Some TBS stat files missing

[vb4]

I downloaded each link that said ALL under the sets of 6 piece EGTB's and after I counted them I was missing about 38 files. 32 positions of the 3-3 format didnt exist in the master(ALL) but 4 of the 32 were found in the individual list. 6 files in the 4-2 format were missing in the master list but were all found in the individual list. Several examples are: kbnknp - kqnkpp - kqrkbp - kqnkqp to name a few. Can someone explain why all are not there and where can I go to get the complete list of the TBS stat files.

Thanks,

Les

[Kirill Kryukov]

Hi Les! If some TBS files are missing it means that simply I don't have them. If you will find them, please post here and I will put them to the web-site. Thanks!

[vb4]

Hi Les! If some TBS files are missing it means that simply I don't have them. If you will find them, please post here and I will put them to the web-site. Thanks!

Hi Kirill,

I will see what I can do but I hope they wont be to difficult to find. I will keep you posted as to what I find. Perhaps others out here can shed some light on this too. Should I post the positions that I know you are missing ???


Talke care,

Les

[Kirill Kryukov]

Thanks Les! Just post the TBS files if they should be added to the page.

[guyhaw]

Les,

See what you can lift from ftp://ftp.cis.uab.edu/pub/hyatt/TB/tbs/ and then provide a list of which .tbs files you are missing. I have them all.
I would expect this URL to provide all 3- to 5-man, 3-3, 4-2 and 4-2p .tbs files + most 3-3p ones [but obviously not the 'last 16' 3-3p .tbs files and maybe some others].
g

[vb4]

Les,

See what you can lift from ftp://ftp.cis.uab.edu/pub/hyatt/TB/tbs/ and then provide a list of which .tbs files you are missing. I have them all.
I would expect this URL to provide all 3- to 5-man, 3-3, 4-2 and 4-2p .tbs files + most 3-3p ones [but obviously not the 'last 16' 3-3p .tbs files and maybe some others].
g

Hi Guy,

Thanks for your willingness to help. Here is the list of TBS stat files that are missing and I can't find anyone who has them. They must exist from whoever generated them but I dont know who that is.

krnkbp
krpkbb
kbbkpp
kbnknp
knnkpp
kqpkbb
kqpkrr
kqpkqp
kqnkpp
kqqkrp
kqqkpp
kqrkqp
kqrkbp
krbknp
krbkbp
krrkpp
krrknp
kqpkbn
kqrknp
kqrkpp
kqbkrp
kqbkbp
kqbknp
kqbkpp
kqnkqp
kqnkrp
kqnkbp
kqnknp

Once again Guy thanks for your help in this matter. All the positions are from the 3-3 set. I dont need the actual tablebase files but am interested in the tbs files.

Thanks,

Les

[guyhaw]

This zipfile should include all 3-3p Nalimov DTM .tbs statistics

[vb4]

This zipfile should include all 3-3p Nalimov DTM .tbs statistics

Hi Guy,

Thank you so much for those files!!! You are very resourceful. Now all we need is to stay on top of Josh in hope that he is able to generate the remaining 5-1 egtb's.

Thanks again Guy,

Les

[guyhaw]

Les,
I do not include myself in your "we": 5-1(p) EGTs hold little interest unless one is creating 5-2(p) EGTs.
In fact, Marc B has generated KxxxxKQ where x = B or N, so he has generated KxxxxK under the same constraints, but to the DTZ metric rather than the DTM metric.
DTZ == "Depth to (move-count) Zeroing (move)", i.e. to P-push, conversion of force and/or mate. This is a better metric in the sense that it produces increasingly more compressible EGTs.

What will be needed in the end is a version of Nalimov's access-code, incorporated in chess engines, that reflects Eugene's latest thinking on the 8-bit/16-bit-choice for KQPK(B/R)P. Perhaps only CRAFTY is up to date at the moment.
g