Generating all 6-men EGDB (Chessmaster).

[Juan P. Naar]

Several years later, the Chessmaster GUI is the one I'm still using the most. Some months ago I decided to embark on the journey of generating all the 6-men EGDB using two somewhat old PCs, a six core Westmere-EP Xeon and a quad core Haswell i7 (I'm about 2/3 through!)

My goal is to complete all 6-men EGDB and then generate some important 7-man pieces (has someone generated a 7 man EGDB set with FEG before?). However I need to know if my 6-men list of possible pieces combinations has the right number:

5 vs 1: 70 sets (from KQQQQK to KPPPPK)
3 vs 3: 225 sets (from KQQKQQ to KPPKPP)
4 vs 2: 175 sets (from KQQQKQ to KPPPKP)
2 vs 4: 175 sets (from KQKQQQ to KPKPPP)

Total possible sets: 645.

Am I missing any other combination to complete all the 6 men sets?

Also, I could upload them if someone else needs them, if someone is willing to host them, that is.

[AdminX]

Several years later, the Chessmaster GUI is the one I'm still using the most. Some months ago I decided to embark on the journey of generating all the 6-men EGDB using two somewhat old PCs, a six core Westmere-EP Xeon and a quad core Haswell i7 (I'm about 2/3 through!)

My goal is to complete all 6-men EGDB and then generate some important 7-man pieces (has someone generated a 7 man EGDB set with FEG before?). However I need to know if my 6-men list of possible pieces combinations has the right number:

5 vs 1: 70 sets (from KQQQQK to KPPPPK)
3 vs 3: 225 sets (from KQQKQQ to KPPKPP)
4 vs 2: 175 sets (from KQQQKQ to KPPPKP)
2 vs 4: 175 sets (from KQKQQQ to KPKPPP)

Total possible sets: 645.

Am I missing any other combination to complete all the 6 men sets?

Also, I could upload them if someone else needs them, if someone is willing to host them, that is.

I had the tool used to create them once apond a time, however when I tried to locate it again I was not able to find it. The main problem I had with them was I was not able to use them outside of the Chessmaster GUI for "TheKing" to work with. Sorry if my response is a little off topic.

[JVMerlino]

Several years later, the Chessmaster GUI is the one I'm still using the most. Some months ago I decided to embark on the journey of generating all the 6-men EGDB using two somewhat old PCs, a six core Westmere-EP Xeon and a quad core Haswell i7 (I'm about 2/3 through!)

My goal is to complete all 6-men EGDB and then generate some important 7-man pieces (has someone generated a 7 man EGDB set with FEG before?). However I need to know if my 6-men list of possible pieces combinations has the right number:

5 vs 1: 70 sets (from KQQQQK to KPPPPK)
3 vs 3: 225 sets (from KQQKQQ to KPPKPP)
4 vs 2: 175 sets (from KQQQKQ to KPPPKP)
2 vs 4: 175 sets (from KQKQQQ to KPKPPP)

Total possible sets: 645.

Am I missing any other combination to complete all the 6 men sets?

Also, I could upload them if someone else needs them, if someone is willing to host them, that is.

Did TheKing even use EGDBs in the search? Unless Johan modified TheKing after CM9000 (the last version I worked on) to do so, you won't get any better play from the engine (and negligible improvement overall). In CM9000, it was the GUI that checked the EGDBs for a move only when there were 6 or fewer men on the board; TheKing did not use the EGDBs in the search.

And regarding generating 7-men, I don't think FEN supported it at all.

jm

[Ajedrecista]

Hello Juan Pablo:

Several years later, the Chessmaster GUI is the one I'm still using the most. Some months ago I decided to embark on the journey of generating all the 6-men EGDB using two somewhat old PCs, a six core Westmere-EP Xeon and a quad core Haswell i7 (I'm about 2/3 through!)

My goal is to complete all 6-men EGDB and then generate some important 7-man pieces (has someone generated a 7 man EGDB set with FEG before?). However I need to know if my 6-men list of possible pieces combinations has the right number:

5 vs 1: 70 sets (from KQQQQK to KPPPPK)
3 vs 3: 225 sets (from KQQKQQ to KPPKPP)
4 vs 2: 175 sets (from KQQQKQ to KPPPKP)
2 vs 4: 175 sets (from KQKQQQ to KPKPPP)

Total possible sets: 645.

