CCRL Discussion Board

More compact format: 8/3b2KP/8/2R5/8/5R2/pk6/7r w - - acd 29/81; acn 62258191501; acs 432000; bm Ra5; c0 "Árpád Rusz - Sakkvilág, 2010"; ce 1355; pv Ra5 a1=N Rf2+ Kb1 Rf1+ Rxf1 Rxa1+ Kxa1 h8=Q Be6 Kh7+ Ka2 Qa8+ Kb3 Qb8+ Kc4 Qc7+ Kd5 Qa5+ Ke4 Qb4+ Kf3 Qf8+ Ke2 Qe8 Kd3 Qxe6 Rf4 Kg6 Kd4 Kg5 R...

FEN: 8/3b2KP/8/2R5/8/5R2/pk6/7r w - - 0 1 A certain magician says: 4/14 00:00 3.508 1.169.000 +1.10 Rf3f2+ Kb2b3 Rc5a5 Rh1xh7+ Kg7xh7 4/14 00:00 5.862 1.465.000 +1.40 Rc5a5 Rh1xh7+ Kg7xh7 Bd7e6 Rf3f2+ Kb2c3 Ra5a3+ Kc3b4 5/15 00:00 6.809 1.702.000 +1.40 Rc5a5 Rh1xh7+ Kg7xh7 Bd7e6 Rf3f2+ Kb2c3 Ra5a3+...

Thanks.
Those were my general impressions.

What is the magic incantation to download it. The download does not work for me.

I have a notion:
When you rewrite the 5 man and generalize it for larger counts, how about making it totally general so that 7,8,9 etc. man files can also be generated.
It may sound silly now, but 5 years from now we'll have petabyte hard drives for $100.

I have some positions that could benefit greatly from 7 man EGTB (notably from MES1234). Are there any chess engines that are functional with 7 man EGTB files that are publicly available? I guess not, but it couldn't hurt to ask.

To use the Nalimov tablebase files, you have to get permission first. It is also problematic as far as licence (e.g. it cannot be used in GPL projects, for example). Miguel Ballicora's format is the wave of the future. It will be released with a licence that everyone can use. Every other format is a...

He also has some big Linux arrays of some kind also. Les' message spells out the details.

I don't know if it is of any value, but Les Fernandez has a vendor who will lend time on a big array of GPU cards for computation of EGTBs. I have no idea how difficult it would be to port EBTB systems to GPU cards (for instance, they can't do recursion if I recall correctly) but it might be worth t...

It's DTM
I am not terribly familiar with the internals as yet.
I suppose that some patience will be needed until the big unveiling.

I have been discussing a compression scheme with him.

If he decides it is a good idea, we should get about 10:1 compression so the 5 men files will consume 3.8 GB instead.

Gaviota EGTB will soon be released with Berkeley license. It means that any chess programmer will be able to have Tablebase access without the very long waits to contact Eugene and also they can be used for any purpose. Currently, it only generates up to 5 man files, but there are plans to perform t...

Not all are directly pertinant, and some are even about endgame calculations for other games besides chess. But perhaps something here will be interesting. http://cap.connx.com/chess-papers/A%20memory%20efficient%20retrograde%20algorithm%20and%20its%20application%20to%20chinese%20chess%20endgames.ra...

Source code & MS VC++ workspaces:
http://cap.connx.com/chess-engines/new- ... wtbgen.zip

bzip2 compressed 64 bit binary:
http://cap.connx.com/chess-engines/new- ... EN.EXE.bz2

The only ones I am aware about are HGM's and Eugene Nalimov's.
Are there any others?

Post subject: EGTB compression (H. G. Muller tablebase...) Posted: Fri Jan 25, 2008 7:36 am -------------------------------------------------------------------------------- Given the following uncompressed H. G. Muller tablebase file: 01/24/2008 11:25 PM 268,439,552 KBB_KN From here: http://www.quic...