BB+ on the matter

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

Moderators: hgm, Rebel, chrisw

Milos
Posts: 4190
Joined: Wed Nov 25, 2009 1:47 am

Re: BB+ on the matter

Post by Milos »

Don wrote:In your own words, "Then there was Rybka, the monster beta that crushed everything." I might add, "including Fruit."
And again you distort the facts. Rybka 1 beta was stronger than Fruit 2.1 exactly for bitboard conversion+10-20elo=60elo.

Rybka is not so much stronger than Fruit as you wish to present. Juri already demonstrated that.
And even though everyone had access to Fruit's ideas, noone but Vas used Fruit's code.
What you are suggesting all the time is a fallacy. You have access to all the ideas of SF still your program is significantly weaker than SF (the strength difference is 2 times what was between Fruit 2.1 and Rybka 1 beta), why?
Because you didn't clone SF. If you did, with ideas from Komodo you would probably surpass R4, maybe even Houdini.
Roger Brown
Posts: 782
Joined: Wed Mar 08, 2006 9:22 pm

Re: BB+ on the matter

Post by Roger Brown »

Don wrote: Please don't take my statement out of context.
Hello Don,

May I ask for the same respect in terms of my statements? I have NEVER asserted, stated, alleged that Rybka is a clone of Fruit. Take your time and play back the tape. Hit the stop button when you get to that part.
Don wrote: Every program benefited from Fruit because it was a trail-blazer. Your assertion is that Rybka is so strong because it's a clone of Fruit which does not explain why it is so much stronger than Fruit.
Perhaps you have me confused me with someone else? Repeating something which is not true will not somehow make it so. Incidentally, you agree that all programs benefited from Fruit. What we seem to be debating is the quantum of that benefit. In your statements you claim that it is very little. I am merely asking for the proof of that. You seem to believe that merely asking the question means I am asserting something.
Don wrote: When I say that there is nothing of any real importance in Fruit the context is the fact that Rybka is far stronger than Fruit - it's just no contest. So what I am calling things of of "real importance" is whatever it is that makes Rybka so much stronger than Fruit.
I fail to see how I am taking you out of context but alright....
Don wrote: What you need to explain is how Rybka can be so much stronger than Fruit and yet claim that it is really nothing more than a clone of Fruit and that Vas did nothing but make a copy of something. That is the assertion that you have yet to talk about.
(a) Where did I make this claim? I see, if you keep saying it then somehow it will morph to become true. You make an assertion. I ask for proof and before you know it, I am the one asserting all sorts of things in your mind.

(b) Again with the assertion thing. Don, you plastered persons as being emotional and responding to weak arguments but you are failing to supply anything substantially different. What happens is that you probably hear a voice which is questioning some article of faith that you hold dear and mistake the question for someone attacking Vas or his accomplishments. Really?

I am going to envy the writer of a strong chess program? Why? I am not in that game at all. I might as well envy Messi his footballing genius. At least that game I pretend to play.

(c) I cannot explain why Rybka is stronger than Fruit. I am not technically qualified or able or intelligent enough. Do you recall me asking you to explain precisely the source of that strength difference? Apart from simply asserting that it exists? Or indeed, anyone so willing.
Don wrote: You seem to believe that all you have to do is take the currently existing strongest program and then just add a few hundred ELO and you have created the new strongest program in the world without having to do any work. That's why it's impossible to take you seriously.
Oh boy, now you are hurting my fragile feelings. So you refuse to take me seriously. I admit that I was contemplating suicide for a moment but I think I will pass. For the record, seeing that you are someone who simply states and believes and asserts and tells me what I believe (which I think is arrogant) let me tell you where I am:

I believe that Vas is a gifted programmer. I believe that in Rybka he has created a monstrously strong program that is dominating computerchess at the present moment. I believe that I live at an extremely fruitful (pun intended) time in computerchess when engines such as yours and Crafty and Critter and Gaviota are free to enjoy.

I believe that there has been more smoke than can be reasonably justified or understood, particularly when with a few lines of code Vas could have conclusively discredited all the supposed clones/derivatives once and for all.

That is what I believe and what I have always wanted - a clearing of the air. So you can stop telling me what I believe or what I am trying to say.

I have said it.

Later.
Uri Blass
Posts: 10297
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: BB+ on the matter

Post by Uri Blass »

Milos wrote:
Don wrote:In your own words, "Then there was Rybka, the monster beta that crushed everything." I might add, "including Fruit."
And again you distort the facts. Rybka 1 beta was stronger than Fruit 2.1 exactly for bitboard conversion+10-20elo=60elo.

