MatPlus.Net

 Website founded by
Milan Velimirović
in 2006

14:32 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 General SPAM in MPR
 
You can only view this page!
Page: [Previous] [Next] 1 2 3 4
(1) Posted by Sergiy Didukh [Wednesday, Mar 10, 2010 15:50]; edited by Sergiy Didukh [10-03-10]

SPAM in MPR


I suggest to post here any critical comments concerning the contents of the magazine. I hope this will make the magazine even better, clean it from the spam material.

In the latest MPR number one candidate for SPAM AWARD is the useless article about EGTB. Again and again Mr.Roycroft doubts that composers deserve complete authorship of any position with 6 pieces and less because they all are in EGTB with guaranteed outcome.
Since Mr.Roycroft’s views are so radical he has been many times asked to demonstrate how EGTB can help to compose studies and this way give weight to his words. The new article reminds me of a ‘User’s guide’: a lot of unimportant information with a little notice at the end to go to the closest service centre in case of problems. If you take the EGTB to the ‘service centre’ you’ll be told that it wasn’t designed to perform the ‘Search’ function, it’s just a device for checking some doubtful lines that sometimes happen in studies. Since this is the only help from EGTB, it’s understandable why there are so many compliments to it in the article. However, the composer in me says that it is not a better tool than any chess program (Fritz, Rybka). These programs evaluate positions in seconds. When the lines are too complicated this can take longer time, but such lines ruin studies, so composers don’t need to know their outcome, composers get rid of them!
 
(Read Only)pid=4971
(2) Posted by Juraj Lörinc [Wednesday, Mar 10, 2010 17:12]

Perhaps this would be better placed in the section Magazine > General (comments on the magazine)?
 
 
(Read Only)pid=4972
(3) Posted by Marek Kwiatkowski [Wednesday, Mar 10, 2010 18:25]

Each author of studies should have now "& (e.g.) John Smith" (computer + chess engine + EGTB). Nobody can prove that don’t use computer to make studies. Thus, 1.66 for a study it is 0.66 in gift. Many of us don’t know that study composers have now such powerful tools.
But, it is fact that a judge should price only the final effect. The question “how it was made” has no sense, even if it is a 6-pieces (or less) position.
 
   
(Read Only)pid=4973
(4) Posted by Sarah Hornecker [Wednesday, Mar 10, 2010 19:09]

Well, for the orthodox problemist (#2, #3, #n), 1.00 of the 1.00 points are a gift if in an EGTB position. At least according to Roycroft's ideas of composing. :-)
 
   
(Read Only)pid=4975
(5) Posted by Miodrag Mladenović [Wednesday, Mar 10, 2010 21:04]

So if someody creates a database of all possible positions and soles #2, #3, #4 etc. we should stop composing problems completely because whatever we create will be already done in the database? This does not make any sense. Whoever creates database like this one (and I am sure one day computers will be enough powerfull to do this within a real time) should better create some logic to extract real problems and published them somewhere. I thnk that any problem stored in the database but not published somewhere can be ignored like it never existed.
 
   
(Read Only)pid=4976
(6) Posted by Jacques Rotenberg [Thursday, Mar 11, 2010 01:14]

There is also another way to explain things :

The help of computers is not seen as a reason to discard short problems

What difference is there between a kind of program, and another ?

Technically the fact that the EGDB and the usual solving programs work in a complete different way seens not to be relevant.

Composers have no matter how works the solving program they use !

When a new and powerful tool is created it changes the world, that's true.

For me it's just an exciting new task : to compose interesting 6 pieces moremovers or studies.
 
   
(Read Only)pid=4979
(7) Posted by Mihai Neghina [Thursday, Mar 11, 2010 10:21]

[Just a personal opinion] Even more interesting is to compose (orthodox) studies/puzzles that mock chess engines -- i.e. studies/puzzles which are (badly) misevaluated by computers.
 
   
(Read Only)pid=4980
(8) Posted by Sarah Hornecker [Thursday, Mar 11, 2010 12:09]; edited by Sarah Hornecker [10-03-11]

Such as your very good one, or many a Kasparyan study:

3k4/2pq3p/pp5R/4P3/P6P/2PN4/1PN3K1/8
(= 9+6 )

Mihai Neghina
MatPlus 35 (2009), correction by Peter Martan in Glarean magazine 2009
White to move and win

http://www.chichitza.com/sah/constrictor.html
 
   
(Read Only)pid=4981
(9) Posted by Frank Richter [Thursday, Mar 11, 2010 13:10]

Why not simply 1.Rd6!? and we win with the advantage of 2 knights?
 
