Have you decided why Vas

[Kurt Utzinger]

Who cares about that number, really.

I do not care
Kurt

[Dr.Ex]

[d] 2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 11

How long does it like to play Kb8 in this position? I mean, come on. Isn't it stupid to walk directly into an attack?

[Dr.Ex]

Rybka is IMHO a brutal effective and fast searcher. It is as dumb as Toga for example. It has very little knowledge about chess.

I don't know about older Rybka prior to Rybka 3 but for Rybka 3 you are simply wrong.

Also giving one single position where Rybka does not do well** and saying that because of that Rybka is stupid, then obviously you are wrong again.


**And in fact Rybka 3 does do well in this:

1k1r3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 w - - 0 12

It thinks this position, which is easily won for white, is better for black.
Just like all other dumb engines like Toga etc. think it is better for black.
That says it all.

That does not says it all since Rybka 3 believes white is better:

BTW i'm using a slow single CPU computer with 32-bit.
Rybka 64-bit with an Octal Skulltrail QX9775 would be around 16 times faster, but this is irrelevant anyway as even my slow Rybka 3 32-bit finds the solution instantly and gives a score suggesting that white is winning.
So you were wrong.

Analysis by Rybka 3 1-cpu 32-bit :

1.b4-b5 a6xb5 2.Ba4xb5 Rh8-f8 3.Qd1-e2 Nh6-g4 4.Nb1-d2 Be7-a3 5.Nd2-b3 Ng4xe3 6.Qe2xe3 b7-b6 7.Ra1-e1 Rd8-e8 8.Nb3-c1
= (-0.05) Depth: 11 00:00:01 0kN
1.b4-b5 a6xb5 2.Ba4xb5 Rh8-f8 3.Qd1-e2 Nh6-g4 4.Nb1-d2 Be7-a3 5.Nd2-b3 Ng4xe3 6.Qe2xe3 b7-b6 7.Ra1-e1 Rd8-e8 8.Nb3-c1
= (-0.05) Depth: 12 00:00:01 1kN
1.b4-b5 a6xb5 2.Ba4xb5 Rh8-f8 3.Qd1-e2 Nh6-g4 4.Nb1-d2 Be7-a3 5.Nd2-b3 Ng4xe3 6.Qe2xe3 b7-b6 7.Ra1-e1 Rd8-e8 8.Nb3-c1
= (0.07) Depth: 13 00:00:09 176kN
1.b4-b5 a6xb5 2.Ba4xb5 Bf5-e4 3.Nb1-d2 Nh6-g4
= (0.19) Depth: 14 00:02:13 2704kN
1.b4-b5 a6xb5 2.Ba4xb5 Nh6-g4 3.Be3-c1 h7-h5 4.h2-h3 Ng4-h6 5.Qd1-a4 Rd8-e8 6.Bc1-a3 Be7xa3 7.Nb1xa3 Nh6-g8 8.Bb5xc6 Qd7xc6 9.Qa4xc6 b7xc6 10.Ra1-c1 Ng8-e7 11.c3-c4
+/= (0.28) Depth: 15 00:05:38 6885kN

+0.28 for white according to Larry Kaufman means that Rybka 3 believes white's expected score is around 62% for this position.
Or that white's winning chances are 24% more than the winning chances of black.

No it doesn't do well.

Analysis by Crafty 20.14:

12.b5 axb5 13.Lxb5 Sg4 14.Ld2 Lg6 15.exf6 gxf6 16.Da4 Le8 17.f5 e5 18.h3 e4 19.hxg4 exf3 20.gxf3
+/= (0.69) Tiefe: 16/24 00:00:21 47112kN

You obviously don't know what you are talking about.

You said Rybka believes that black is better in this position.
I have showed you that what you were saying was WRONG!
Didn't you see that? Why you avoided answering this?

No, it was not wrong at all. It should evaluate this position as clearly better for white from the get go. Not after deep search. In this Blitz game it evaluated the Black position as -0.50.

Also you don't know what 0.69 for Crafty means and what 0.28 for Rybka means. I will not try to explain. You should find it yourself....

I also said that even IF Rybka didn't believe that white is better, then this is just a single position.
And from a single position you generalized that Rybka is stupid.
No comments needed for that.