Rybka is not so much stronger than Fruit as you wish to present. Juri already demonstrated that.
And even though everyone had access to Fruit's ideas, noone but Vas used Fruit's code.
What you are suggesting all the time is a fallacy. You have access to all the ideas of SF still your program is significantly weaker than SF (the strength difference is 2 times what was between Fruit 2.1 and Rybka 1 beta), why?
Because you didn't clone SF. If you did, with ideas from Komodo you would probably surpass R4, maybe even Houdini.
Rybka1 beta was more than 60 elo stronger than fruit2.1
From the CCRL 40/40 list
I can see 92 elo difference even if you consider only 32 bits hardware and
I can see 124 elo difference if you consider 64 bit hardware.

Rybka 1.0 64-bit 2920 +16 −16 68.7% −127.9 34.8% 1383
Rybka 1.0 32-bit 2888 +16 −16 60.8% −72.7 40.8% 1381
Fruit 2.1 2796 +26 −26 51.9% −15.9 37.3% 496

Juri demonstrated nothing because I do not consider strelka as a clone of fruit.
I am not going to discuss the question what people can be consider as a derivative but the facts are that strelka has a lot of code not from fruit.

I also do not believe that
"noone but Vas used Fruit's code."

I am not sure if Vas used Fruit code but even if we agree for the discussion that he did it
I see no proof that other programmers did not use fruit's code and only got 10 elo improvement or even reduction in elo(because using fruit's code does not mean starting from the full program and you can use some part of fruit code but get reduction in playing strength because of inferior move generator that you do not copy from fruit or because of other similiar reasons).
K I Hyams
Posts: 3584
Joined: Fri Mar 31, 2006 11:21 pm

Re: BB+ on the matter

Post by K I Hyams »

Graham Banks wrote:
K I Hyams wrote:
SzG wrote:
K I Hyams wrote: Perhaps it was all a cunning plan, designed to convince idiots like me that he does in fact have principles.
SzG wrote: That's what I intended to suggest.
Well it is always a possibility, I suppose. In fact it seems like useful "evidence" with which to launch a further smear attack on him. Damned if he does seek Nalimov's permission, damned if he doesn't seek Nalimov's permission. I note that nobody else saw fit to make such a slur, perhaps they thought that it was too cheap. Either way, if you don't want to further exploit it yourself, point out the possibility to Graham, he may not have thought of it.

I seem to remember that Houdart made the resources that he used perfectly clear and at an early stage.
It seems like the only time you make appearances in CCC is when it's to take potshots at me.
Use of the forum search facility would prove that suggestion to be incorrect. I have nothing personal against you although I do appear to have a below average tolerance for those who behave in the way that you behave.

In fact, the truth is the exact opposite of your suggestion. I visit CCC with the intention to NOT respond to your posts. I read less than half of those that you write and the motivation for reading those that I do open is to find out which crusade you are on at that particular moment; morbid curiosity.

I enjoy reading intelligent posts and the reason why I came across your latest nonsense is because Matthew Hull writes intelligently. Matthew does not normally take “potshots” at you, and the last time that I replied to a post that you had written was about 6 months ago. Perhaps you should ask yourself why both of us broke our normal posting habits.
User avatar
mhull
Posts: 13447
Joined: Wed Mar 08, 2006 9:02 pm
Location: Dallas, Texas
Full name: Matthew Hull

Re: BB+ on the matter

Post by mhull »

Graham Banks wrote:According to your reasoning, nobody who gets paid for what they do makes a contribution to society. That's crazy. :shock:
A contribution would be something that is worth far more than they were paid. The suggestion that Vas has contributed something that is far greater than his remuneration is hero worship, especially since most of his product was built upon the shoulders of real contributors who never charged a dime for their groundbreaking work.

So no, Vas has not made any contributions to computer chess worthy of the name "contribution". He donates nothing to the computer chess community above and beyond his remuneration.
Matthew Hull
User avatar
Don
Posts: 5106
Joined: Tue Apr 29, 2008 4:27 pm

Re: BB+ on the matter

Post by Don »

mhull wrote:
Graham Banks wrote:According to your reasoning, nobody who gets paid for what they do makes a contribution to society. That's crazy. :shock:
A contribution would be something that is worth far more than they were paid. The suggestion that Vas has contributed something that is far greater than his remuneration is hero worship, especially since most of his product was built upon the shoulders of real contributors who never charged a dime for their groundbreaking work.