Am I missing any other combination to complete all the 6 men sets?

Also, I could upload them if someone else needs them, if someone is willing to host them, that is.

According to this site:

Number of endgames

The number of endgames with N = 6 pieces is (1/2)*(N+7 over N-2) + (1/2)*((N+6)/2 over (N-2)/2) = (1/2)*(13 over 4) + (1/2)*(6 over 2) = 365. (a over b) is the binomial coefficient. A list with the number of endgames knowing the number of pieces can be found here: 1 for 2 pieces, 5 for 3 pieces, 30 for 4 pieces, 110 for 5 pieces, 365 for 6 pieces, 1001 for 7 pieces and so on.

I guess that 1/2 factor is for not counting white and black twice. For example, with 3 pieces and according to the same site: (1/2)*(10 over 1) = 5, which should be KQk, KRk, KBk, KNk and KPk (uppercase for white, lowercase for black). Of course there are Kkq, Kkr, Kkb, Kkn and Kkp, so you can get rid of the factor 1/2.

However, you follow a different counting system taking colours into account, so 365 can not be directly applied. I do not understand why you distinguish '4 vs 2' and '2 vs 4' in your post while you do not do the same for '5 vs 1' and '1 vs 5'.

------------

There is a number of sites that offer 6-piece tablebases for download if you want to compare set by set. Just mention few of them:

http://tablebase.sesse.net/

http://kirill-kryukov.com/chess/tablebases-online/

I already have got a summary of statistics of endgames. In fact they are 365 without taking into account colours. I hope no typos:

Code: Select all

5 vs 1 (70 sets):

KQQQQK
KQQQRK
KQQQBK
KQQQNK
KQQRRK
KQQRBK
KQQRNK
KQQBBK
KQQBNK
KQQNNK
KQRRRK
KQRRBK
KQRRNK
KQRBBK
KQRBNK
KQRNNK
KQBBBK
KQBBNK
KQBNNK
KQNNNK
KRRRRK
KRRRBK
KRRRNK
KRRBBK
KRRBNK
KRRNNK
KRBBBK
KRBBNK
KRBNNK
KRNNNK
KBBBBK
KBBBNK
KBBNNK
KBNNNK
KNNNNK
KQQQPK
KQQRPK
KQQBPK
KQQNPK
KQQPPK
KQRRPK
KQRBPK
KQRNPK
KQRPPK
KQBBPK
KQBNPK
KQBPPK
KQNNPK
KQNPPK
KQPPPK
KRRRPK
KRRBPK
KRRNPK
KRRPPK
KRBBPK
KRBNPK
KRBPPK
KRNNPK
KRNPPK
KRPPPK
KBBBPK
KBBNPK
KBBPPK
KBNNPK
KBNPPK
KBPPPK
KNNNPK
KNNPPK
KNPPPK
KPPPPK

Code: Select all

3 vs 3 (120 sets):

KQQKQQ
KQQKQR
KQQKQB
KQQKQN
KQQKRR
KQQKRB
KQQKRN
KQQKBB
KQQKBN
KQQKNN
KQRKQR
KQRKQB
KQRKQN
KQRKRR
KQRKRB
KQRKRN
KQRKBB
KQRKBN
KQRKNN
KQBKQB
KQBKQN
KQBKRR
KQBKRB
KQBKRN
KQBKBB
KQBKBN
KQBKNN
KQNKQN
KQNKRR
KQNKRB
KQNKRN
KQNKBB
KQNKBN
KQNKNN
KRRKRR
KRRKRB
KRRKRN
KRRKBB
KRRKBN
KRRKNN
KRBKRB
KRBKRN
KRBKBB
KRBKBN
KRBKNN
KRNKRN
KRNKBB
KRNKBN
KRNKNN
KBBKBB
KBBKBN
KBBKNN
KBNKBN
KBNKNN
KNNKNN
KQQKQP
KQQKRP
KQQKBP
KQQKNP
KQQKPP
KQRKQP
KQRKRP
KQRKBP
KQRKNP
KQRKPP
KQBKQP
KQBKRP
KQBKBP
KQBKNP
KQBKPP
KQNKQP
KQNKRP
KQNKBP
KQNKNP
KQNKPP
KQPKQP
KQPKRR
KQPKRB
KQPKRN
KQPKRP
KQPKBB
KQPKBN
KQPKBP
KQPKNN
KQPKNP
KQPKPP
KRRKRP
KRRKBP
KRRKNP
KRRKPP
KRBKRP
KRBKBP
KRBKNP
KRBKPP
KRNKRP
KRNKBP
KRNKNP
KRNKPP
KRPKRP
KRPKBB
KRPKBN
KRPKBP
KRPKNN
KRPKNP
KRPKPP
KBBKBP
KBBKNP
KBBKPP
KBNKBP
KBNKNP
KBNKPP
KBPKBP
KBPKNN
KBPKNP
KBPKPP
KNNKNP
KNNKPP
KNPKNP
KNPKPP
KPPKPP

