The near future of computer chess

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

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
User avatar
Don
Posts: 5106
Joined: Tue Apr 29, 2008 2:27 pm

Re: The near future of computer chess

Post by Don » Fri Nov 18, 2011 9:52 pm

jdart wrote:If you think of chess programs as works of art, then what you are saying makes sense. But I don't think of them in this way. They are technical artifacts, and unlike works of art, their relative quality (or at least a major dimension of it, playing strength) is measurable.

--Jon
"Art" is an artificial term and the definition of it is abstract and changes from year to year.

What I am saying is that a well developed and strong chess program is much more than the sum of it's parts.

The human body is composed of a bunch of cells that individually are interesting, but doesn't by themselves amount to anything as spectacular as a human being. Same basic principles. It's all about how it's put together and chess programs are not designed like Mr Potato head - you don't just put together the parts and expect them to work together. Many ideas that work well in one program will not work at all in another.

A fast car is a good example. You can go out and purchase performance parts but to get the very most performance they have to be matched very precisely. You cannot just bolt any old piece on and expect the best results or even good results.

Don

User avatar
hgm
Posts: 23605
Joined: Fri Mar 10, 2006 9:06 am
Location: Amsterdam
Full name: H G Muller
Contact:

Re: The near future of computer chess

Post by hgm » Fri Nov 18, 2011 10:08 pm

jdart wrote:If you think of chess programs as works of art, then what you are saying makes sense. But I don't think of them in this way. They are technical artifacts, and unlike works of art, their relative quality (or at least a major dimension of it, playing strength) is measurable.

--Jon
You speak for yourself. Micro-Max is definitely a work of art! 8-)

User avatar
Don
Posts: 5106
Joined: Tue Apr 29, 2008 2:27 pm

Re: The near future of computer chess

Post by Don » Fri Nov 18, 2011 10:27 pm

hgm wrote:
jdart wrote:If you think of chess programs as works of art, then what you are saying makes sense. But I don't think of them in this way. They are technical artifacts, and unlike works of art, their relative quality (or at least a major dimension of it, playing strength) is measurable.

--Jon
You speak for yourself. Micro-Max is definitely a work of art! 8-)
Personally, I think of all chess programs as works of art because they engage the creative thinking process - in my opinion more than a painting or sculpture or cleverly worded essay.

But art is actually almost a derogatory term for a chess program because I feel that chess programs go beyond art. A classic work of art, such as a painting, is subject to interpretation and ambiguity. As they say, "beauty is in the eye of the beholder" and much art is judged differently by different people. All I have to do is say it's crap and my viewpoint is just as legitimate as someone else's. Such is art.

But a good chess program has to prove itself. You can fawn over it and experts can give their opinions, but it has the ability to prove it's own legitimacy. Of course we can criticize it's style - but at the end of the day that doesn't matter if it can win. We can also praise it's style but at the end of the day if it cannot win there is not that much to impress - or if does impress it's all subject to someones interpretation.

Of course a chess program can be impressive in other ways too. Franz was able to produce programs that ran with 256 bytes of RAM and very little ROM. To me such a program can also be viewed as a work of art.

Your programs are in the same category of course. Like any work of art it is finely constructed but more than art it is much more than just aesthetically pleasing, it is does this WHILE performing a function and it cannot just be put together any old way - it requires true craftsmanship.

User avatar
mhull
Posts: 12376
Joined: Wed Mar 08, 2006 8:02 pm
Location: Dallas, Texas
Full name: Matthew Hull

Re: The near future of computer chess

Post by mhull » Fri Nov 18, 2011 10:39 pm

Don wrote:Ed,

There are two types of people in any field. The majority only know how to work with things that are provided for them - they are not the people that actually make things happen. The world we live in today puts a huge amount of emphasis on giving the illusion of expertise to people who have no expertise and often computers are behind this. The concept is to try to help the person focus on doing something without having to be concerned about the low level details.

That is not a formula that ever leads to excellence. There will never be a paint by number kit that makes you a great artist for example. It only presents the illusion of competence. Providing a starting program that is at the very top can make everyone an expert, but it's hard to believe that you consider that desirable.

In fact I can hardly believe you even made the comment you did because it makes it seem like you don't have any concept of what actually goes into a chess program. What you are suggesting is that starting with a fully developed program and then passing it along to the pretenders to fine tune is how progress in this field will happen and that is completely ridiculous.

There is also the problem of incest. In the short time everyone has been really excited over the "clones", a new incredibly strong program coming out almost every week! Wow! It presented us with the illusion that every day we were making rapid progress with rich innovation, etc. We have seen the enthusiasts playing long matches between these programs and pretending they were all original works, but in the end it's just an illusion. Do you really have 20 unique programs or do you have a bunch of different versions of the same program?