So no, Vas has not made any contributions to computer chess worthy of the name "contribution". He donates nothing to the computer chess community above and beyond his remuneration.
Well then this means Shredder has made no contribution, Frtitz, Junior, Sjeng, Naum, Fruit (which went commercial) and going back Genius and the entire Richard Lang line of product or Dan and Kathy Spracklen, nor Belle (because Ken Thompson kept his product to himself), and so on. But this is so ludicrous that it's laughable.

That also means that University professors, teachers of any kind who are payed, physicists and scientist have made no contribution to society because they get payed for their work. I have to ask you what contribution YOU think you are making to society if any.

If you don't even believe teachers are making a valuable contribution to society then try to imagine what things would be like if every one of them were fired from their jobs and we did not have teachers. You make it almost impossible to take anything you say seriously.
User avatar
mhull
Posts: 13447
Joined: Wed Mar 08, 2006 9:02 pm
Location: Dallas, Texas
Full name: Matthew Hull

Re: BB+ on the matter

Post by mhull »

Don wrote:
mhull wrote:
Graham Banks wrote:According to your reasoning, nobody who gets paid for what they do makes a contribution to society. That's crazy. :shock:
A contribution would be something that is worth far more than they were paid. The suggestion that Vas has contributed something that is far greater than his remuneration is hero worship, especially since most of his product was built upon the shoulders of real contributors who never charged a dime for their groundbreaking work.

So no, Vas has not made any contributions to computer chess worthy of the name "contribution". He donates nothing to the computer chess community above and beyond his remuneration.
Well then this means Shredder has made no contribution, Frtitz, Junior, Sjeng, Naum, Fruit (which went commercial) and going back Genius and the entire Richard Lang line of product or Dan and Kathy Spracklen, nor Belle (because Ken Thompson kept his product to himself), and so on. But this is so ludicrous that it's laughable.
One can always claim it's laughable when it isn't. To the extent that these guys published or shared their science, they are contributors. Most of the names you mention here publish exactly nothing for the benefit of the computer science community.
Don wrote:That also means that University professors, teachers of any kind who are payed, physicists and scientist have made no contribution to society because they get payed for their work.
I qualified my remarks to contributions, the value of which is beyond the pay-grade of the teacher/professor/whatever. You can't make that argument for Vas, SMK or the Spraklens. The fun-factor of their products is certainly something, but it's not a "contribution" in any scientific research sense, since they don't publish or otherwise share information.
Don wrote:I have to ask you what contribution YOU think you are making to society if any.
Pointing out fallacious arguments and correcting your understanding of what constitutes a contribution. ;)
Don wrote:If you don't even believe teachers are making a valuable contribution to society then try to imagine what things would be like if every one of them were fired from their jobs and we did not have teachers. You make it almost impossible to take anything you say seriously.
The idea that teachers are a holy order is just a progressive jedi mind trick to prevent people from realizing they are actually filling our kids minds full of BS about American history and the virtues of communism. And if your kids actually learn something useful like reading, writing and arithmetic, well that's purely accidental.
Matthew Hull
Albert Silver
Posts: 3019
Joined: Wed Mar 08, 2006 9:57 pm
Location: Rio de Janeiro, Brazil

Re: BB+ on the matter

Post by Albert Silver »

mhull wrote:
Don wrote:
mhull wrote:
Graham Banks wrote:According to your reasoning, nobody who gets paid for what they do makes a contribution to society. That's crazy. :shock:
A contribution would be something that is worth far more than they were paid. The suggestion that Vas has contributed something that is far greater than his remuneration is hero worship, especially since most of his product was built upon the shoulders of real contributors who never charged a dime for their groundbreaking work.

