Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / And the code is TOTALLY different.
1 2 3 4 Previous Next  
Parent - By Venator (Silver) Date 2012-02-03 13:00
Taking many ideas is not code copying. That seems the problem you are still stuck at.
Parent - By AWRIST (****) Date 2012-02-08 09:06
For me as observer here is the proof that Hyatt is a liar in his campaign against Vas. I write this prior to the appearance of the second part of the Levy interview in CB. Actually on talkchess and openforum Hyatt and Watkins are desperately explaining why it shouldnt be taken as a contradiction in itself if Levy had declared in his first part of the interview that it wouldnt be about verbatim = literal codecopying!! So we have a total breakdown of the whole accusation. Until now (here Feb 1st) it always was about line after line literal codecopying and now (Feb 8th) we have the new concept of non-literal codecopying which should also be forbidden... but let's see what the ICGA mafia will reveil next.

Perhaps it's only about Vas didnt cooperate and showed them a bit of his tricks so that the SMK and Uniacke whiners wouldnt have looked so weak and out-dated with their formerly so masterly engines. Of course Junior all the same. In the meantime Junior 13 is treated like a multi-handicapped blind ex-warrior on rehabilitation: "please challenge me if you have interest in a long timecontrol of 1 move per day with your Houdini or Stockfish against my dear J13, please... - sorry, but that is too funny but it's real from this very day!
Parent - - By Adam Hair (**) Date 2012-02-01 23:08
I will ask you. Taking ideas is okay. But is it okay to take them all from one source? Surely it is possible to take too much from one source. Can we decide how much is okay and how much is too much?

Can I take a book you created, make some changes, and announce it to be my work? It seems to me that I need to do a certain amount of work, a sufficient amount of changes in order to claim it as my own. The question is, how much?

By the way, if I did try to do this, the resulting mess would be such that you would disowned it anyway :grin:
Parent - By Ugh (*****) Date 2012-02-01 23:17
We can make a good stab at deciding how much is ok. Good question. But we won't do it when confronted again and again with ridiculous cut and paste allegations when it is patently clear there is no section of either ASM or manufactured C code that can be placed side by side and matched in any realistic way. There are arguably sections of code based on similar ideas that can, but until we can get past the blocking position there can be no realistic progress.
Parent - - By Rebel (****) Date 2012-02-01 23:51
I think Vas learned the most of Crafty, Bob is the first one Vas thanks in the release notes of Rybka 1, see Dann Corbit's document.

http://www.top-5000.nl/rybka-icga-ban-opinion.htm

Sometimes reality is stranger than fiction.
Parent - - By bob (Gold) Date 2012-02-02 00:32
Totally impossible.  banned for life "knows a guy" that said Crafty's code is "poorly written."  Apparently banned for life doesn't understand enough about programming to reach his own conclusion.  In any case, how could he learn anything from such poorly written code???
Parent - - By Banned for Life (Gold) Date 2012-02-02 00:51
You should learn to read. What you are misremembering was actually a quote from Anthony Cozzie:

Anyway, from an engineering standpoint Strelka (rybka) is fantastic.
It packs most of the search and evaluation of Fruit/Crafty into
something almost 3 times faster.  For those of you that have never
written code, 3 times is quite a bit, considering that Fruit and
Crafty are already reasonably well coded.


As you can see, Anthony does not say that Crafty is poorly written. He says it is reasonably well coded.
Parent - - By bob (Gold) Date 2012-02-02 05:01
I said YOU said it was poorly written.  Learn to read.  Remember what you write.  "you know a guy that looked at the source..."  remember that nonsense?

I'll search for the post if you don't...
Parent - - By Banned for Life (Gold) Date 2012-02-02 23:24
You are misremembering again, a sign of advancing senility. Please search for the post. At least this will keep you off the street.
Parent - By Lukas Cimiotti (Bronze) Date 2012-02-05 09:44

>You are misremembering again


True :lol:
The post he didn't find was written by me: http://rybkaforum.net/cgi-bin/rybkaforum/topic_show.pl?pid=370560#pid370560
Parent - By Banned for Life (Gold) Date 2012-02-04 06:52
How are you doing with that search? It's too bad you're so confused.

