Protector 1.2.9 executables with egtb access

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

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Volker Pittlik
Posts: 597
Joined: Wed Mar 08, 2006 8:10 pm
Location: Murten / Morat, Switzerland

Protector 1.2.9 executables with egtb access

Post by Volker Pittlik » Thu Oct 22, 2009 11:11 am

I've taken a look at the latest Protector version (091021). No doubt there are influences from other engines. That's not my point. Also that it does not compile with gcc. That's a minor issue.

According to the Readme
... the permission to use and distribute Eugene Nalimov's and Andrew Kadatch's code for accessing Nalimov tablebases along with the sources of Protector is pending...
There are instructions how to compile the program without tablebase support. There are also instructions for those who
...have the permission to use or include their ( Nalimov's and Kadatch's) code...
how to make a version which is enabled to access egtbs.

These instructions include to copy files from Crafty (egtb.cpp, tbdecode.h and lock.h).

My first question is: is it legal to copy files from Crafty and use them in other programs? I presume the answer is no.

Furthermore there are some executables in the package which seem to use the tablebases:

Code: Select all

256208455.056 POLYGLOT FEN 8/4R3/4K1P1/7k/8/8/8/5r2 w - - 0 97...
1256208455.088 Engine->Adapter: info depth 37 seldepth 37 time 1 nodes 19144 pv g6g7 f1g1 score mate 19 tbhits 1300
My second question is: is it legal to use that executables? Is it even legal to spread them without the permissions?

Maybe only the Readme isn't up-to-date. Can someone please enlighten me?

vp

User avatar
Graham Banks
Posts: 33138
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

The author's post from the Russian forum

Post by Graham Banks » Fri Oct 23, 2009 8:49 pm

The Protector author posted in the Russian forum:
Since the first release of Protector there has been some debate if Protector should be regarded as a Fruit clone. Of all comments written on this topic I would like to pick one that summarizes the complexity of the question pretty well. It can be found at

http://talkchess.com/forum/viewtopic.ph ... 95&t=29581

I didn't write Protector from scratch in order to have an "original" engine and I don't plan to modify Protector towards that goal. This would be a futile task anyway. The answer to the question if Protector is a Fruit clone depends on the personal notion of 'clone' -- my impression is there will always be some people who are not satisfied by the 'distance' between the Protector and the Fruit sources. Currently I don't plan to enter Protector in a serious tournament and thus it isn't an important topic for me.

The Fruit and Toga sources are freely available and cannot be unpublished anymore. Everyone who wants to judge about cloning needs to draw his own line of separation but if it is drawn with too much distance from Fruit this will restrict other programmers severly. I don't believe that this was Fabien Letouzey's intention when he published Fruit -- well, my opinion."

As already mentioned above I have modified several things in Protector that orginally have been analogous to Fruit/Toga. I will do so in the future but my only goal was, is and will be to create a strong chess program that is as competitive as possible -- version 1.2.8 scores about 80% vs. Fruit 2.1 on Linux (the latest open source Fruit version); pretty good for a 'clone', isn't it?. To a certain extend I can understand people who didn't find anything 'new' or 'original' in the Protector source code -- I was also disappointed when I first saw the Fruit source code. There wasn't much new in it but nonetheless Fruit played much stronger than many other programs who seemed to comprise almost the same ingredients.
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz

User avatar
Graham Banks
Posts: 33138
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Correction - it's from the readme file

Post by Graham Banks » Fri Oct 23, 2009 9:17 pm

The author's comments are actually from the readme file.
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz

User avatar
slobo
Posts: 2331
Joined: Mon Apr 09, 2007 3:36 pm

Re: The author's post from the Russian forum

Post by slobo » Fri Oct 23, 2009 9:18 pm

Graham Banks wrote:The Protector author posted in the Russian forum:
Since the first release of Protector there has been some debate if Protector should be regarded as a Fruit clone. Of all comments written on this topic I would like to pick one that summarizes the complexity of the question pretty well. It can be found at

http://talkchess.com/forum/viewtopic.ph ... 95&t=29581

I didn't write Protector from scratch in order to have an "original" engine and I don't plan to modify Protector towards that goal. This would be a futile task anyway. The answer to the question if Protector is a Fruit clone depends on the personal notion of 'clone' -- my impression is there will always be some people who are not satisfied by the 'distance' between the Protector and the Fruit sources. Currently I don't plan to enter Protector in a serious tournament and thus it isn't an important topic for me.

The Fruit and Toga sources are freely available and cannot be unpublished anymore. Everyone who wants to judge about cloning needs to draw his own line of separation but if it is drawn with too much distance from Fruit this will restrict other programmers severly. I don't believe that this was Fabien Letouzey's intention when he published Fruit -- well, my opinion."