Nope, I have played countless games against the previous version.

[Jim Walker]

b5 Game over

Yes if black is Elo 1200.

Black was Rybka 3. Don't you see there is no counterplay at all for black?
Black will be crushed on the Qeenside, no matter how he plays.

Yes but for me to win with white black must be 1200 Elo !

[geots]

Anyone who would dredge this worn-out topic back up from the depths of hell is completely shit out of anything to do, i.e.- W_H_O -- C_A_R_E_S

[bob]

Have we not already discussed this matter 100 times?

I believe Vas can show what he decides to.
- Vas is not running an academic project.
- Vas has not signed a UN treaty to show NPS in certain way.
- If Vas is trying to hide something so be it. He has already said that he does it in his own way. Vas is not following any other programs or oracles. Just look around how other commercial businesses run.

Commercial entities _do_ use a standard dictionary of terms however. A pipe 1" ID is the same around the world. A bolt of size 8mm x 35 mm x 1.75 pitch is the same around the world. A bolt is a bolt. A switch is a switch. A node is a node. So that explanation just does not "cut it".

- If you don't like Vas's method don't buy his program.
- And if someone has not bought his program, it's seems strange that person sneers at Vas every chance he gets.

Where have I "sneered" at him. He has written an outstanding program. But not everything he does is "golden". Of course he can obfuscate whatever he wants. He can fix his engine so it only gives the best move in a position, no score, no PV, no nothing, if he wants. And people are going to discuss why...

This dislike for Rybka from certain persons have been obvious from very beginning, almost showing signs of jealousy.

Don't know (nor care) who you are talking about there. I've been the best in the past. I've been passed more than once. That happens. Happened in the 80's. The 90's. It will happen again to myself or others in the 2000's.

All of a sudden Vas comes to the computer chess and beats programs (20-0) that been around 30 years. And Vas does it in his own way. For some author this can be frustrating. Instead of showing any sort of acknowledgment for Vas, who has taken computer chess to a new level, lets pounce on the only thing that can be pounced on (compared to author's own program) every time, time and time again.

Sorry, but I simply answered a question that was asked. Nothing more, nothing less...

[LarsA]

Has anyone here decided why Vas has chosen not to show the true kns in Rybka? Some of the Vas salesmen here are saying that Rybka 3 is showing even a slower kns than Rybka 2.3.2a. We all know 3 has more knowledge thus slower but Vas has still chosen to mask the kns. Does anyone have any idea why Vas decided to deceive us?

5250 kns Deep Fritz
4780kns Glaurung 2.0.1
4000 kns Deep Sjeng
2300 kns Deep Shredder 11 x64
1740 kns Zappa Mexico II
1450 kns Deep Hiarcs Paderborn
425 kns Rybka 2.3.2a mp x64
375 kns Rybka 3

[d] rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq -

This is just ridiculous. The reason for the "deceit", as you call it, is obviously that Vas is trying to hide rybka's secrets. As a sole developer trying to support a family in the marginal field of engine making, keeping these secrets is incredibly important.
Some protocol writer decided some years ago that he wanted the user to see his engine's kns. Vas thinks this is revealing too much and scales the number down, it's as simple as that. If you don't like it, don't buy the engine. Welcome to the world of competitive business.

[Rolf]

All of a sudden Vas comes to the computer chess and beats programs (20-0) that been around 30 years. And Vas does it in his own way. For some author this can be frustrating. Instead of showing any sort of acknowledgment for Vas, who has taken computer chess to a new level, lets pounce on the only thing that can be pounced on (compared to author's own program) every time, time and time again.

Sorry, but I simply answered a question that was asked. Nothing more, nothing less...

In my eyes an academic who writes this must follow a dishonest path of agenda. Because there is no such thing as unconditional answering without hidden motivation. And by chance I remember the true reason for your sort of anti-talk against Vas. You've told me yourself time ago.

The reason for you is that Vas had visited CCC, discussed a lot, but above all learned a lot, just because others like you shared their knowledge, what is for you as academic professor the most natural thing to do unless you are not in a double bind with IBM people who cheated on Kasparov. Psychologically cheated him, but you didnt agree because yxou are not a psychologist and therefore dont undewrstand what I was talking about a research client who they cheated because suddenly they tranformed the research thing into a dirty secret service like process. You said but he was fool enough not to protect himself against such methods and I contradicted you by showing that there is no way for a proud chess master to suddenly reflect on cheating by otherwise sypathetic people. If Kasparov would have known this he wouldnt have played at all because without the nominally best player all their cheats wouldnt have made sense.