   
(Read Only)pid=4982
(10) Posted by Sarah Hornecker [Thursday, Mar 11, 2010 13:11]

Most probably 1...c7xd6 would make the win a bit difficult...
 
   
(Read Only)pid=4983
(11) Posted by Frank Richter [Thursday, Mar 11, 2010 13:15]

Of course ...
 
   
(Read Only)pid=4984
(12) Posted by Jacques Rotenberg [Thursday, Mar 11, 2010 22:41]

@ Mihai
"[Just a personal opinion] Even more interesting is to compose (orthodox) studies/puzzles that mock chess engines -- i.e. studies/puzzles which are (badly) misevaluated by computers."

ok, but why 'more' interesting ?

This is interesting, and this is interesting too.
 
   
(Read Only)pid=4986
(13) Posted by Kevin Begley [Friday, Mar 12, 2010 04:12]

I didn't bother to read Mr.Roycroft's article.

I am well aware what his position is on EGTB positions, but perhaps somebody can explain to me exactly what is he advocating for?
I expect he must be intent upon amending the Codex (retroactively??), to change the current position on EGTB positions...
The Codex clearly does not consider his view legitimate, currently -- has he conceded this point?

If not...
The Mat Plus editors have a responsibility to inform their readers that, while Mr.Roycroft may be something of an authority on EG studies, his opinion on EGTB positions is NOT supported by the Codex (past or present).

Chess problems are clearly defined in the Codex -- which lists no minimum number of units necessary (to constitute an endgame study).
Therefore, failure to properly credit the author of such a problem, would constitute plagiarism.
Failure for judges to accept (or properly evaluate) such problems (unless expressly forbidden in thematic tournaments), constitutes an injustice.

I support Mr.Roycroft's right to make his case (wrong, though it may be -- it fails to recognize the evolution of "composition").
In fact, I'm not in favor of the "SPAM" label, on this thread, because alternative viewpoints should be more welcomed.

If he dodges the crux of the issue, Mat Plus editors were unwise to allow him to duck the consensus of experts in their forum.

I haven't read the article, so I'm only guessing why so many here have suggested that it misses the mark.

Thanks.
 
   
(Read Only)pid=4987
(14) Posted by Sergiy Didukh [Friday, Mar 12, 2010 08:29]

Kevin, if you read this article you'll see that it's not a grounded and clear exposition of an opposite point of view. It's a kind of spam that gets on composers' nerves, it's a suspicion of plagiarism of everyone who has studies with 6 pieces and didn't mention the help of EGTB. What help?
 
   
(Read Only)pid=4988
(15) Posted by Kevin Begley [Friday, Mar 12, 2010 11:12]; edited by Kevin Begley [10-03-12]

OK, sigh, maybe it's worse than I first guessed.
Mat Plus should have slapped a disclaimer on his article.

Mr.Roycroft's view is not official PCCC policy (neither by Codex, nor by precedent); and discrediting authors based upon unsubstantiated claims of EGTB-Anticipation would constitute a plagiarism.

What's next... somebody will start crediting old problems to Chessmaster/Rybka/Fritz/Popeye/Natch/Alybadix/Win Chloe/etc?
Because, hey, why wait for the slothy PCCC to make a ruling -- if Mr.Roycroft gets away with it, anybody can do it!

There is a governing body, and I would encourage Mr.Roycroft to start abiding by their codex.
I understand that he may disagree with the codex -- but he should amend it first (by demonstrating a delegate majority), before he advocates widespread discreditation of fellow composers.

It's disrespectful to do this; particularly, given the venue-illusion that Mr.Roycroft speaks with some authority on the matter.
 
 
(Read Only)pid=4989
(16) Posted by Sarah Hornecker [Friday, Mar 12, 2010 11:15]; edited by Sarah Hornecker [10-03-12]

Interestingly I won a prize with an EGTB study in EG (!).