The only "guy" I quoted that looked at the Crafty source was Anthony Cozzie, and he stated that Crafty was reasonably well coded but that Rybka 1 Beta was much faster.
Parent - By Banned for Life (Gold) Date 2012-02-04 21:19
You're such a clown! Blaming your senility on others!
Parent - - By Venator (Silver) Date 2012-02-02 15:46 Edited 2012-02-02 15:50
I will ask you. Taking ideas is okay. But is it okay to take them all from one source? Surely it is possible to take too much from one source. Can we decide how much is okay and how much is too much?

Why not? If taking ideas is OK, then there is no question about 'taking too much'. And if you think the rules should be about 'taking too much ideas from one source', it should be in the rules beforehand, clearly defined. And not be decided afterwards.

Can I take a book you created, make some changes

You can play 5,000-50,000 comp-comp games with my book, at home or at Playchess, merge these games into a book, alter some priorities (or even better: have the GUI do this for you), add some own moves and you have your own book. There is nothing I can do about this, nor do I think this is illegal in any way. And I will not take any action or attack you for doing this. I wouldn't even care.
Parent - - By Ugh (*****) Date 2012-02-02 16:05
Well, I dunno. I am somehow offended by the idea of taking all ideas from one source rather more than taking a bunch of ideas from several sources. But I agree that taking ideas is ok, so how do I resolve my unease (of using one source only)? I think it has to be because we are not taking the ideas specifically, but because we are taking the recipe or list of ideas.

So, now we care not about the implementation of the ideas, for they can be implemented any which way in this scenario, but in how much of the list was taken, the how much is too much question, but also now, the weightings given to each idea (which had been abstracted out, but are now, for the purposes of recipe comparison, going to have to be de-abstracted back in again) for we all know that messing with the quantities in a recipe drastically affects the nature of the whole.
Parent - By Adam Hair (**) Date 2012-02-02 18:08
I have to say that I like the recipe analogy. My wife loves to collect recipes, but is a bit timid about cooking new dishes. So, she will ask me to cook them. I will look at the recipe and use the ingredients it calls for, but I do not go by the amounts specified in the recipe. Instead, I use my own sense of taste to judge the correct amounts of the herbs and spices. And I will add or subtract ingredients if I feel it will improve the dish.  At times, the dish I actually make is only superficially related to the recipe.
Parent - - By Adam Hair (**) Date 2012-02-02 17:51
I do not know if Vas took too much from Fruit. But I do know it is possible to take too much from one source. If I took all the ideas from one engine, wrote the code myself, but imparted little or none of my own ideas into the resulting engine, then I have done little more than to translate the original code into my own.

As for rule #2, I believe that it was assumed that it clearly stated the intent. Like many, many other situations, it looks as if the people who formulated that rule did not see this possible occurrence.

For the book example, you give what could be interpreted to be a minimum amount of work necessary to transform the book into my "own". For another book builder, this may not be sufficient. Even for yourself, there may be a point where you would care.
Parent - By Venator (Silver) Date 2012-02-02 18:04
But I do know it is possible to take too much from one source. If I took all the ideas from one engine, wrote the code myself, but imparted little or none of my own ideas into the resulting engine, then I have done little more than to translate the original code into my own.

It has never been about ideas. The issue was code copying. To take your example: Vas took many ideas from Fruit (which he claimed from the beginning) and added 400 elo in 5 years. That is putting A LOT of one's own ideas in the end result.

For the book example, you give what could be interpreted to be a minimum amount of work necessary to transform the book into my "own". For another book builder, this may not be sufficient. Even for yourself, there may be a point where you would care.