Many people have shown why your take against Vas is premature. You yourself didnt defend science against your friends when they were making money with IBM but you pretend that now you must do so in case of Vas and that doesnt simply fly. Becausae as I told you, Vas is probably, besides yourself, the most interactive programmer who does communicate with everyone who asks questions. He gave away for free most of his progs, just like you must do for academic reasons and the law. But he wouldnt be forced to do so but he does.

But I contradict you for another reason too. We once had the debate about secret service and or being in service for the industry and I am certain that you know what this means for academics too. That then science is second ranked after the first which is military or the country's benefits.

And now you dislike a programmer who has made extreme profits out of what he had learned here and nobody else could compete. And in such a situation instead that you admire him you begin to support people who ask nasty questions. While you know better than me that Vas' advantage isnt coming from his hiding of knodes or such nonsense. That is in truth only the propaganda method to nag Vas and to put him down. Because most here, except you, dont even know how exactly such a prograsm is working. Never could I read such stupid propaganda questions from other collegues so that I can conclude that this is a consequence of a sort of anti free speech debating culture.

It disturbs the whole group process if our senior big one puts himself into opposition aganst the best programmer around for reasons of low motivated envy - and that as an academic who should show himself as the best available role model of the field.

Excuse me for the clarity but please try to reconsider your position.

This was written in one rush so that typos are inevitable. Sorry to all.

[bob]

All of a sudden Vas comes to the computer chess and beats programs (20-0) that been around 30 years. And Vas does it in his own way. For some author this can be frustrating. Instead of showing any sort of acknowledgment for Vas, who has taken computer chess to a new level, lets pounce on the only thing that can be pounced on (compared to author's own program) every time, time and time again.

Sorry, but I simply answered a question that was asked. Nothing more, nothing less...

In my eyes an academic who writes this must follow a dishonest path of agenda. Because there is no such thing as unconditional answering without hidden motivation. And by chance I remember the true reason for your sort of anti-talk against Vas. You've told me yourself time ago.

Rolf, believe what you want to believe. Someone asked a _factual_ question. I gave a _factual_ answer. It doesn't matter to me whether you believe that or not. This is a topic for chess programmers more than anything, because _we_ understand what an engine reveals about its internal design when it displays data at the end of a search. And we also realize that if that is the one thing that sets you apart from the rest, then certainly you want to keep it to yourself. I have kept _many_ things to myself until after a major event has completed, because I wanted to have my new idea as an advantage, knowing that others had come with their new ideas as their advantage.

But once someone that understands engine design gets interested, it is _very_ difficult to hide new ideas for a long time. I explained my book-learning approach and wrote a paper about it. But anyone could have figured out the basic ideas with just the executable file and the ability to play some games. And when you think about how it works, if you understand it at all, you'd see how easy it would be to figure it out even if you didn't understand it initially.

So there is no motive involved in my answering a question, other than the motive to actually answer it and move on..

The reason for you is that Vas had visited CCC, discussed a lot, but above all learned a lot, just because others like you shared their knowledge, what is for you as academic professor the most natural thing to do unless you are not in a double bind with IBM people who cheated on Kasparov. Psychologically cheated him, but you didnt agree because yxou are not a psychologist and therefore dont undewrstand what I was talking about a research client who they cheated because suddenly they tranformed the research thing into a dirty secret service like process. You said but he was fool enough not to protect himself against such methods and I contradicted you by showing that there is no way for a proud chess master to suddenly reflect on cheating by otherwise sypathetic people. If Kasparov would have known this he wouldnt have played at all because without the nominally best player all their cheats wouldnt have made sense.

That is an operational mode that I don't like, yes. But then I grew up in the fun years of computer chess where we all discussed ideas with each other and never had "secrets" that lasted beyond the next ACM or WCCC tournament. But what that has to do with this issue is beyond me as they are completely unrelated.

