authors should do, after all the data from BB has been produced, is to give credit to Vas for creating the incredible code that gives their engine its strength. So far I have yet to see any one of the clones give Vas any credit.

As for Ippolit being based on R3 ... well at least you seem to admit that it has to be based on R3. I think several hundred posts on this very forum were based on tooth and nail denial of this very fact not so long ago. Only after BB posted his report, and Zach, Chris and the author of DS as well as many other established and respected engine authors opinions came out, has this silly denial been laid to rest. Only a total idiot would today claim that Ippolit and its derivatives did not come from R3.
Rybka 1 contained at least 30% of the code (verbatim copied) of Fruit and at least 80% of ideas. In total and practical sense Rybka 1 was bitboard Fruit, no more no less.
Before stealing a lot of Fruit code and most of its ideas Rybka bF (before Fruit) was a mediocre and totally amateur engine obviously written by someone with very little if any chess knowledge. Rybka aF (after Fruit) suddenly gained over 600 elos. Anyone with at least half a brain understands this. The rest have moral issues (since they obviously have the whole brain).
Freaking unbelievable. I really did not expect anyone to start questioning VR chess skills, but this forum is full of surprises.

Miguel
Further versions of chess engine are never written from scratch and probability that all further Rybka versions still contain Fruit code is extremely high.

Rybka 3 even today still contains at 30% of Fruit ideas (a well know fact obvious from BB report and to anyone that actually did disassemble it).

Ippolit might be based on ideas from disassembled Rybka 3. However, the percentage is certainly less than 50%. Ippolit also contains the same 30% of ideas from Fruit that Rybka 3 has.
Ippolit contains literary 0% of code from Rybka 3.
So it's safe to claim that Rybka 3 is closer to Fruit than is Ippo to Rybka 3.

Moreover, reverse engineering is a completely legal form of research and who knows how many programs Vas dissembled in order to "steal" ideas. Zappa Mexico is one for example.

So the most morally compromised author and engine in the world of all the established engines is Vas and Rybka.

Judging any other engine or author while this shameful engine and author are justified is hypocritical, immoral and dishonest.

Post Reply