There are even easier ways: write a program that substracts all the lines from the book, makes PGN's out of these lines and then merges it into a book. You think I really care at some point? Maybe, yes. A simple 1 to 1 copy, sold somewhere, would be too much. But if one clever guy would use the methods I mentioned to make his own book, sell it or give it away as a download, I would not care at all. In any case, I wouldn't spend years of my life and thousands of postings to hunt such a guy :-).
Parent - By Ugh (*****) Date 2012-02-02 18:16 Edited 2012-02-02 18:18
"If I took all the ideas from one engine, wrote the code myself, but imparted little or none of my own ideas into the resulting engine, then I have done little more than to translate the original code into my own. "

ok, so you postulate taking the sub-ideas, writing them yourself in some different technology, recombining them to give you a "resulting engine", and this gives you a bad feeling.

Fair enough, but, in this case, let's look a little deeper. Say, the original engine has ideas ABCDEFG. Each of those "idea modules" will have input and outputs; since we consider evaluation, the inputs will mostly be some aspects of a chess position, and the outputs will be mostly evaluation scores, we know all weights are different and the modules do things differently as well. Thus we have no input/output matching at this level for any idea module. Nor do we have relative matching between the outputs of the modules. A' B' C' etc are non-matching black boxes with differing inputs/outputs and known different technology. All we know is that they somehow implementing similar ideas in different ways and with different weights and have dissimilar outputs.

We also know that, in this case, our resulting engine doesn't use all of ABCDEFG, it uses some only.

We also know in this case, our resulting engine uses its own ideas, and some ideas from elsewhere, so we have perhaps ABCDEXYZ, where XYZ certainly don't match ideas-wise and ABCDE don't match input/output-wise or technology-wise. Of course we can argue about relative proportions of idea usage, where on the grey scale and so on.

A giant mess, but peering forward to the inevitable conclusion, what else can you see but NOT GUILTY?
Parent - - By michiguel (***) Date 2012-02-02 21:00
Your question is a very good question. Taking ideas from one source? many? is it the same? I do not know, and probably there is no simple answer for ICGA purposes, but the question is relevant. The problem is, rule #2 is terribly worded and was designed eons ago with only cut&paste clones in mind. The part that says "e.g. programs that play nearly the same moves" (or something like that) is extremely naive.

Let's suppose that I write a program following Ed's document about how rebels plays. That would probably be a program that is 80-90% identical. There was zero code copy because I did not even see it. Is that ok? I believe yes... what would be the difference if I saw the code?

Miguel
Parent - - By Barnard (Bronze) Date 2012-02-02 22:46
supongo que me entenderas si escribo en español,porque me expreso mejor que en ingles...

hay una grandisima diferencia en lo que dices,y aqui un ejemplo:

1)voy a un congreso de automoviles de la marca Mercedes-Benz,y en el se explica como se construyen sus coches,sus cualidades,caracteristicas tecnicas,etc...y yo a partir de esa informacion,contruyo un coche

2)ahora lo mismo,pero tomando yo los planos de un coche Mercedes-Benz,donde se detalla tecnicamente pieza por pieza totalmente el montaje del coche y su fabricacion...y fabrico yo el coche con esos datos

no es lo mismo,verdad?en el primer ejemplo,uso las ideas,en el segundo ejemplo,estaria usano los planos(en el caso del ajedrez,estaria copiando el codigo)...asi que no,no es lo mismo el ejemplo que tu pusiste sobre 1)leer el documento de Ed y programar un engine que juegue 80-90 % identico acerca de como juega o 2)leyendo el codigo fuente de Rebel y a partir de ahi programando tu el engine
Parent - - By michiguel (***) Date 2012-02-02 23:53

