MatPlus.Net

 Website founded by
Milan Velimirović
in 2006

16:27 UTC
ISC 2024
 
  Forum*
 
 
 
 

Username:

Password:

Remember me

 
Forgot your
password?
Click here!
SIGN IN
to create your account if you don't already have one.
CHESS
SOLVING

Tournaments
Rating lists
1-Jan-2024

B P C F





 
 
MatPlus.Net Forum Internet and Computing Yet another chess problem database
 
You can only view this page!
Page: [Previous] [Next] 1 2
(21) Posted by Dmitri Turevski [Thursday, Mar 12, 2009 18:03]

Pleased to serve you, Siegfried:
http://dt.dewia.com/yacpdb/?id=167060
 
   
(Read Only)pid=3215
(22) Posted by Paz Einat [Thursday, Mar 12, 2009 23:46]; edited by Paz Einat [09-03-12]

A quick look at the databse... I noted that my problem #6326 doesn't have the full intention. The thematic try 1.Rb5? is hidden away. This try leads to reciprocal change of the mates following the unpins of pawn d7 (1...Qd3 & 1..Qe3) - the main point of the problem. Thus, also the theme is not mentioned. Similarly, in #6343 the reciprocal changes, mentioned by the letters, is not in the themes.
 
   
(Read Only)pid=3216
(23) Posted by Dmitri Turevski [Friday, Mar 13, 2009 05:50]

Why didn't you fix that?

This pattern recognition can be tricky, if you let it loose it will find things that are not really there, and if you get it too strict it will fail to find what was intended. In this case it failed to recognize that "dualized" (by underpromotions) variations d8Q/d8R and d8Q/d8B were intended to be the same move. Not sure about the second problem, i guess it is because departure squares of Q and R are different in the phases.

Anyway, i have fixed both.
 
   
(Read Only)pid=3217
(24) Posted by Paz Einat [Friday, Mar 13, 2009 10:22]

As I mentioned this was a quick look. Is it possible for anyone to fix things?
 
   
(Read Only)pid=3218
(25) Posted by Sarah Hornecker [Friday, Mar 13, 2009 10:45]

Dmitri Turevski has installed some Wiki software, so everyone can edit. You need JavaScript enabled, though, and also for a captcha site when you want to save edits.
 
 
(Read Only)pid=3219
(26) Posted by Paz Einat [Friday, Mar 13, 2009 11:13]

OK, played with it a bit and got the hang of it. Very nice!!
 
   
(Read Only)pid=3220
(27) Posted by Hauke Reddmann [Friday, Mar 13, 2009 17:24]

Yikes, 220 problems on my HDD and only 42 listed here!
This calls for some edits after my holiday :-)

Is there also a dupe-finder installed? Probably it's much faster
when I input everything without looking what is already there.
(Especially since I have themes, anticipations etc. ready)

Hauke
 
 
(Read Only)pid=3221
(28) Posted by Dmitri Turevski [Friday, Mar 13, 2009 19:30]

Paz, i'm really glad that you liked it.

 QUOTE 
Is it possible for anyone to fix things?

This is the very point of it.

Hauke, good thing you have a holiday coming. I hope the adding button will be ready when it's over.
 
   
(Read Only)pid=3223
(29) Posted by Harry Fougiaxis [Monday, Sep 21, 2009 09:35]; edited by Harry Fougiaxis [09-09-21]

Well done on your nice effort, Dmitri. In case that you have not noticed, I'd like to point out that it seems there has been a serious error in importing certain FEN and/or PGN files, where the letter N (for knight) was unfortunately parsed as of nightrider.
 
   
(Read Only)pid=4124
(30) Posted by Dmitri Turevski [Monday, Sep 21, 2009 11:56]

Thank you very much, Harry. To my shame, i didn't notice that, could you please give me a link to some corrupted problem?
 
   
(Read Only)pid=4126
(31) Posted by Harry Fougiaxis [Monday, Sep 21, 2009 12:38]

ID 2847, 2848, 15037, 66328, 68008 are a few examples; I did a random check only.
 
   
(Read Only)pid=4128
(32) Posted by Dmitri Turevski [Monday, Sep 21, 2009 13:13]

All look fine to me (eg http://dt.dewia.com/yacpdb/?id=15037 etc). Perhaps you are checking old revisions?
Before fairy pieces support was added both N and S could be used to indicate a knight, after that (~ end of July) all Ns were auto-converted to Ss and few problems with nightriders were added: http://dt.dewia.com/yacpdb/?id=258194
 
   
(Read Only)pid=4129
(33) Posted by Harry Fougiaxis [Monday, Sep 21, 2009 13:39]

My bad, false alarm, sorry! As you mention, I was in fact looking at the old diagram(s) included in the revision history...
 