Code: Select all

4 vs 2 (175 sets):

KQQQKQ
KQQQKR
KQQQKB
KQQQKN
KQQRKQ
KQQRKR
KQQRKB
KQQRKN
KQQBKQ
KQQBKR
KQQBKB
KQQBKN
KQQNKQ
KQQNKR
KQQNKB
KQQNKN
KQRRKQ
KQRRKR
KQRRKB
KQRRKN
KQRBKQ
KQRBKR
KQRBKB
KQRBKN
KQRNKQ
KQRNKR
KQRNKB
KQRNKN
KQBBKQ
KQBBKR
KQBBKB
KQBBKN
KQBNKQ
KQBNKR
KQBNKB
KQBNKN
KQNNKQ
KQNNKR
KQNNKB
KQNNKN
KRRRKQ
KRRRKR
KRRRKB
KRRRKN
KRRBKQ
KRRBKR
KRRBKB
KRRBKN
KRRNKQ
KRRNKR
KRRNKB
KRRNKN
KRBBKQ
KRBBKR
KRBBKB
KRBBKN
KRBNKQ
KRBNKR
KRBNKB
KRBNKN
KRNNKQ
KRNNKR
KRNNKB
KRNNKN
KBBBKQ
KBBBKR
KBBBKB
KBBBKN
KBBNKQ
KBBNKR
KBBNKB
KBBNKN
KBNNKQ
KBNNKR
KBNNKB
KBNNKN
KNNNKQ
KNNNKR
KNNNKB
KNNNKN
KQQQKP
KQQRKP
KQQBKP
KQQNKP
KQQPKQ
KQQPKR
KQQPKB
KQQPKN
KQQPKP
KQRRKP
KQRBKP
KQRNKP
KQRPKQ
KQRPKR
KQRPKB
KQRPKN
KQRPKP
KQBBKP
KQBNKP
KQBPKQ
KQBPKR
KQBPKB
KQBPKN
KQBPKP
KQNNKP
KQNPKQ
KQNPKR
KQNPKB
KQNPKN
KQNPKP
KQPPKQ
KQPPKR
KQPPKB
KQPPKN
KQPPKP
KRRRKP
KRRBKP
KRRNKP
KRRPKQ
KRRPKR
KRRPKB
KRRPKN
KRRPKP
KRBBKP
KRBNKP
KRBPKQ
KRBPKR
KRBPKB
KRBPKN
KRBPKP
KRNNKP
KRNPKQ
KRNPKR
KRNPKB
KRNPKN
KRNPKP
KRPPKQ
KRPPKR
KRPPKB
KRPPKN
KRPPKP
KBBBKP
KBBNKP
KBBPKQ
KBBPKR
KBBPKB
KBBPKN
KBBPKP
KBNNKP
KBNPKQ
KBNPKR
KBNPKB
KBNPKN
KBNPKP
KBPPKQ
KBPPKR
KBPPKB
KBPPKN
KBPPKP
KNNNKP
KNNPKQ
KNNPKR
KNNPKB
KNNPKN
KNNPKP
KNPPKQ
KNPPKR
KNPPKB
KNPPKN
KNPPKP
KPPPKQ
KPPPKR
KPPPKB
KPPPKN
KPPPKP

Apart from the '5 vs 1' and '1 vs 5' issue, the rest looks fine for me:

a) 4 vs 2 has 175 sets and there are other 175 for 2 vs 4 using your notation.

b) Regarding 3 vs 3, just doubling 120 to get 240 is misleading because duplicates of both white and black sides must be removed. There are exactly 15 of them, so 120*2 - 15 = 225, like your number. Duplicates: the same in characters 1-3 than in characters 4-6 ==> KQQKQQ, KQRKQR, KQBKQB, KQNKQN, KQPKQP, KRRKRR, KRBKRB, KRNKRN, KRPKRP, KBBKBB, KBNKBN, KBPKBP, KNNKNN, KNPKNP and KPPKPP.

