Fritz 15.0

[bob]

Fritz 15 is not based on Fritz, it's based on Rybka. It is the Rybka 5. Only the name changed, for marketing reasons. So it should be compared to Rybka 4.1, not Fritz 14.

And Fritz 14 is not Fritz either, it's Pandix, a different engine by a different author. The last Fritz by Franz Morsch was Fritz 13. F14=Pandix F15=Rybka. What next ? Seeing that they are still 200 elo behind SF, and that bridging the gap is not going to be possible by conventional means, they might decide to make F16 a SF clone. After all, Rybka started from Fruit, why not restart from SF ? And chessbase is selling free syzygy for an extortionate price, why not sell free SF too ?

Hello Lucas,

I am concerned about that last point.

Stockfish's licence explicitly indicates that it can be sold as long as the source is supplied.

Why (I know, I know, freedom of ideas etc.) would you and the SF team include that clause for a commercial juggernaut like Chessbase to take advantage of?

Just curious...

I had asked Ronald about the selling of his terrific, free bases by Chessbase, but he is of a similar bent as yourselves.



Later.

I think Lucas, Ronald and others recognize that there is "legally right" and "morally right". The two are NOT the same thing in many cases. It is absolutely egregious to sell EGTBs for much more than the medium costs, etc., as the idea is you are saving someone the hassle of downloading a ton of data. But not THAT much. As far as selling SF goes, they have not minded selling Houdini don't forget, a clear ripoff of the ippolit/robolito completely free engine that might be a ripoff of another engine for all we know...

Have you ever walked down a major street in Net York, been accosted by all the camera store salesman standing in the doorways, those that want to sell you a $500 f1.2 50mm "Zeiss" camera lens that is a pure ripoff made in China? Sleazy is as sleazy does...

Houdini a rip-off of Ippolit/robolito? Isn't Houdini 4 still stronger than the strongest Ippolit and Robolito despite the fact that H4 has been out for two years?

I'm very glad to hear that indeed Fritz 15 is a uci engine. Judging from the result of the 1CPU test from Ingo it probably is the long-awaited Rybka 5. Hopefully Fritz 16 will surpass Gull 3, maybe even Houdini 4.

Houdini was a direct ripoff. You can find much discussion and disassembly of early versions and direct comparisons to ippo/robbo on open chess....

Nothing says one can't copy and then improve. It has happened multiple times in fact..

[mcostalba]

Why (I know, I know, freedom of ideas etc.) would you and the SF team include that clause for a commercial juggernaut like Chessbase to take advantage of?

If Chessbase wants to derive FXX from SF they have to release the sources together with it, and I don't think releasing the sources makes a lot of sense for a commercial entity and also I don't think they will take the risk to expose them to a legal action in case they take SF and just release the binaries.

What they can do is looking into SF sources for new ideas and hints, but this is already what everybody does, including of course Komodo (by they authors admission). And this is perfectly fine with me BTW.

Competing with an open source engine against closed ones of course gives you some clear starting handicap and in the past many people thought this idea was crazy. Today many accepts that the trade off that a lively community development (that requires, by its own nature open sources) out weights the handicap of sharing the info with competitors in just one direction, and I agree totally with this view: in the long term it is very difficult for the same single/small team to keep the pace because lone developers get tired/reduce focus in the years, while in a healthy community there is always some fresh mind coming in when a veteran get tired.

For instance, it is a long time that I lost my energy on finding and testing for improving ELO, today I am almost only focused on ancillary and code maintenance stuff, because this is what I like more and requires less time and effort. Luckily new faces stepped in with passion and fresh minds. And it is not only me. I have seen in these years that the heydays of valuable contributors have a life span of about 1-2 years after this period they usually lose interest. More time I spend on chess engines and more I appreciate the value in keeping focused for long years, and among these people Bob is clearly the champion, so he has all my respect for this, although I can say (hoping without hurting him) that probably the energy and mental focus is very distant from his heydays now.

That's why I think in the long term open source will win: a live open source community lives longer that each one of its single components.

[whereagles]

Marco, are there people actively contributing to SF elo improvements? Seems like current focus is like 100% SMP stuff

[reflectionofpower]

Trust me they will not sell a cheaper UCI only engine!

yeap,I 2nd that

[Leto]

It seems to be out now. My partner at CEGT is already testing the 1CPU version for CEGT Blitz. I just checked the website and it seems the download version is out now I'm downloading it at the moment. But I haven't received my activation key, should I have gotten it by now, I preordered on the 17th?