> supongo que me entenderas si escribo en español,porque me expreso mejor que en ingles...
>
> hay una grandisima diferencia en lo que dices,y aqui un ejemplo:
>
> 1)voy a un congreso de automoviles de la marca Mercedes-Benz,y en el se explica como se construyen sus coches,sus cualidades,caracteristicas tecnicas,etc...y yo a partir de esa informacion,contruyo un coche
>
> 2)ahora lo mismo,pero tomando yo los planos de un coche Mercedes-Benz,donde se detalla tecnicamente pieza por pieza totalmente el montaje del coche y su fabricacion...y fabrico yo el coche con esos datos
>
> no es lo mismo,verdad?en el primer ejemplo,uso las ideas,en el segundo ejemplo,estaria usano los planos(en el caso del ajedrez,estaria copiando el codigo)...asi que no,no es lo mismo el ejemplo que tu pusiste sobre 1)leer el documento de Ed y programar un engine que juegue 80-90 % identico acerca de como juega o 2)leyendo el codigo fuente de Rebel y a partir de ahi programando tu el engine


No tengo una respuesta pero la analogia no me parece que adapte 100% a este caso.
1) los planos de MB no estan disponibles al publico
2) el codigo fuente de un motor no tiene el mismo nivel de detalle tecnico que tu caso
3) estoy asumiendo que se usaron los conceptos y no una copia de los detalles tecnicos, o sea, para ir a tu caso, el plano se uso para ver la forma del auto en lugar de ver las dimensiones y el material de los tornillos.

Tu caso es muy claro, pero en el caso de mi pregunta se cae en un area gris muy extrania, sobretodo porque las reglas de la ICGA nunca fueron claras.

Saludos,
Miguel
Parent - - By Barnard (Bronze) Date 2012-02-03 00:43
bueno,matizare mas aun:

no es lo mismo asitir a un congreso de MB y con la informacion alli publica hacer un coche,que encontrarme los planos del coche en una basura,o robarlos,o conseguirlos como sea,y a partir de los planos,construirlo

en el primer caso,escucharia en ese congreso ideas,y correria a mi cargo la implementacion de esas ideas,cosa perfectamente licita

en el segundo caso,no usaria ideas y las implementaria con mi propio codigo(por usar el lenguaje de programacion de los engines),sino que cogeria directamente el codigo y con este codigo que otro escribio haria el engine...

la cuestion es bien simple:tomar ideas E IMPLEMENTARLAS CON TU PROPIO CODIGO ES LICITO,sin embargo COGER CODIGO DE OTROS Y PONER ESTE CODIGO EN TU ENGINE NO ES LICITO
Parent - - By michiguel (***) Date 2012-02-03 02:06

> bueno,matizare mas aun:
>
> no es lo mismo asitir a un congreso de MB y con la informacion alli publica hacer un coche,que encontrarme los planos del coche en una basura,o robarlos,o conseguirlos como sea,y a partir de los planos,construirlo
>
> en el primer caso,escucharia en ese congreso ideas,y correria a mi cargo la implementacion de esas ideas,cosa perfectamente licita
>
> en el segundo caso,no usaria ideas y las implementaria con mi propio codigo(por usar el lenguaje de programacion de los engines),sino que cogeria directamente el codigo y con este codigo que otro escribio haria el engine...
>
> la cuestion es bien simple:tomar ideas E IMPLEMENTARLAS CON TU PROPIO CODIGO ES LICITO,sin embargo COGER CODIGO DE OTROS Y PONER ESTE CODIGO EN TU ENGINE NO ES LICITO


En este caso en particular, VR implemento ideas con su propio codigo.