All in all, I get 365 without taking into account colours and 715 (your 645 plus another 70 from the case '1 vs 5') taking into account them.

Regards from Spain.

Ajedrecista.

[JVMerlino]

The FEG format actually uses many more files, because it has one set of files for each material combination, which comprises one set for each winning side, each of which has a pair for side to move, each of which has another pair for index/data. So, in the 5-man set there are 920 (so 230 "sets") files because, for example, there are sets for both KQPKQ and KQKQP.

KQKQP__B.J00
KQKQP__B.JIX
KQKQP__W.J00
KQKQP__W.JIX
KQPKQ__B.J00
KQPKQ__B.JIX
KQPKQ__W.J00
KQPKQ__W.JIX

jm

[Juan P. Naar]

Did TheKing even use EGDBs in the search? Unless Johan modified TheKing after CM9000 (the last version I worked on) to do so, you won't get any better play from the engine (and negligible improvement overall). In CM9000, it was the GUI that checked the EGDBs for a move only when there were 6 or fewer men on the board; TheKing did not use the EGDBs in the search.

And regarding generating 7-men, I don't think FEN supported it at all.

jm

According to the FEG readme: "To choose the best move Chessmaster searches 1 ply ahead, or sometimes
2 ply if only btm data is installed." "For 5 pieces we can save some space by omitting either White to move
(wtm) or Black to move (btm). The search for best move must then be
extended from 1 ply to 2 ply. This will cause much more random access,
but is still doable in about 1 second. It will reduce storage to less
than half (using the smallest files of either wtm or btm)."

I'm aware I won't get much stronger play from the engine, this is more like a fun project. And regarding generating 7-men, I'll just try and see what happens

Hello Juan Pablo:

According to this site:

Number of endgames

The number of endgames with N = 6 pieces is (1/2)*(N+7 over N-2) + (1/2)*((N+6)/2 over (N-2)/2) = (1/2)*(13 over 4) + (1/2)*(6 over 2) = 365. (a over b) is the binomial coefficient. A list with the number of endgames knowing the number of pieces can be found here: 1 for 2 pieces, 5 for 3 pieces, 30 for 4 pieces, 110 for 5 pieces, 365 for 6 pieces, 1001 for 7 pieces and so on.

I guess that 1/2 factor is for not counting white and black twice. For example, with 3 pieces and according to the same site: (1/2)*(10 over 1) = 5, which should be KQk, KRk, KBk, KNk and KPk (uppercase for white, lowercase for black). Of course there are Kkq, Kkr, Kkb, Kkn and Kkp, so you can get rid of the factor 1/2.

However, you follow a different counting system taking colours into account, so 365 can not be directly applied. I do not understand why you distinguish '4 vs 2' and '2 vs 4' in your post while you do not do the same for '5 vs 1' and '1 vs 5'.

------------

There is a number of sites that offer 6-piece tablebases for download if you want to compare set by set. Just mention few of them:

http://tablebase.sesse.net/

http://kirill-kryukov.com/chess/tablebases-online/

I already have got a summary of statistics of endgames. In fact they are 365 without taking into account colours. I hope no typos:

Code: Select all

5 vs 1 (70 sets):

KQQQQK
KQQQRK
KQQQBK
KQQQNK
KQQRRK
KQQRBK
KQQRNK
KQQBBK
KQQBNK
KQQNNK
KQRRRK
KQRRBK
KQRRNK
KQRBBK
KQRBNK
KQRNNK
KQBBBK
KQBBNK
KQBNNK
KQNNNK
KRRRRK
KRRRBK
KRRRNK
KRRBBK
KRRBNK
KRRNNK
KRBBBK
KRBBNK
KRBNNK
KRNNNK
KBBBBK
KBBBNK
KBBNNK
KBNNNK
KNNNNK
KQQQPK
KQQRPK
KQQBPK
KQQNPK
KQQPPK
KQRRPK
KQRBPK
KQRNPK
KQRPPK
KQBBPK
KQBNPK
KQBPPK
KQNNPK
KQNPPK
KQPPPK
KRRRPK
KRRBPK
KRRNPK
KRRPPK
KRBBPK
KRBNPK
KRBPPK
KRNNPK
KRNPPK
KRPPPK
KBBBPK
KBBNPK
KBBPPK
KBNNPK
KBNPPK
KBPPPK
KNNNPK
KNNPPK
KNPPPK
KPPPPK