   
(Read Only)pid=4130
(34) Posted by Sarah Hornecker [Monday, Sep 21, 2009 19:46]

Anyway, FEN and PGN should use the international standards now where S is used for knight. What will they do when knights are finally replaced by nightriders, as suggested by Pierre Drumare?
 
   
(Read Only)pid=4135
(35) Posted by [Tuesday, Sep 22, 2009 21:10]

'S' for Knight is not in any international standard. A convention, maybe.

The closest thing there is, is the FIDE handbook, part E.I.01B.C, where algebraic notation
is described. And in C.3 there's provision for a player using any set of character used in his
country -- but that does not allow for S in English notation. German, sure. Scandinavian
language, also. And perhaps a few more. But that's for 'real' chess, which is the area both
PGN and FEN are targeted at.

It would be up to the problem world to set its own standards in this respect -- and I think it was
mentioned briefly in the discussions at the Computer Matters group meeting at Wageningen. But I believe
the XML-based format was thought to make this form of notation obsolescent, and it was not
pursued any further.

It's desirable to have a standard, I agree. Doesn't really matter if Knights are represented with N
instead of S, and Nightrighters with X or 9.
 
   
(Read Only)pid=4138
(36) Posted by Steven Dowd [Tuesday, Sep 22, 2009 22:23]

Yet the German product Gustav uses "N" for knight in its FEN as does Alybadix...... correct?
 
   
(Read Only)pid=4140
(37) Posted by Olaf Jenkner [Tuesday, Sep 22, 2009 23:08]

Years ago when I made Gustav for Windows I was looking for a FEN-definition.
I googled and found KQRBNP. It seems logical, because Forsyth was an english man (or scottish...).
Now I wonder that some software or websites use the german S (Springer/Knight).

Olaf
 
   
(Read Only)pid=4141
(38) Posted by [Thursday, Sep 24, 2009 09:42]; edited by [09-09-24]

Olaf Jenkner writes:

>Years ago when I made Gustav for Windows I was looking for a FEN-definition.
>I googled and found KQRBNP. It seems logical, because Forsyth was an english man (or scottish...).

It's more complex than that. Forsyth's original suggestion used 'Kt', and underlined letters to indicate
black (See C.N. 5205 at http://www.chesshistory.com/winter/winter39.html.)

Early applications of it also used boldface and italics as alternatives -- underlines were more difficult
to do for the typesetter. Here's an example from Rowland and Rowland: The Problem Art, 2nd edition (1897):

8 | B 7 | 2 s 3 S 1 | 2 s k 2 p 1 | 1 p 3 B 1 | 1 P 3 P 2 | 2 S 2 Q | 7 K

(Slater, 1. pr, Milwaukee Telegraph, s.a., #2)

(but I can't add the italics correctly here) with the additional explanation:

'The Black pieces should be represented by small italic letters, and the white by Roman capitals.
This system appeared in the "Souther Trade Gazette" (Kentucky) and has been adopted by the
"Glasgow Weekly Herald;" and is probably the most convenient for printer and reader.'

Glasgow Weekly Herald is the newspaper where Forsyth published his original proposal. And the spaces
between the pieces are there in the original.

There's also a note: 'It will be noticed that S is given for Kt; this is to prevent confusion with K for King.'
(S also seems to have been used by Brownson in some of the early problem monographs he published
-- I would guess it may have been used in his Chess Journal as well.)

A look at some old newspapers will show the need for clarity in this respect: occasionally small
letters are dropped or broken up. In plain text, this doesn't make as much difference as the loss
of a single 't' would make in a Forsyth position. (The opposite situation is also known: when
black pieces on black squares get all muddied up due to imperfect printing technique.)


FEN, it should perhaps be noted, serves another purpose than Forsyth notation: it is intended to
show complete state from a chess game: a FEN position, apart from the position in modern Forsyth
notation must also include a) side to move, b) castling status, c) en passant capture availability,
d) half-move clock and e) full move number. FEN does not allow for any of these to be indicated
as 'unknown' or 'irrelevant', which forces any user/software to lie about information that is not explicit
in the problem.

So a problem in a PGN file, which relies on FEN positions, read by a solving program that does not
know about castling convention may fail to find solutions or cooks if the castling status field or
en passant status is not 100% correct.
 
   
(Read Only)pid=4148
(39) Posted by Iļja Ketris [Thursday, Oct 29, 2009 00:19]

The notation for Knight in modern (computer-friendly one, without italics and underscore) FEN is "N", period.

Any extension of FEN which incorporates fairy pieces must use something else for Nightrider, and must generally be based on English names for fairy pieces. Milan's extension, used on this site, is, unfortunately, presentation-oriented: it says "rook rotated 90° CW", but has no information regarding the nature of the piece.
 
 
(Read Only)pid=4304

No more posts
Page: [Previous] [Next] 1 2

MatPlus.Net Forum Internet and Computing Yet another chess problem database