Miguel
Parent - - By Barnard (Bronze) Date 2012-02-03 02:26

>En este caso en particular, VR implemento ideas con su propio codigo


bueno,puedes decirme que pruebas tienes que implanto ideas con su propio codigo,si no has tenido acceso al codigo fuente de rybka porque vas se ha negado a facilitarlo,e incluso dijo que perdio todos los codigos fuentes para que no se los pudieran pedir mas?

dime,como sabes que solo tomo ideas y las implemento con su propio codigo si no has tenido acceso al codigo fuente de rybka?y no me digas que lo has desensambaldo,porque si segun los defensores de vas el desensamblado no es suficiente para demostrar que lo copio el codigo,tampoco es suficiente para demostrar que no lo copiase...

dime,has tenido acceso al codigo fuente de rybka para afirmar que vas solo tomo ideas pero las implemento con su propio codigo original 100%?
Parent - - By michiguel (***) Date 2012-02-03 02:49

>> En este caso en particular, VR implemento ideas con su propio codigo
> bueno,puedes decirme que pruebas tienes que implanto ideas con su propio codigo,si no has tenido acceso al codigo fuente de rybka porque vas se ha negado a facilitarlo,e incluso dijo que perdio todos los codigos fuentes para que no se los pudieran pedir mas?
>
> dime,como sabes que solo tomo ideas y las implemento con su propio codigo si no has tenido acceso al codigo fuente de rybka?y no me digas que lo has desensambaldo,porque si segun los defensores de vas el desensamblado no es suficiente para demostrar que lo copio el codigo,tampoco es suficiente para demostrar que no lo copiase...
>
> dime,has tenido acceso al codigo fuente de rybka para afirmar que vas solo tomo ideas pero las implemento con su propio codigo original 100%?


Es al reves, no he visto ninguna prueba que haya usado codigo foraneo para implementar las ideas.

Miguel
Parent - - By Barnard (Bronze) Date 2012-02-03 21:37
miguel,te pongo el mismo ejemplo que le puse el otro dia a otra persona...

es de noche,en la calle no hay nadie,y tu estas paseando...llevas una pistola en la mano,pasa alguien cerca de ti y le pegas un tiro...nadie te ve,no hay grabaciones,no hay testigos,no hay ninguna prueba,pero tu has matado a esa persona...

dime,que no se hayan podido reunir suficientes pruebas te convierte en inocente?
Parent - - By michiguel (***) Date 2012-02-03 22:18

> miguel,te pongo el mismo ejemplo que le puse el otro dia a otra persona...
>
> es de noche,en la calle no hay nadie,y tu estas paseando...llevas una pistola en la mano,pasa alguien cerca de ti y le pegas un tiro...nadie te ve,no hay grabaciones,no hay testigos,no hay ninguna prueba,pero tu has matado a esa persona...
>
> dime,que no se hayan podido reunir suficientes pruebas te convierte en inocente?


Estas asumiendo que conoces la verdad, y no es asi.

Miguel
Parent - - By Barnard (Bronze) Date 2012-02-03 22:24
cierto,no la conozco,ni tu tampoco,ni nadie tampoco...aqui el unico que sabe la verdad,es vas,y no ha hecho nada por ayudar a esclarecerla cuando todo el tema este se descubrio,al reves,se escondio durante meses...no suelen ser precisamente los inocentes los que se esconden y no dan la cara,sino los culpables
Parent - - By michiguel (***) Date 2012-02-03 22:39

> cierto,no la conozco,ni tu tampoco,ni nadie tampoco...aqui el unico que sabe la verdad,es vas,y no ha hecho nada por ayudar a esclarecerla cuando todo el tema este se descubrio,al reves,se escondio durante meses...no suelen ser precisamente los inocentes los que se esconden y no dan la cara,sino los culpables


Cuando aparecio hace unos dias, le tergiversaron las respuestas.