The world has really been dumbed down a lot and people just accept what is spoon fed to them these days. Your way is the way of the world today and I'm disappointed in you for this. I just cannot believe you are becoming the advocate for this cookie cutter non-creative, non-innovative, pre-packaged kit style approach to computer chess.

Now you SAY you don't like it, but it's the way it is. But the surprising thing is that seem to think it's a superior approach, that you and I are outdated and that we need to get with the program so to speak. I think you are a very confused person.
+1

Your post is the gold standard in this thread. It hits the bullseye exactly. What a breath of fresh air!

The idea expressed at the top of the thread, suggesting we all give-in to and compromise with corruption makes me sick to my stomach.
Matthew Hull

K I Hyams
Posts: 3508
Joined: Fri Mar 31, 2006 9:21 pm

Re: The near future of computer chess

Post by K I Hyams » Fri Nov 18, 2011 10:46 pm

Rebel wrote: 1. You want to label the new guy Roberto Munter a terrorist for not willing to reinvent dozens of wheels and being honest about that ?
Nobody is going to criticise Roberto Munter for not being "willing to reinvent dozens of wheels and being honest about that." The crucial issue is what he does after he has legitimately modified a program.

He is likely to be criticised if he subsequently does any or all of the following:
#. Lies about having modified the program and claims it as completely original.
#. Sells it as an original piece of work.
#. Diverts income from sales away from the person whose program he modified and into his own pocket.
#. Enters the program in tournaments that are reserved for original programs.
#. Diverts prize money and/or trophies that were intended for programmers of original engines into his own pocket.
#. Violates the licences of those who open their code under restricted conditions.
#. Tricks people into testing his engine by claiming it was original.
#. Makes unsubstantiated claims and hypocritical claims that other people have copied his “original” engine.

If Rajlich has used Fruit code, he may have done all of those things. If Houdart has used Ippolit code, he may have done at least 2 of them. Whereas standing “on giants shoulders” is perfectly acceptable, it is the behaviour in which Rajlich and Houdart may have indulged subsequent to standing on giants shoulders that causes some of us problems. It is the absence of that behaviour that causes some of us to view Thomas Gaksch, the man who modified Fruit code to produce Toga, in a different light.

Carotino
Posts: 216
Joined: Fri Jun 11, 2010 8:40 am
Location: Italy
Contact:

Re: The near future of computer chess

Post by Carotino » Fri Nov 18, 2011 11:04 pm

The problem of Ippolit is that it's:

- without any license;
- without any limitation;
- without any real author (There are just fancy names, eg. Roberto Pescatore, etc.);

This is a Blessing or Curse? :twisted:

User avatar
kranium
Posts: 1824
Joined: Thu May 29, 2008 8:43 am

Re: The near future of computer chess

Post by kranium » Fri Nov 18, 2011 11:42 pm

Rebel wrote:
Rebel wrote: That you fight a lost battle like the music and video industry. You can't beat the clones with old rules not fit to stand the pressure anno 2011. That is better to do something now that you still have influence then to wait the problem to grow above your head and then realize you have lost with no influence at all.

It's a matter of vision. But if you want to live in the pre-internet past go ahead. It's mainly the old 80's and 90's generation that want to put their head in the sand.
bob wrote: [1] SO we can't defeat the terrorists outright. We should give up.

[2] We can't stop copying completely, so we should just open the floodgates and let everyone copy and enter.

[3] What a wonderful world you must live in. The music and video industries are hardly "losing" their battles. They are fighting on behalf of the artists making the videos or music. As they should.
1. You want to label the new guy Roberto Munter a terrorist for not willing to reinvent dozens of wheels and being honest about that ?

2. You know I don't want that. http://www.top-5000.nl/rule2.htm

3. You should take a look at youtube. You find everything. The music and movie industry are helpless. They have given up on youtube long time ago defeated by volume. These are the days of internet, it has come with new realities.

Bob, as 63 year old you certainly must know that to win a fight you can not win is to give up the lost fight and look for alternatives. I would not have said that 3-4 months ago but I foresee that before 2020 the situation will be drastically different as the new generation will take over, that's for sure. We are living now in a transition period, a fight between the old 80's and 90's pre-internet generation and the new 2005+ generation and it's better to give up now and find new solutions than being excluded from the innovatory process and laughed at for not recognizing the inevitable.

Also... one might wonder living in a world with strong source code all over the internet just a few mouse clicks away if it is reasonable to demand new chess programmers to write everything from scratch (because we had to!!) while at the same time it is allowed for the established programmers to freely take from Rybka's legacy of 400 elo points and never mention it. It's too bizarre for words and hypocritical. It's much better to have a transparent CC world with new rules, whatever those rules will be eventually.