[Roger Brown]

Why (I know, I know, freedom of ideas etc.) would you and the SF team include that clause for a commercial juggernaut like Chessbase to take advantage of?

If Chessbase wants to derive FXX from SF they have to release the sources together with it, and I don't think releasing the sources makes a lot of sense for a commercial entity and also I don't think they will take the risk to expose them to a legal action in case they take SF and just release the binaries.

What they can do is looking into SF sources for new ideas and hints, but this is already what everybody does, including of course Komodo (by they authors admission). And this is perfectly fine with me BTW.

Competing with an open source engine against closed ones of course gives you some clear starting handicap and in the past many people thought this idea was crazy. Today many accepts that the trade off that a lively community development (that requires, by its own nature open sources) out weights the handicap of sharing the info with competitors in just one direction, and I agree totally with this view: in the long term it is very difficult for the same single/small team to keep the pace because lone developers get tired/reduce focus in the years, while in a healthy community there is always some fresh mind coming in when a veteran get tired.

For instance, it is a long time that I lost my energy on finding and testing for improving ELO, today I am almost only focused on ancillary and code maintenance stuff, because this is what I like more and requires less time and effort. Luckily new faces stepped in with passion and fresh minds. And it is not only me. I have seen in these years that the heydays of valuable contributors have a life span of about 1-2 years after this period they usually lose interest. More time I spend on chess engines and more I appreciate the value in keeping focused for long years, and among these people Bob is clearly the champion, so he has all my respect for this, although I can say (hoping without hurting him) that probably the energy and mental focus is very distant from his heydays now.

That's why I think in the long term open source will win: a live open source community lives longer that each one of its single components.

Hello Marco,

Thank you for your response.

I understand it, I do.

I have corresponded with people like yourself, Bob, Tord and Fabien so I understand the open source philosophy.

I guess I find the thought that a commercial entity can rip free software and make a huge profit, repugnant.

And yes, people like Bob and Miguel and Jon are champions to keep doing this after all these years...

Later.

[shrapnel]

It seems to be out now. My partner at CEGT is already testing the 1CPU version for CEGT Blitz. I just checked the website and it seems the download version is out now I'm downloading it at the moment. But I haven't received my activation key, should I have gotten it by now, I preordered on the 17th?

Let us know if its stronger than Komodo and Stockfish. If its weaker, I'm not interested in buying it.

[AdminX]

I am interested in learning more about these two points.

[Frank Quisinsky]

Hi Marco,

congratulation for that what you do with all the supporters around SF.

Only this one:
It's absolutely clear that a work by a Group of People will be strong. Have a look what we do with Arena and over 200 supporters in Arena development time. Or what we do with Winboard, later UCI. It was a hard work in the beginning of the protocols to give the Information ... please programmers, used the free protocols. At this time the Commercials are very strong with his own not free available protocols. OK, winboard is much older as the Commercials protocols but in the beginning a hand full used it only. Thinking on the mails Tim Mann sent ... please Frank, stopped it. I have to many work to added daily all the engines in my list of compatible engines if you make such a Publicity on your Winboard site. At this time ChessBase need a banner on the own Winboard site for his own new Publicity for the new webpage and server.

Same for Arena ...
OK, the sources aren't free but each one can give the ideas and Martin try to implemented it.

Commercials are very ignorant and arrogant at this time. I read here from ChessTiger programmer ... never an amateur can beat one of the Top engines.

For that reason I like SF development, I supported Arena, Winboard, FRC, UCI, DGT and so one.

Right is ...
The work by a group is allways much stronger as the work by a lone fighter. But the work by lone fighter is very interesting too. Today many engines with own ideas are around 200 - 400 Elo stronger as commercials in the past.

OK, with all the public ideas in combination with own ideas much more is possible.

Keep up the good work around SF.

Best
Frank

All the fast draw games with standard settings I dislike here. But to 95% I like SF and the nice style the engine produced on such a high and gigantic niveau. You are right with your opinions ... if I am thinking on our interview.

[shrapnel]

What they can do is looking into SF sources for new ideas and hints, but this is already what everybody does, including of course Komodo (by they authors admission). And this is perfectly fine with me BTW.

The Komodo authors certainly did admit it. What you've forgotten to mention is that Mark Lefler, Lead Programmer of the Komodo Team, went on to say in that particular post that he had discovered something that greatly increased Komodo's strength ( or words to that effect) and that he was surprised that the huge Stockfish Team hadn't discovered it yet !( or words to that effect).
He also implied that as he was working on a Commercial Engine, he saw no reason to share his discovery Publicly !
What do you feel about that ?