Code: Select all

3 vs 3 (120 sets):

KQQKQQ
KQQKQR
KQQKQB
KQQKQN
KQQKRR
KQQKRB
KQQKRN
KQQKBB
KQQKBN
KQQKNN
KQRKQR
KQRKQB
KQRKQN
KQRKRR
KQRKRB
KQRKRN
KQRKBB
KQRKBN
KQRKNN
KQBKQB
KQBKQN
KQBKRR
KQBKRB
KQBKRN
KQBKBB
KQBKBN
KQBKNN
KQNKQN
KQNKRR
KQNKRB
KQNKRN
KQNKBB
KQNKBN
KQNKNN
KRRKRR
KRRKRB
KRRKRN
KRRKBB
KRRKBN
KRRKNN
KRBKRB
KRBKRN
KRBKBB
KRBKBN
KRBKNN
KRNKRN
KRNKBB
KRNKBN
KRNKNN
KBBKBB
KBBKBN
KBBKNN
KBNKBN
KBNKNN
KNNKNN
KQQKQP
KQQKRP
KQQKBP
KQQKNP
KQQKPP
KQRKQP
KQRKRP
KQRKBP
KQRKNP
KQRKPP
KQBKQP
KQBKRP
KQBKBP
KQBKNP
KQBKPP
KQNKQP
KQNKRP
KQNKBP
KQNKNP
KQNKPP
KQPKQP
KQPKRR
KQPKRB
KQPKRN
KQPKRP
KQPKBB
KQPKBN
KQPKBP
KQPKNN
KQPKNP
KQPKPP
KRRKRP
KRRKBP
KRRKNP
KRRKPP
KRBKRP
KRBKBP
KRBKNP
KRBKPP
KRNKRP
KRNKBP
KRNKNP
KRNKPP
KRPKRP
KRPKBB
KRPKBN
KRPKBP
KRPKNN
KRPKNP
KRPKPP
KBBKBP
KBBKNP
KBBKPP
KBNKBP
KBNKNP
KBNKPP
KBPKBP
KBPKNN
KBPKNP
KBPKPP
KNNKNP
KNNKPP
KNPKNP
KNPKPP
KPPKPP

Code: Select all

4 vs 2 (175 sets):

KQQQKQ
KQQQKR
KQQQKB
KQQQKN
KQQRKQ
KQQRKR
KQQRKB
KQQRKN
KQQBKQ
KQQBKR
KQQBKB
KQQBKN
KQQNKQ
KQQNKR
KQQNKB
KQQNKN
KQRRKQ
KQRRKR
KQRRKB
KQRRKN
KQRBKQ
KQRBKR
KQRBKB
KQRBKN
KQRNKQ
KQRNKR
KQRNKB
KQRNKN
KQBBKQ
KQBBKR
KQBBKB
KQBBKN
KQBNKQ
KQBNKR
KQBNKB
KQBNKN
KQNNKQ
KQNNKR
KQNNKB
KQNNKN
KRRRKQ
KRRRKR
KRRRKB
KRRRKN
KRRBKQ
KRRBKR
KRRBKB
KRRBKN
KRRNKQ
KRRNKR
KRRNKB
KRRNKN
KRBBKQ
KRBBKR
KRBBKB
KRBBKN
KRBNKQ
KRBNKR
KRBNKB
KRBNKN
KRNNKQ
KRNNKR
KRNNKB
KRNNKN
KBBBKQ
KBBBKR
KBBBKB
KBBBKN
KBBNKQ
KBBNKR
KBBNKB
KBBNKN
KBNNKQ
KBNNKR
KBNNKB
KBNNKN
KNNNKQ
KNNNKR
KNNNKB
KNNNKN
KQQQKP
KQQRKP
KQQBKP
KQQNKP
KQQPKQ
KQQPKR
KQQPKB
KQQPKN
KQQPKP
KQRRKP
KQRBKP
KQRNKP
KQRPKQ
KQRPKR
KQRPKB
KQRPKN
KQRPKP
KQBBKP
KQBNKP
KQBPKQ
KQBPKR
KQBPKB
KQBPKN
KQBPKP
KQNNKP
KQNPKQ
KQNPKR
KQNPKB
KQNPKN
KQNPKP
KQPPKQ
KQPPKR
KQPPKB
KQPPKN
KQPPKP
KRRRKP
KRRBKP
KRRNKP
KRRPKQ
KRRPKR
KRRPKB
KRRPKN
KRRPKP
KRBBKP
KRBNKP
KRBPKQ
KRBPKR
KRBPKB
KRBPKN
KRBPKP
KRNNKP
KRNPKQ
KRNPKR
KRNPKB
KRNPKN
KRNPKP
KRPPKQ
KRPPKR
KRPPKB
KRPPKN
KRPPKP
KBBBKP
KBBNKP
KBBPKQ
KBBPKR
KBBPKB
KBBPKN
KBBPKP
KBNNKP
KBNPKQ
KBNPKR
KBNPKB
KBNPKN
KBNPKP
KBPPKQ
KBPPKR
KBPPKB
KBPPKN
KBPPKP
KNNNKP
KNNPKQ
KNNPKR
KNNPKB
KNNPKN
KNNPKP
KNPPKQ
KNPPKR
KNPPKB
KNPPKN
KNPPKP
KPPPKQ
KPPPKR
KPPPKB
KPPPKN
KPPPKP