Seriously, the time will come the new generation of chess programmers will give us oldies the middle finger for not using what's freely available blocking progress. We would like as those farmers refusing to use milk machines and go bankrupt for not recognizing the times.

Also put yourself in their shoes as an 20 year young university student anno 2011 fascinated by CC, would you start from scratch because a couple of old men want that ?
+1

thank-you Ed!

and now there's the incredible open-source/public domain IvanHoe (www.ippolit.wikispaces):

development ongoing for 2.5 years now
the program has grown from a basic skeleton engine to a full-featured analysis tool
they have added feature after feature...including many original technical innovations
Windows and Linux support
unique SMP code,
montecarlo analysis,
sophisticated hashing options,
compiling tracing,
Magic Bitboards,
ZugZwang detection,
Large page/slab memory,
Eval and Material explanation modes,
chess 960 support,
a complete bitbase solution that surpasses Nalimov EGTB,
Gaviota TB support,
JAVA GUI,
etc.
etc.

and for selfishly sharing this tremendous program and great CC progress...normally we must (should!) thank a group of Russia/Italian/Albanian programmers, and the Decembrists:
Yakov Petrovich Golyadkin, Igor Igorovich Igoronov, Roberto Pescatore Yusuf Ralf Weisskopf, Ivan Skavinsky Skavar plus Decembrists (all)...
but instead, (paradoxically), the program is arbitrarily blacklisted by the misguided abd uninformed amateur 'good old boy' testing groups: CCRL, CEGT, abd IPON!

Interestingly enough, many believe Rybka 4 benefited (took many things?) from the publication of strong IvanHoe
and according to Richard Vida (who has decompiled significant portions both programs):
it appears the establishment 'hero' Vas R. has liberally availed himself of this tremendous 'Ippolit' resource (like many others) with the release of Rybka 4:

"It is too soon to draw any conclusions but to my surprise R4/IPPO similarity is far greater than R3/IPPO. R4 was released quite a bit after IPPO, so one my draw (very wild) conclusions."
and:
"As for IPPOLIT... It seems that the authors took a special care not to copy anything literally. As I mentioned earlier, I am now digging into R3 internals, and the deeper I go the more I am convinced that IPPOLIT was written from scratch (although with heavy R3 influence)"

http://open-chess.org/viewtopic.php?f=5 ... 124#p13124

Clearly, this 'extreme right' entrenched CCC establishment has a singular interest: to preserve their embedded power/influence, i.e. 'status quo' at all cost,
hence the 'blacklisting' of Ippolit...

fortunately, we must be patient...we all know this abuse of influence is ultimately in vain, and ultimately change is inevitable.

Thanks for communicating that so aptly!

User avatar
kranium
Posts: 1824
Joined: Thu May 29, 2008 8:43 am

Re: The near future of computer chess

Post by kranium » Sat Nov 19, 2011 12:24 am

for anyone interested...

below are images of the 'ComradesGUI' (for general analysis)
and 'RobboExplorative' (for endgame analysis w/ Robbobases)

they are the 2 Java GUIs included with the just released IvanHoe 999946 Beta

Image
Image

you can download the KLO compiled IvanHoe 999946 package here:
http://chess.cygnitec.com/engine/ivanhoe/

absolute top quality (as usual!) work by King L.
Last edited by kranium on Sat Nov 19, 2011 12:29 am, edited 2 times in total.

User avatar
fern
Posts: 8755
Joined: Sun Feb 26, 2006 3:07 pm

Re: The near future of computer chess

Post by fern » Sat Nov 19, 2011 12:26 am

I understand your point fully, Don. Perhaps, instead, you do not understand mine. I am not defending this or that guy copying a program. I do see that there is a craft in all this, an organic entity where the sequence of codes, iterations , etc are the core of it, not just the subroutines.
I just do what more or less Ed S. did: to point out that in the general process of the progress of an activity this kind of behavior is common and part of it, with all the distortions, stealing, robbery, bad faith, etc.
So what I have done is to say, "look, this is the way of all flesh".
Some guys are stolen; some other are cheated; some progress are real hoaxes, some guys really make improvements of substance, some other do not, etc, etc.
In the middle of all that there are sheer robbery as the one you have mentioned..
Oh, this is long to explain for me and for you. Hard in this way, posting. Perhaps we should meet in a bar and talk about this..:-)

Fern

Christopher Conkie
Posts: 6068
Joined: Sat Apr 01, 2006 7:34 pm
Location: Scotland
Contact:

Re: The near future of computer chess

Post by Christopher Conkie » Sat Nov 19, 2011 3:26 am

fern wrote:Perhaps we should meet in a bar and talk about this..:-)

Fern
A Greek bar? Just a suggestion.....

Go look.

Post Reply