As already mentioned above I have modified several things in Protector that orginally have been analogous to Fruit/Toga. I will do so in the future but my only goal was, is and will be to create a strong chess program that is as competitive as possible -- version 1.2.8 scores about 80% vs. Fruit 2.1 on Linux (the latest open source Fruit version); pretty good for a 'clone', isn't it?. To a certain extend I can understand people who didn't find anything 'new' or 'original' in the Protector source code -- I was also disappointed when I first saw the Fruit source code. There wasn't much new in it but nonetheless Fruit played much stronger than many other programs who seemed to comprise almost the same ingredients.
And now, Graham, your home-work:

Read the Protector author post, once more, and imagine it was posted by the Ippolit´s author.
"Well, I´m just a soul whose intentions are good,
Oh Lord, please don´t let me be misunderstood."

User avatar
Graham Banks
Posts: 33138
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: The author's post from the Russian forum

Post by Graham Banks » Fri Oct 23, 2009 9:20 pm

slobo wrote: And now, Graham, your home-work:

Read the Protector author post, once more, and imagine it was posted by the Ippolit´s author.
Fruit is open source, Rybka is not.
Slight difference don't you think?
You can pick the crooks in this forum from a mile away. :wink:
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz

User avatar
slobo
Posts: 2331
Joined: Mon Apr 09, 2007 3:36 pm

Re: The author's post from the Russian forum

Post by slobo » Fri Oct 23, 2009 9:26 pm

Graham Banks wrote:
slobo wrote: And now, Graham, your home-work:

Read the Protector author post, once more, and imagine it was posted by the Ippolit´s author.
Fruit is open source, Rybka is not.
Slight difference don't you think?
You can pick the crooks in this forum from a mile away. :wink:
Ippolite is more like Fruit than like Rybka. If you or anyone else don´t agree, prove your points.

You think you are very smart?
"Well, I´m just a soul whose intentions are good,
Oh Lord, please don´t let me be misunderstood."

gladius
Posts: 538
Joined: Tue Dec 12, 2006 9:10 am

Re: The author's post from the Russian forum

Post by gladius » Fri Oct 23, 2009 10:09 pm

slobo wrote:Ippolite is more like Fruit than like Rybka. If you or anyone else don´t agree, prove your points.
This is unbelievable. Clearly you have not looked at fruit and ippolit's source. From a programming point of view, they could not be more different.

What is your proof that it is similar to Fruit (besides the somewhat close hashtable implementation)?

User avatar
slobo
Posts: 2331
Joined: Mon Apr 09, 2007 3:36 pm

Re: The author's post from the Russian forum

Post by slobo » Fri Oct 23, 2009 10:35 pm

gladius wrote:
slobo wrote:Ippolite is more like Fruit than like Rybka. If you or anyone else don´t agree, prove your points.
This is unbelievable. Clearly you have not looked at fruit and ippolit's source. From a programming point of view, they could not be more different.

What is your proof that it is similar to Fruit (besides the somewhat close hashtable implementation)?
It´s strength. All super-strong chess engines clones are based on Fruit: Toga, Rybka and Ippolit. The same proof the accusers have against Ippolit.
"Well, I´m just a soul whose intentions are good,
Oh Lord, please don´t let me be misunderstood."

Karlo Bala
Posts: 301
Joined: Wed Mar 22, 2006 9:17 am
Location: Novi Sad, Serbia

Re: The author's post from the Russian forum

Post by Karlo Bala » Fri Oct 23, 2009 11:31 pm

slobo wrote:
gladius wrote:
slobo wrote:Ippolite is more like Fruit than like Rybka. If you or anyone else don´t agree, prove your points.
This is unbelievable. Clearly you have not looked at fruit and ippolit's source. From a programming point of view, they could not be more different.

What is your proof that it is similar to Fruit (besides the somewhat close hashtable implementation)?
It´s strength. All super-strong chess engines clones are based on Fruit: Toga, Rybka and Ippolit. The same proof the accusers have against Ippolit.
There is no doubt that new era in computer chess started with Fruit and Glaurung
Best Regards,
Karlo Balla Jr.

Volker Pittlik
Posts: 597
Joined: Wed Mar 08, 2006 8:10 pm
Location: Murten / Morat, Switzerland

Re: The author's post from the Russian forum

Post by Volker Pittlik » Sat Oct 24, 2009 2:58 am

Thanks for your replies. But it was not my intention to provoke a new hassle.

The author himself "admits" he uses other open source programs. The problem I see is if it is legal to use and spread a program which is enabled to use the Nalimov egtbs without his permission. And the executables in the package obviously do that.

vp

Post Reply