Problemiste freeware
Matthieu L.
Problemiste is a freeware program entirely dedicated to the chess
problem:
Solves orthodox chess problems: direct mates, helpmates, selfmates
from 2 up to 6 moves,
Writes and prints the solutions with many options,
Prints diagrams (after installing the TT font in the windows/system
directory)
Manages archives which can contain thousands of problems,
And many other features (twins, searching of positions, etc...)
Problemiste supports English, Spanish and Italian languages (Select
"Fichier | Langues" menu item)
PGN Import / Export
French help file.
Don French
This is the first chess problem program that I have seen that does many of
the things I originated in Problematic (www.geocities.com/~donfrench). It
is in general a nice improvement on Problematic, which was written for DOS
many years ago.
Problemiste still can't solve my two compositions, "The problem the computer
couldn't solve" and "Another problem the computer couldn't solve", however.
Is it possible that Problematic is still the only program that can solve
this kind of chess problem?
Don French
Matthieu L. wrote in message <364DD322...@easynet.fr>...
Jori Ostrovskij
>
> Looks good, Mattieu. I found a few minor problems, but generally very nice.
> This is the first chess problem program that I have seen that does many of
> the things I originated in Problematic (www.geocities.com/~donfrench). It
> is in general a nice improvement on Problematic, which was written for DOS
> many years ago.
>
> Problemiste still can't solve my two compositions, "The problem the computer
> couldn't solve" and "Another problem the computer couldn't solve", however.
> Is it possible that Problematic is still the only program that can solve
> this kind of chess problem?
I don't know what kind of problem you are trying to solve, but you
should try Popeye.
It can be found in ftp://ftp.leo.org/pub/rec/games/chess/popeye/. If it
cat't solve your problem, then you have either too much fairy in it
(conditions, pieces, stipulations) or it is too complex to be solved. In
latter case you can try Ilkka Blom's Alybadix and Fairybadix, which are
the fastest problem solving softwares available. They are commercial and
do not support as much fairy chess as popeye, but they are _very_ fast.
- Jori -
Anders Thulin
Jori Ostrovskij <jo...@ifi.uio.no> wrote:
>(conditions, pieces, stipulations) or it is too complex to be solved. In
>latter case you can try Ilkka Blom's Alybadix and Fairybadix, which are
>the fastest problem solving softwares available. They are commercial and
>do not support as much fairy chess as popeye, but they are _very_ fast.
Last time I looked, they we're also non-Windows, so getting many
solutions into, say, a Word document required some careful planning.
Check problemiste ... it's surprisingly fast.
You might want to check that you do get version 1.80. I tried a few
hours ago, but I got version 1.70 instead.
--
Anders Thulin Anders....@telia.se 013-23 55 32
Telia ProSoft AB, Teknikringen 6, S-583 30 Linkoping, Sweden
Anders Thulin
>I don't know what kind of problem you are trying to solve, but you
>should try Popeye.
Both problems are e.p. capture problems, and so require proof of
last move.
It's an interesting, but not a serious lack of functionality.
Don French
>last move.
>
> It's an interesting, but not a serious lack of functionality.
>
Oh, I quite agree. In fact, Problematic doesn't actually prove the legality
of the move. It only looks for a possible e.p. capture if it could find no
other solution. Then, if it finds pawns in the appropriate places to
suggest a possible e.p. capture, it attempts it. If it succeeds, it
pronounces that the solution. There are two problems with this approach.
If a problem with an e.p. key has a second solution or a cook, it will find
only the other solution. And the e.p. capture might not in fact be legal
and the problem would give a false positive on soundness. But these
considerations were in my opinion minor in comparison to finding an e.p. key
when it was the only way to solve a problem. Much better than pronouncing
all e.p. capture problems unsound.
Anders Thulin
Don French <donfrenc...@photographer.net> wrote:
>Oh, I quite agree. In fact, Problematic doesn't actually prove the legality
>of the move. It only looks for a possible e.p. capture if it could find no
>other solution. Then, if it finds pawns in the appropriate places to
>suggest a possible e.p. capture, it attempts it. If it succeeds, it
>pronounces that the solution.
I'm disappointed ... :-) I figured out your second 'computer can't
solve' problem after a while (and, at last, even why Pa2, a piece
that seemed thoroughly idle, was there), and thought that a computer
might be able to do that kind of reasoning as well ...
>If a problem with an e.p. key has a second solution or a cook, it will find
>only the other solution. And the e.p. capture might not in fact be legal
>and the problem would give a false positive on soundness. But these
>considerations were in my opinion minor in comparison to finding an e.p. key
>when it was the only way to solve a problem. Much better than pronouncing
>all e.p. capture problems unsound.
A possible alternative might be to announce the e.p. capture(s) as
tentatively possible, but that *real* brain capacity is needed to
decide the question finally.
There's a similar problem with castling, of course, only in reverse,
as castling is assumed to be legal, unless proved illegal, while
e.p. is assumed to be illegal, unless proved legal.
Problemists do it with difficulty ...
Don French
not a trivial task to do a retrograde analysis of a position to definitively
determine the last move. I think it might make an interesting topic for a
master's thesis in computer science. I wouldn't mind working on it myself
some day, but I am busy doing other things at this time. When I wrote
Problematic I took the approach I did as an easy-to-accomplish compromise.
Your suggestion of announcing the e.p. solution as a tentative possibility
is a good one.
If anyone does take on this task, I suggest this problem by Sam Loyd,
composed in 1894, as a good test:
/1b3SrR/P1pp2pK/1p2p1RQ/5pP1/1P3pk1/1P6/2PPP2P/S7/ Mate in 4.
It is the first known example of an e.p. capture key where the WK does not
reside on the fifth rank, thereby making the elimination of the single
square advance of the black pawn quite a bit more difficult.
Don French
Anders Thulin wrote in message <72rb3r$mad$1...@jupiter.linkoping.trab.se>...