(= 6+5 )

Siegfried Hornecker
EG 2008(?), study 16.032
4th prize
White to move and win

1.Sb3! b1Q+ 2.Kxb1 Rxb3+ 3.Ka2 Rb2+! 4.Kxa3 Rb8 5.Bb7+!! Rxb7 6.Rd8+ Rb8 7.Rxb8+ Kxb8
(= 3+2 )

8.h5! Sc6! 9.h6 Se5 10.h7 Sg6 11.Sh6!! Sh8!! 12.Kb4 Kc7 13.Kc5 Kd7 14.Kd5 Ke7 15.Ke5 Sg6+! 16.Kf5 Sh8 17.Kg5 Ke6 18.Sf5 Ke5 (Kf6 19.Kh6 etc.) 19.Se7 Ke6 20.Sc8 Ke5 21.Kh6 Kf6 22.Sd6 Sg6 23.Se4+ Kf7 24.Sc5/i Sh8 25.Sd7 Sg6 26.Se5+ Sxe5 27.h8Q wins (27...Sf7+???)

i - or 24.Sf2 Sh8 25.Sg4 Sg6 26.Se5+ Sxe5 27.h8Q wins



EDIT: Now saw Kevin's reply above. Here's a bold case of database mining I found last year!

(= 2+3 )

White to move and draw

1.Rf4? g3 2.Rg4 Rc3 3.Kf7 Kc2 4.Kg6 Kd2 5.Kh5 Ke2 6.Kh4 Kf2 7.Kh3 Rf3!zz
1.Rf5!! g3 2.Rg5 Rc3 3.Kf7 Kc2 4.Kg6 Kd2 5.Kh5 Ke2 6.Kh4 Kf2 7.Kh3 Rf3 8.Rg4!zz Rf8 9.Rf4+ Rxf4 stalemate

Imagine who was the author...
It was Nikolay Dmitrievich Grigoriev, and the study won 4th prize in "64" 1937. Shame on him! :-)

More info about the great composer: http://www.chesspro.ru/_events/2008/grigoriev.html
 
 
(Read Only)pid=4990
(17) Posted by Mihai Neghina [Friday, Mar 12, 2010 11:26]; edited by Mihai Neghina [10-03-12]

"ok, but why 'more' interesting ?"
@Jacques
It is just an opinion which I consider true, but this does not imply that everyone necessarily has to agree with it. I'll try to explain my reasons:
First, I am quite new to the puzzle composing world, and computer misevaluation was so far a minor goal (and difficulty checkpoint) for my studies.
Second, I feel chess is a little betrayed when "solvers" switch on engines before finding the idea/solution. With tough-for-computer studies, the fun doesn't end when engines are started, and these can be seen as "advanced chess" studies.
Third, engines may defeat humans in regular chess, but they cannot win such winning positions or draw such drawing positions. Such studies mock engines, and it's a way of breaking the blind trust in computer evaluations felt by a fairly large number of chess enthusiasts.

Unfortunately, there are only a handful of themes leading to computer misevaluations...

@Siegfried
Thank you for posting the study.
 
 
(Read Only)pid=4991
(18) Posted by Administrator [Friday, Mar 12, 2010 14:26]

 QUOTE 
Kevin: I haven't read the article, so I'm only guessing why so many here have suggested that it misses the mark.


The article (both parts) is available now in DL Archive; here is a direct link:
http://www.milanvel.net/pub/Polemics_Roycroft.pdf

Whoever disagrees with it (or is even annoyed by it) is welcome to respond. After all, the discussion was what Mr. Roycroft wanted in the first place.

Re "SPAM": Mr. Roycroft kindly responded to the invitation by the editor of Mat Plus Review for additional explanation, so "spam" is the last thing his text can be associated with.
 
   
(Read Only)pid=4992
(19) Posted by Sarah Hornecker [Friday, Mar 12, 2010 15:26]

I don't find the posting but you promised at the Fairy Kings article linking to not make this (i.e. articles in MPR for download) a regular thing. :-)

Just a reminder.
 
   
(Read Only)pid=4993
(20) Posted by Sergiy Didukh [Saturday, Mar 13, 2010 08:05]

To my mind, Siegfried is very lucky to get the 4th prize for his study in EG. This study has too many artistic drawbacks and the judge clearly pointed them out in his comments: disharmony of the both parts, a long tiresome and technical win in the finale. It should have been a commendation at maximum. But it's good that the study is shown in this thread, because it demonstrates how many traps have to be avoided when developping a position that pops up from analysis or 'surfing' through different lines. The main Siegfried's mistake here is that the 5-man position is not interesting at all. Yes, it's long, dual-free, but not interesting - not good for a study.
Grigoriev's study shows how to make excellent works of arts.
 
   
(Read Only)pid=4994

Read more...
Page: [Previous] [Next] 1 2 3 4

MatPlus.Net Forum General SPAM in MPR