So no, Vas has not made any contributions to computer chess worthy of the name "contribution". He donates nothing to the computer chess community above and beyond his remuneration.
Well then this means Shredder has made no contribution, Frtitz, Junior, Sjeng, Naum, Fruit (which went commercial) and going back Genius and the entire Richard Lang line of product or Dan and Kathy Spracklen, nor Belle (because Ken Thompson kept his product to himself), and so on. But this is so ludicrous that it's laughable.
One can always claim it's laughable when it isn't. To the extent that these guys published or shared their science, they are contributors. Most of the names you mention here publish exactly nothing for the benefit of the computer science community.
I think you are wrong. You wish to narrow the word contribution only toward teaching others how you did it, but that is a bogus argument IMHO. For example, from Shredder 6 to 8, and the other competing Fritzes and whatnot, we saw progress in the order of 20-30 elo per YEAR in software. It had gotten to the point where this was believed to be what one could expect, and if one made 40 Elo, it was a miracle. By simply trouncing Shredder 8, as well as keeping a breathtaking rhythm of 100 Elo per year for years, Vas proved this wrong. Interestingly, as soon as he had proven this, everyone started doing the same, realizing this thinking was nonsense. Even if he did not explain how, his program was also the first one to demonstrably show how much one could benefit from a 64-bit OS, which until then had been the subject of discussions and not results. Creating and sharing the top program bar-none is not a lack of contribution. It may not teach others how to do it, but it has had a huge positive impact on computer chess and chess itself.
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
Dann Corbit
Posts: 12541
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: BB+ on the matter

Post by Dann Corbit »

Albert Silver wrote:
mhull wrote:
Don wrote:
mhull wrote:
Graham Banks wrote:According to your reasoning, nobody who gets paid for what they do makes a contribution to society. That's crazy. :shock:
A contribution would be something that is worth far more than they were paid. The suggestion that Vas has contributed something that is far greater than his remuneration is hero worship, especially since most of his product was built upon the shoulders of real contributors who never charged a dime for their groundbreaking work.

So no, Vas has not made any contributions to computer chess worthy of the name "contribution". He donates nothing to the computer chess community above and beyond his remuneration.
Well then this means Shredder has made no contribution, Frtitz, Junior, Sjeng, Naum, Fruit (which went commercial) and going back Genius and the entire Richard Lang line of product or Dan and Kathy Spracklen, nor Belle (because Ken Thompson kept his product to himself), and so on. But this is so ludicrous that it's laughable.
One can always claim it's laughable when it isn't. To the extent that these guys published or shared their science, they are contributors. Most of the names you mention here publish exactly nothing for the benefit of the computer science community.
I think you are wrong. You wish to narrow the word contribution only toward teaching others how you did it, but that is a bogus argument IMHO. For example, from Shredder 6 to 8, and the other competing Fritzes and whatnot, we saw progress in the order of 20-30 elo per YEAR in software. It had gotten to the point where this was believed to be what one could expect, and if one made 40 Elo, it was a miracle. By simply trouncing Shredder 8, as well as keeping a breathtaking rhythm of 100 Elo per year for years, Vas proved this wrong. Interestingly, as soon as he had proven this, everyone started doing the same, realizing this thinking was nonsense. Even if he did not explain how, his program was also the first one to demonstrably show how much one could benefit from a 64-bit OS, which until then had been the subject of discussions and not results. Creating and sharing the top program bar-none is not a lack of contribution. It may not teach others how to do it, but it has had a huge positive impact on computer chess and chess itself.
You can cook a fine dinner and feed it to me.
You can show me how to cook it myself.
You can show me how to cook it and then feed me.
You can teach me how to teach others to cook it.
All are contributions related to a meal. I think we have something analogous here.
User avatar
mhull
Posts: 13447
Joined: Wed Mar 08, 2006 9:02 pm
Location: Dallas, Texas
Full name: Matthew Hull

Re: BB+ on the matter

Post by mhull »

Albert Silver wrote:I think you are wrong. You wish to narrow the word contribution only toward teaching others how you did it, but that is a bogus argument IMHO. For example, from Shredder 6 to 8, and the other competing Fritzes and whatnot, we saw progress in the order of 20-30 elo per YEAR in software. It had gotten to the point where this was believed to be what one could expect, and if one made 40 Elo, it was a miracle. By simply trouncing Shredder 8, as well as keeping a breathtaking rhythm of 100 Elo per year for years, Vas proved this wrong. Interestingly, as soon as he had proven this, everyone started doing the same, realizing this thinking was nonsense. Even if he did not explain how, his program was also the first one to demonstrably show how much one could benefit from a 64-bit OS, which until then had been the subject of discussions and not results. Creating and sharing the top program bar-none is not a lack of contribution. It may not teach others how to do it, but it has had a huge positive impact on computer chess and chess itself.
Then I think you should call it an impact, but you shouldn't call it a contribution. Doing so diminishes actual contributions that were freely given. Contributions are charitable. Contribution implies sharing without strings. Contributions by paid individuals are over and above pay grade.

But holding secrets for competitive advantage isn't a contribution by any stretch of the imagination. That's not to say that's bad. But it's not a contribution.
Matthew Hull