Many people have shown why your take against Vas is premature. You yourself didnt defend science against your friends when they were making money with IBM but you pretend that now you must do so in case of Vas and that doesnt simply fly. Becausae as I told you, Vas is probably, besides yourself, the most interactive programmer who does communicate with everyone who asks questions. He gave away for free most of his progs, just like you must do for academic reasons and the law. But he wouldnt be forced to do so but he does.

I can't follow that rambling line of thought. What does "making money with IBM" have to do with anything here? Hsu and Campbell were _always_ open with what they were doing, if that is your angle of attack. Hsu called and asked for a copy of my Dissertation as they were looking for ideas for their two-level parallel search. They sent me copies of things they had written. Things like the singular extension paper, etc... So where, exactly, is this going since they have been inactive for over 10 years now???

And what is that about "for academic reasons and the law?" I could sell a version of crafty if I wanted. In fact I helped a software company write a small chess program a year or so ago and was paid for it. Not a soul cares. So that I really don'[t follow. We have people that leave the university all the time and start private companies based on something they discovered.

But I contradict you for another reason too. We once had the debate about secret service and or being in service for the industry and I am certain that you know what this means for academics too. That then science is second ranked after the first which is military or the country's benefits.

And that is related to this exactly how??

And now you dislike a programmer who has made extreme profits out of what he had learned here and nobody else could compete. And in such a situation instead that you admire him you begin to support people who ask nasty questions. While you know better than me that Vas' advantage isnt coming from his hiding of knodes or such nonsense. That is in truth only the propaganda method to nag Vas and to put him down. Because most here, except you, dont even know how exactly such a prograsm is working. Never could I read such stupid propaganda questions from other collegues so that I can conclude that this is a consequence of a sort of anti free speech debating culture.

Now you go _way_ too far. I don't "dislike" Vas. I have never even met him. I didn't like the non-commercial / commercial flip-flop. But it wasn't the first, and won't be the last. But "dislike" is absolutely _not_ the right word.

It disturbs the whole group process if our senior big one puts himself into opposition aganst the best programmer around for reasons of low motivated envy - and that as an academic who should show himself as the best available role model of the field.

What does "envy" have to do with this? I followed Slate and Atkin around for years before I passed 'em. I followed Ken for a few years before I passed him. I used that as motivation, not as envy, at least in a bad sense. So where does that come from?

Excuse me for the clarity but please try to reconsider your position.

This was written in one rush so that typos are inevitable. Sorry to all.

what is to "reconsider"? Nothing I wrote was speculation, it was all simple fact based on lots of analysis done by others when this story first broke. And no, I didn't "break" the story as I don't own a copy of any commercial program and won't own one. So I should "reconsider" stating facts and instead do exactly what???

[bob]

NPS = total nodes searched / total time used

it is just as simple as that...

If this is a de facto standard among chess programmers, it is still not completely clear to me:

(1) What about the positions at the end of the variations each, after the last moves? In other words, those where no more moves are being made, at a certain point during calculation time? By common sense, I think a node is something like a street junction where something branches off, still.

The classic definition of a node could be stated as "a node is what you get when you make a move and update the game board to a position that is not the same as it was before you made the move." It doesn't matter whether a node has branches leading out of it or not, all that matters is if you actually follow any of those branches, and to do so you just make the move the branch (arc) represents, and that takes you to a new node...

(2) Also, I have read that chess engines make pseudolegal moves, internally. Maybe that leads to a similar question for the node count, if the resulting positions of such moves are discarded (as I assume), and not evaluated or at least not searched deeper.

That's a matter of opinion and it is such a small fraction of the nodes searched that it doesn't really matter. Some programs only search legal moves. Which would mean that the ones searching illegal moves might search more total nodes. Or not. But so long as you increment the counter when you make a move, you will be in exactly the right neighborhood...

(3) Another possible distinction which would make sense to me is: Decide if hash table hits are counted or not. Because if the kN/s should serve as a performance figure, I am not sure if these should be included, or only those where the evaluation has to be calculated.

Again, each call to MakeMove(), or as an alternative, each call to Search()/Quiesce(), is a node. If you make a move, visit a new node, and get a hash hit, that's all you count. No way to know how much work that hash hit prevented you from doing and trying to count that.

(4) Same for tablebase hits.

Same answer also.