Naum 4.6 (64-bit only) free to you all!

[Ron Murawski]

Good news...

And thanks to Alex Naumov and Graham Banks !

Btw, as far as I know,
Naum 4.2 seems to be derivative or clone of Rybka
So...Naum 4.6 will be tested in SCCT, only in case of +100 Elo over Rybka 3

Thanks for understanding!)

My best wishes,
Sedat

First I heard of that. If my memory serves me correctly Naum was on the scene before Rybka?

Hey man !

May I help you ? Both (Rybka & Naum) are Fruit derivatives !



http://computer-chess.org/doku.php?id=c ... ngine_list

Like bonus (I don't want a psycho breakdown off CCC members ) a full carved in stone (very hard work) & old monastery of IASI:



Best regards,

Herr Silvian von IASI

I did not see a convincing evidence for these claims.

Hi Uri,

Do not confuse the terms 'derivative' and 'clone'. A derivative engine starts out by modifying the source code of another engine. In many cases the code morphs enough to be considered as acceptable by the community. Naum is one of those engines.

In the past Alex posted that his first Naum code was based on Fruit. That is when I added the note about it. He has never stated that his latest code is no longer based on Fruit. Perhaps the engine has been totally rewritten and is 100% original. All I know is that the engine originally was based on Fruit code.

FYI: No engine known to be a clone is listed on the Chess Engine List.

Ron

[reflectionofpower]

I compiled the FRC version although you should test it first to see if it works properly.
As for the post on chess2u.com and since Ray made me dig out the source code and compile it, I also compiled 32-bit version and included linux and mobile versions of 4.2 in the package. It would me to much work to make linux/mobile compiles of 4.6 and demand for those versions is small anyway.

http://speedy.sh/UTcRK/naum46.rar

Regards,
Alex

I have always been into compression and still am.

This is just for interests sake. I am not complaining about your work.

naum46.rar ------------- 1,664 KB (Graham's file)
naum46.rar (RAR5) ----- 1,221 KB
naum46.arc -------------- 1,145 KB
naum46.7z --------------- 1,105 KB
naum46.rk --------------- 989 KB

The RAR5 archive format does better compression than RAR standard.

ARC is nice at level 9 compression.

7z is consistently beating out the RAR5 and is just as fast from what I have been seeing for a while now.

WINrk utilizes multiple cores to compress/decompress and does take longer but it is way faster than PAQ or KGB.

If you're a compression nut then download PeaZip program ( http://bit.ly/1vuQbcu ). It has some of these compression schemes as well as some other exotic ones and it can integrate into the explorer shell.

[Guenther]

...

In the past Alex posted that his first Naum code was based on Fruit. That is when I added the note about it. He has never stated that his latest code is no longer based on Fruit. Perhaps the engine has been totally rewritten and is 100% original. All I know is that the engine originally was based on Fruit code.

FYI: No engine known to be a clone is listed on the Chess Engine List.

Ron

Sorry Ron, this cannot be true. I never saw Alex ever saying he started from a Fruit base for whatever version.
If that would have been true he would have been forced realeasing the sources and no way he could have added his own license as he did.
I also found nothing about this in the readme files I still have for some versions.

And BTW have you followed the thread about Gojemat? I still saw it listed lately at the Wiki list.

Guenther

[ernest]

The author, Alex Naumov, has asked me to provide the download link for you all.

Strange!, still yet no post on Naum 4.6 performance...

So I publish my small test, Naum 4.6 vs former Naum 4.2 (all 64-bit, of course)
Dual core, TC 2'+1", 5-men Nalimov, 5-move Book (PB5m)

256 games
For Naum 4.6: +69 -21 =166 or 152-104 (59.4%)

Nice improvement (almost +70 "family" Elo) !!!
Let's see what the big rating guys will find...

[Ron Murawski]

...

In the past Alex posted that his first Naum code was based on Fruit. That is when I added the note about it. He has never stated that his latest code is no longer based on Fruit. Perhaps the engine has been totally rewritten and is 100% original. All I know is that the engine originally was based on Fruit code.

FYI: No engine known to be a clone is listed on the Chess Engine List.

Ron

Sorry Ron, this cannot be true. I never saw Alex ever saying he started from a Fruit base for whatever version.
If that would have been true he would have been forced realeasing the sources and no way he could have added his own license as he did.
I also found nothing about this in the readme files I still have for some versions.

And BTW have you followed the thread about Gojemat? I still saw it listed lately at the Wiki list.

Guenther

Hi Guenther,

Gojemat was removed from the list as soon as I read your posts about it. Look in the 'Wiki Page Changes' section from Sept 30
http://computer-chess.org/forum/index.p ... ad&id=2920

It seems there is too much confusion when I mention 'derivative'. I have a much looser definition than you do. I did not mean to imply that Naum is built on Fruit code. The original engine that Alex worked with (which may have never been released) was based on Fruit. I am certain I read that somewhere.

After thinking about this, I have decided that the comment is unfair to Alex. I will remove it. To Alex: I am sorry and I regret adding the comment.

I'm not perfect and I don't claim to be. If you (or anyone else) notices anything that seems wrong on the Computer Chess Wiki, then let me know about it and I will fix it. The CC Wiki is a huge site and it's difficult to make certain every fact is correct.

Best regards,
Ron

[Guenther]

...

I'm not perfect and I don't claim to be. If you (or anyone else) notices anything that seems wrong on the Computer Chess Wiki, then let me know about it and I will fix it. The CC Wiki is a huge site and it's difficult to make certain every fact is correct.

Best regards,
Ron

Ron, keep up the good work with the CC Wiki pages!

I know how difficult that work is and from time to time I think I should revive my RWBC chronology pages, but each time I shrink back from the amount of work it would need to update after four years meanwhile.
(and not to speak about the investigations needed to verify the programs from illegal derivates/clones, which sure are more now, than in the past)

Best regards,
Guenther

[CositasBuenas]

That's really great. Thank you! But could anyone provide the community (me) with a linux compilation? It would be much appreciated.
Michael

[Ovyron]

Sorry for the bump, I just had to thank Alex for his job on this engine, as I didn't know this version was released until today.

I've been using Naum 4.2 on a daily basis for my analysis of correspondence games, and have found that it's a very useful engine that very often "out-evaluates" the Stockfishes, the Komodos, the Houdinis, the Critters and the Rybkas, being able to hold on to the analysis of a move that turns out to be best that no other engine liked.

It's also very useful in drawish positions, showing useful values and original selection while other engines are happy showing 0.00 for many move alternatives which is absolutely useless for analysis. Naum bests them.

It's great to see another release of a superb engine, hopefully people are able to realize it and use it for more than just engine matches.

[Ovyron]

Bumping just to thank Alex again, this time for providing the 32bit versions of the engine.

Recently my 64bit OS became unusable, and so did Naum 64-bit, so it's a nice gesture to release these compiles that would work in other systems or chess variants

[A Distel]

The author, Alex Naumov, has asked me to provide the download link for you all.

Thanks Alex!

http://speedy.sh/2xFbC/naum46-x64.rar

Thanks... Banks!