Apart from the '5 vs 1' and '1 vs 5' issue, the rest looks fine for me:

a) 4 vs 2 has 175 sets and there are other 175 for 2 vs 4 using your notation.

b) Regarding 3 vs 3, just doubling 120 to get 240 is misleading because duplicates of both white and black sides must be removed. There are exactly 15 of them, so 120*2 - 15 = 225, like your number. Duplicates: the same in characters 1-3 than in characters 4-6 ==> KQQKQQ, KQRKQR, KQBKQB, KQNKQN, KQPKQP, KRRKRR, KRBKRB, KRNKRN, KRPKRP, KBBKBB, KBNKBN, KBPKBP, KNNKNN, KNPKNP and KPPKPP.

All in all, I get 365 without taking into account colours and 715 (your 645 plus another 70 from the case '1 vs 5') taking into account them.

Regards from Spain.

Ajedrecista.

Thanks a lot for your input! I'm more confident now. I only take into account the first 70 "5 vs 1" sets because FEG won't generate sets named like "KKQQQQ" which would be the theoretical "1 vs 5", maybe because of the piece order.

[JVMerlino]

Did TheKing even use EGDBs in the search? Unless Johan modified TheKing after CM9000 (the last version I worked on) to do so, you won't get any better play from the engine (and negligible improvement overall). In CM9000, it was the GUI that checked the EGDBs for a move only when there were 6 or fewer men on the board; TheKing did not use the EGDBs in the search.

And regarding generating 7-men, I don't think FEN supported it at all.

jm

According to the FEG readme: "To choose the best move Chessmaster searches 1 ply ahead, or sometimes
2 ply if only btm data is installed." "For 5 pieces we can save some space by omitting either White to move
(wtm) or Black to move (btm). The search for best move must then be
extended from 1 ply to 2 ply. This will cause much more random access,
but is still doable in about 1 second. It will reduce storage to less
than half (using the smallest files of either wtm or btm)."

I'm aware I won't get much stronger play from the engine, this is more like a fun project. And regarding generating 7-men, I'll just try and see what happens

Heh, I wrote that readme, using the very extensive notes that Johan provided me. But now I can see how it is easy to get misled by that first sentence. In fact, it is the FEG probing code that is doing the "search" in this case, and it is being called by the GUI.

[Juan P. Naar]

Hello! Today I finished building the last set:

https://imgur.com/wXO42rx

Here's a screenshot of the entire folder. I can't believe how many files it generated!

https://imgur.com/bgUM9wg

Having completed the entire 6 men, I gave it a shot and tried to build a 7 man set. It worked! I wonder how long it will take to complete though

https://i.imgur.com/BWDcQHR.png

EDIT: I can upload all of the log files (LOS, LOF, LOG) if someone wants them. It weights 10MB compressed (2631 files!).

[Dann Corbit]

I have a friend who would probably like the statistics generated.
If you can give me a download link, I would appreciate it.

[Juan P. Naar]

I have a friend who would probably like the statistics generated.
If you can give me a download link, I would appreciate it.

Of course! Here it is:



Enjoy.