Miguel
Parent - By Barnard (Bronze) Date 2012-02-05 02:41
cuando aparecio hace unos dias,despues de estar convenientemente desaparecido en espera de que desapareciese el temporal,no dio respuestas,sino que dio 'vagas' respuestas y solo a lo que le intereso,no a lo que le podia hacer parecer culpable...

cuando bob le pregunto sobre cosas de su programa que eran imposible que estuvieran ahi si no hubiera copiado codigo directamente,y le hubiera pasado desapercibido que lo copio,no dio explicacion a como es que estaba ese fragmento de codigo obsoleto en su programa,y casualidad,tambien en el mismo lugar del crafty de una version antigua (no recuerdo cual,pero no es el famoso 0.0,es otra cosa),y vas,en vez de dar una explicacion de como en un programa actual,en un codigo actual,aparecia un fragmento de codigo obsoleto identico al de crafty y en el mismo sitio que el de crafty,sabes que respondio?que no le apetecia hablar de eso

dime Miguel,eso es dar respuestas?responder a lo que sabes que no tiene relevancia,y que no te va a hacer quedar mal,pero a las cosas que pueden incriminarte y demostrar que eres culpable decir que no te apetece hablar de eso,es dar respuestas?respondeme
Parent - - By Adam Hair (**) Date 2012-02-04 02:41
Good question in return. I have looked at Ed's document, but without much attempt at comprehending it. I assume the main difference between the collection of ideas in Ed's document and actually seeing the code in Rebel is that the code shows how Ed pieced all those ideas together. Which is important. Using the same ideas but in a different manner constitutes something new. Using the same ideas in the same way is non-literal copying.

Many (most? all?) of the ideas in Fruit are found in Rybka in varying degrees. Rybka includes a few other ideas. Were there enough changes, enough additions to make Rybka into solely Vas' work? I do not know. I do think that several other authors constructed their engines in a similar manner to Rybka (using many of the ideas found in Fruit or Strelka)) and feel that their engines are not examples of plagiarism. Other authors feel differently. Who is right? Who is wrong?
Parent - - By Uly (Gold) Date 2012-02-04 02:45
Ahh, this seems like a grey issue that people want to paint white or black, with endless discussions about the exact grey tone that it has.
Parent - By Adam Hair (**) Date 2012-02-04 12:36
Yes.
Parent - - By bob (Gold) Date 2012-02-01 18:46 Edited 2012-02-01 18:50
I do not understand why this is so hard to understand.

1.  A single piece of the ICGA evidence does NOT strongly imply that code was copied.  No one has said this. 

2.  When you take the eval as a whole, with the >75% overlap, plus the PST data, plus the various eval ramping arrays, and consider all of this together, it tells a pretty compelling story.

you (and a couple of others) want to take each piece, by itself, and compare and if it matches, say "but this is such a small piece of code".  Or if it doesn't match exactly, "this is similar but not an exact match so it doesn't mean a thing."  That's not the way any jury I have served on treats evidence in a court trial.  We try to infer from the whole body of evidence.  We might ignore pieces because of what we would consider to be unreliable testimony, or false testimony, or bogus (non-scientific evidence like, maybe, space aliens or vampires).  But we considered all of the evidence that was credible, and used that to reach a verdict.  We did not spend time going through each piece and saying "OK, by itself, this is not convincing, that is not convincing, etc."  Coincidence can only explain a few such things, sooner or later, coincidence becomes suspicious, and that turns into a verdict of guilt

One has to look at everything, consider ALL the similarities, and then infer how all those similarities came about.  Serendipitous development, or copying.  The former is simply too implausible, and was shown to be so in the ICGA report because no two other programs examined had anywhere near that level of overlap in the eval.  Surely, if this serendipitous simultaneous development happened once, it has happened more than once.  Otherwise, believing it was serendipitous is a real stretch, to the point of implausibility.

BTW, I don't think your last 2 plies of "margin" are going to work.  Based on my testing.  You realize that those numbers are a sort of "error margin" and should almost certainly be larger as you get to 5/6 plies from the tips?  I found some values (roughly +5, but I would have to dig thru a mound of data to be exact) that were workable, but which were no stronger in terms of elo.  I don't want to prune, just to "break even".  Seems a little riskier than to not prune instead if there is no gain.
Parent - - By Rebel (****) Date 2012-02-01 20:37

> 2.  When you take the eval as a whole, with the >75% overlap, plus the PST data, plus the various eval ramping arrays, and consider all of this together, it tells a pretty compelling story.


