Six piece EGTB

Discussion of chess software programming and technical issues.

Moderator: Ras

vb4
Posts: 165
Joined: Sat Mar 11, 2006 5:45 am
Location: NY

Six piece EGTB

Post by vb4 »

There has not been much conversation on this subject and I was curious if any of you out here knows of anyone who has completed the set of 5-1 EGTB's. My interest is in the stat files and not the actual tablebase files themselves. I understand there are 32 of these tables and I think someone had mentioned they generated 16 of the 32 remaining ones which means only 16 remaining. Does anyone here know of anyone who has completed the remaining 16 tables or has a generator that can generate them without code modification? I am studying the 3-4-5-6 piece EGTB's and need to have completed sets to finish my work. Any help or guidance would be deeply appreciated.

Les
ernest
Posts: 2053
Joined: Wed Mar 08, 2006 8:30 pm

Re: Six piece EGTB

Post by ernest »

Why double-posting?

See my answer in
http://www.talkchess.com/forum/viewtopi ... 587#287587
Dann Corbit
Posts: 12838
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Six piece EGTB

Post by Dann Corbit »

ernest wrote:Why double-posting?
It was my suggestion. He was not sure if it was a good idea.
Some of the programmer types frequent this forum more often than the CCC one. If it was a newsgroup, he could have put both groups in a header, but there is no option like that for CCC
vb4
Posts: 165
Joined: Sat Mar 11, 2006 5:45 am
Location: NY

Re: Six piece EGTB

Post by vb4 »

Thanks Dann!
vb4
Posts: 165
Joined: Sat Mar 11, 2006 5:45 am
Location: NY

Re: Six piece EGTB

Post by vb4 »

Hi Ernest,

I guess what I am in need of are the set of 5-1 stat files not the actual tables. It is not important to me that current chess engines cant utilize the tables since that is not my interest although that would be nice. Most everyone refers to the 6 piece EGTB as being complete when it doesnt account for the 5-1 set. Do you know of anyone that has the 5-1 completed set of stat files?

Thanks,

Les
ernest
Posts: 2053
Joined: Wed Mar 08, 2006 8:30 pm

Re: Six piece EGTB

Post by ernest »

vb4 wrote:Do you know of anyone that has the 5-1 completed set of stat files?
Complete... I don't know.
But I think they are speaking of those stats in the references I gave.

OK I looked there:
http://kirill-kryukov.com/chess/discuss ... in+Kreuzer

See the tbs51.tar (stats files for 5-1 pawnless Nalimov-like tablebases) by John Kominek (jkominek)

and... here you are:
Attachments
6men-pawnfull-stats.from_tbgen2.tar.gz
4 5-1 pawnfull stats files: kxxxpk.
(1.48 KiB) Downloaded 29 times
6men-pawnless-stats.from_tbgen2.tar.gz
35 5-1 pawnless stats files.
(7.93 KiB) Downloaded 23 times

Maybe even Martin Kreuzer finally published stats from his 5-1 FEG tables

After downloading those stats you are certainly welcome to to register and then post further questions in that remarkable (unique) CCRL Discussion Board/ Endgame Tablebases Forum.

But vb4/Les, what the hell is that, I just see that you have been posting there all the time, why are you bothering people in this CCC Forum??? :?: :?: :?: :evil: :evil: :evil:

...or has Alzheimer caught up with you???
diep
Posts: 1822
Joined: Thu Mar 09, 2006 11:54 pm
Location: The Netherlands

Re: Six piece EGTB

Post by diep »

not so complicated to generate them,
where do you need them for?
vb4
Posts: 165
Joined: Sat Mar 11, 2006 5:45 am
Location: NY

Re: Six piece EGTB

Post by vb4 »

Hi Vincent,

It has been a long time since we have chatted! I am studying relationships between the 2-3-4-5-6 piece egtb's and have all the data except for the dozen or so of the 5-1 stat files. What I have been using till now are the stat files that Eugene and others have generated over the past several years and just want to complete the 6 piece stat files so that I can work with a pure set of data. Over the years I have from time to time posted if anyone plans on completing the set (stat files only) and the reponse has been there is no purpose since there is no reason to since todays engines can solve any 5-1 end game. I know that statement has alot of fact but no one has paid attention to the fact that until I can have all completed stat files I really cant move forward in my studies. The studies may not show anythnig of interest but I believe there are some interesting things within them.

My thoughts are that the 5-1 should converge fairly quickly but I can say for sure.

Thanks Vincent for atleast taking time to show some interest in my post.

Les