Experiment 11 - Similarity between the top engines back then and now.

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.
User avatar
Rebel
Posts: 4299
Joined: Thu Aug 18, 2011 10:04 am

Re: Experiment 11 - Similarity between the top engines back then and now.

Post by Rebel » Fri Mar 15, 2019 7:14 am

Hi Frank,

That's a cool looking picture.

How did you do the analysis work?

Analyze EPD with Arena?
Ignore list: HGM.

Frank Quisinsky
Posts: 4838
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: Experiment 11 - Similarity between the top engines back then and now.

Post by Frank Quisinsky » Thu Mar 21, 2019 8:32 pm

Hi Ed,

sorry for my late answer!
I am looking not often in chess fora for the moment.
I am suffering on the older chess computer fever, you understand ...
And you are one of the guilty persons!

:-)

Arena:
I don nothing with Arena since my work ended here around 2009 or so.

No, no ...
That's a grafic from the opening book project FEOBOS.
Under Shredder GUI 41.614 end positions analysed by 10 engines with 1 minutes per move, 6 cores on Intel i7 4.3Ghz. So it's nice to see if you put a template of the 41.614 postions. For FEOBOS opening book project (ended for around 14 months) we do some of such statistics under Excel. The idea is to find out the balanaced opening positions, 3 moves after ECO code ended for eng-eng testing. But with all the analyzes we can do a lot, so we try a bit under Excel:

FEOBOS:
http://www.amateurschach.de/main/_new-opening-book.htm

So 33.009 positions is the final result with Contempt = 3 (not more as 2 of 10 engines should give as eval 0,00). The topic for contempt is to find out the fast draw 3fold positions. But all this is history. Not important for your new idea.

I find your work interesting to that topic. Since a long time I lost my interest on eng-eng testing because most of the stronger engines today produced the same chess, same strengths, same weaknesses, allways the same with different names of engines. I missed a personality of the programs. Such a personality you gave Rebel (to bear the hollmarks, handwriting ... I can't find the right word in english language, sure you know what I mean) or John Stanback gave Zarkov / Wasp. That's computer chess engines for myself.

Yesterday I am reading CSS and saw the pic you hold the cup (1992).
Good, I like the pic ... again and again, you do a great job and with the years I am older and older I can more and more unterstand the criticism current computer chess devlopments got. Many people lost here interest ...

Be sure ...
I have fun all the time.
At the moment with the older chess computers.
Now 52 I have and in next month I will start my big tourney ... can be run 2 or 3 years, not important.

For you Ed ...
All the best from my site!!!
Hold your fun and ignore modern things, the time can be used better.
Hope I can write that to such a great person you are.

Friendly
Frank
I like computer chess!

Post Reply