I have did the same, took the unburden evidence as a whole and Rybka is way too original.

> BTW, I don't think your last 2 plies of "margin" are going to work.  Based on my testing.


Typical you.

I don't understand, so something is wrong.

The ICGA-Rybka fiasco in a nutshell.
Parent - - By bob (Gold) Date 2012-02-01 20:41
Your argument is, to be polite, NFG.

you can't just extract the original parts and then say "these parts show he is original."  And ignore all the parts that were copied.  Copyright law, nor ICGA rules, state that if you copy something, but then add something original to it, that the new "work" is now considered original.  It is not.

"I don't understand..."  :)

You copied MY numbers, developed with millions of games of tests, and "I don't understand"?  :)

I simply commented that they should get bigger as you retreat from tips, because the farther from the tips, the more critical errors become.   But do what you want...  You don't seem to expend much effort in understanding anything anyway...  It is easier to copy, eh?

Where have we seen that attitude previously?
Parent - - By Rebel (****) Date 2012-02-01 23:20
Ed  - moving from 4 to 6 plies futility pruning gave me a good elo improvement.

Bob - no, that can't be because 6 did not work in Crafty, only 4.

Typical you.

What I don't understand I will deny.

And when I finally understand I will deny I ever said it.

Pattern since 1995. And so predictable.

Perhaps I should quote some Peter Gilgasch RGCC postings, he had a special talent to psych you.
Parent - - By bob (Gold) Date 2012-02-01 23:34
I didn't say a thing about "can't be".  I said exactly this:

BTW, I don't think your last 2 plies of "margin" are going to work.  Based on my testing.  You realize that those numbers are a sort of "error margin" and should almost certainly be larger as you get to 5/6 plies from the tips?  I found some values (roughly +5, but I would have to dig thru a mound of data to be exact) that were workable, but which were no stronger in terms of elo.  I don't want to prune, just to "break even".  Seems a little riskier than to not prune instead if there is no gain.

So, where's that "can't be"?  I do see a "I don't think".  Ever considered knocking off the hyperbole, misquotes, and false statements?

Quote whatever you want, would be nice if you actually quote "me" from time to time, if you are going to say I said something.  At LEAST say I said something I actually said?

Talk about "not changing."  You should grab a mirror...

I certainly understand what my code does, and what those numbers mean.  You just chose to copy 'em.  And then say that I don't understand them.  Typical...
Parent - - By Rebel (****) Date 2012-02-01 23:44

> I didn't say a thing about "can't be".  I said exactly this: BTW, I don't think your last 2 plies of "margin" are going to work.  Based on my testing.


Correct, it's what you said.

Well, it works.

Despite your testing :razz:

And now we will see the start of the process in which you will deny you ever said the above.

Fixed pattern since 1995.
Parent - - By bob (Gold) Date 2012-02-02 00:29
Nope, I said exactly what I claimed I said, which was exactly what I quoted.  NOT what you claimed I said, if you look.  And I tested various combinations with 30K games per value change.  I am sure your "testing" is that accurate and methodical?

:)

I will stick by my results here...  since I know they are trustable for my program's code...
Parent - - By Rebel (****) Date 2012-02-02 06:55
The truth is pretty simple, I found a way to make progress from 4 to 6, while you are stuck at 4.

And it won't surprise me if modern programs are doing more than 6.
Parent - By bob (Gold) Date 2012-02-02 13:56
I'm not "stuck at 4".  You speak without information.  I simply gave you an opinion.  You might want to think about why I suggested that my 1.5,1.5,3,3 does not translate to 1.5,1.5,3,3,3,3 when you go to 6? 

In any case, what you "said I said" and what I actually "said" were not the same thing.  That was the point.  As always...
Up Topic Rybka Support & Discussion / Rybka Discussion / And the code is TOTALLY different.
1 2 3 4 Previous Next  

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill