Website founded by
Milan Velimirović
in 2006
7:43 UTC
| |
MatPlus.Net  Forum General SPAM in MPR |
| |
|
|
|
|
You can only view this page!
| | Page: [Previous] [Next] 1 2 3 4 | | (61) Posted by Michael McDowell [Saturday, Mar 20, 2010 16:00] |
My joint with Edgar Holladay had its origins in a tourney run by Ideal-Mate Review for problems in which the initial position showed a shape of some sort. Gene Albert and Edgar were the judges. I composed the figure 8 helpmate (yes, by the old-fashioned method – it would be a few years before I sat behind a computer for the first time!) and sent it to Gene. I had a reply saying that Edgar had composed the same position, and that he proposed sending it as a joint to Chess Life. I remember feeling that there was something not quite right about a judge composing a problem which would be eligible for the tourney he was judging, but I had no option other than to agree.
I have no evidence that Edgar used a computer.
| | | (62) Posted by Dan Meinking [Saturday, Mar 20, 2010 16:26] |
@Michael,
Thanks for that elucidation! I didn't know the IMR part of the story, but I'm certain neither Gene nor Edgar would do anything dishonest. Wonder if Edgar would've given himself a Prize, but I digress... :-)
I suspected all along that you used the 'old-fashioned method'. And I'm certain that Edgar used 'his method'.
Conclusion: A good problem is a good problem no matter how it was 'discovered'!
| | | (63) Posted by Marek Kwiatkowski [Saturday, Mar 20, 2010 16:44] |
And second conclusion:
good problem = good study = 1.0
| | | (64) Posted by Sergiy Didukh [Saturday, Mar 20, 2010 17:03] |
@Gerhard. I don't have problems with evaluation of all the studies based on its artistic criteria. But if someone wants to organize their division into some groups, he may judge analytical studies separately, the set of criteria will certainly change in this case. A composer will have to decide then which tourney to send his original. But I don't support any divisions, especially those that originate from suspicious minds and fiction.
| | | (65) Posted by Marek Kwiatkowski [Saturday, Mar 20, 2010 17:33]; edited by Marek Kwiatkowski [10-03-20] |
Bye-bye chess aristocracy!
Study composers may only divide their own joy of computer tools. Now, they have a chance to make or get correct compositions.
| | | (66) Posted by Sarah Hornecker [Saturday, Mar 20, 2010 17:36] |
This hits the nail if thought about it! Is every twomover mined because it can be fully tested with a computer? :-)
I admit! I would not have made my twomover of last MatPlus without this help! But there is still artistical value in getting the idea and setting up the twenty or so pieces until one finally has a correct position.
Many thanks, Marek!
| | | (67) Posted by Jean-Marc Loustau [Saturday, Mar 20, 2010 17:58] |
@Gerhard (post 59): This way is not satisfactory, because it depends where is the main idea of the study: if this main idea is in the first part of the study (when the position has 7, 8 or more pieces on the board), then the fact that the study concludes with an EGTB position does not mean they were used intensively.
@Sergiy:
1)You wrote: "I must disagree that the presentation of the solution with isolated main line could emphasize ‘ideas’ or ‘themes’". Well it’s a matter of opinions. You speak about your own criteria, I respect that; my opinion is that sometimes it’s true and sometimes it’s not.
2)You wrote: “It is also wrong to divide studies depending on the use (marginal or intensive) of EGTB”. May be it is a bad idea by itself, and in the best of the worlds it would be a bad idea indeed; in fact the idea was finding a solution to the mood prevailing on this subject. I have read here and elsewhere extremely disagreeable, sometimes hurtful and personal words, from both sides. Tolerance and respect ideas of others is clearly not the rule in chess composition. I think chess composition does not need such fights, so the idea was to separate the two sides so that each side can live in peace and develop its “art”. But as I said it is probably not possible for practical reasons. And I even don’t know if it is wishable (obviously you answered “no”).
3) Speaking of disagreeable words, you wrote: “Can you swear that you don’t use TB + ‘Wilhem’”.
Why have I to swear anything? Am I accused of something? Is this a trial? I said earlier I have not used any other tool: Do you want to say that my word has so little value that I should swear? Sergiy I want to believe that your goal was not to be offensive and this is may be a matter of language, because English is not my native language (nor yours). So I answer to your questions:
- Do I use Wilhem? I did not; by the way I don’t know what is Wilhem; by the way I have no programs like Fritz or other Chessbase or things like that, and I would not be able to create such program (may be when I was younger, but I stopped coding 20 years ago, so today I forgot all the computer languages I knew, and by the way I am not interested).
- Why this material? At the beginning no particular reason; I had just fun with EGTB, in fact just for testing them, with different materials, to understand how it works and so on… So at once I put this material on the board… And by chance I got the end of one the lines of the sudy mentionned (in fact before I found another study, much more simpler with the Pawn on 2nd line). After that it is like I have described. This material is very very balanced so there are numerous kinds of correct draws; when I analyzed to understand things (not only from the study itself) I found other very interesting positions, and I followed the same process. Then it becomes like a tree… And also I discoverd some "tracks" to find more quickly interesting positions. All this manually, of course.
So why did I continue with this material? 4 reasons (without order): 1) almost nothing serious have been done with it, thus it is a free field; 2) the potential is enormous because it is very balanced 3) By taste I prefer to go deeper into a subject rather than to go here and there without logic 4) may be by chance I could help to find some principles useful for theory (this is a little presomptuous… I did not know where I go).
- I want to add that all my studies show something I like, something beautiful to my tastes, I would say all of them have a “theme” in my own meaning of this word (for example a nice positional draw is a theme). I have been quite surprized by the great deal of “themes” allowed by this material; one of the most surprizing I found is a classical problem theme: the “4 corners theme”; in this study the rook plays on each corner in continuous way and leaves it (1 Rh1 2 Ra1 3 Ra8 4 Rh8 5 Rh6 leaving last corner), moreover in some “logical” way (thematic tries). To my taste this result is beautiful by itself (the fact it is dual-free); surely the study is analityc because good understanding requires some difficult analyses… But it was a great pleasure to discover it (there also I have “built”: of course I did not found directly this position, but as I am a problemist, this “theme” is in my “culture”, and it appears from an intermediate position that may be it could be achieved… and it was possible indeed!). If only 1 other person likes it, then I will be an happy man :-)
| | | (68) Posted by Sergiy Didukh [Saturday, Mar 20, 2010 21:14] |
Jean-Marc, I didn't mean to offend you. My insistance was indirectly directed to Gerhard. Seeing that he doesn't want to abandon his suspicions I wanted you to repeat to him again that you developped your study using EGTB and still you made enormous creative human efforts. This proves that any methods of composing are good and welcome. It's an evidence to all people who see some kind of threat to the art from EGTB that they are wrong and they should stop sowing panique among chess fans.
| | | (69) Posted by Jean-Marc Loustau [Saturday, Mar 20, 2010 23:16]; edited by Jean-Marc Loustau [10-03-20] |
Sergiy, I apologize for my misunderstanding. About the question of "Art"... What is "Art"? Entire books could be written on this matter... And I will not try to answer!
| | | (70) Posted by Jacques Rotenberg [Sunday, Mar 21, 2010 06:03] |
However, Sergyi, your post was a bit degrading.
I don't fully agree with Jean-Marc : he has not to explain himself, in any case.
All that recalls me what happened in the 80's with the first computers for solving problems : some great composers wrote papers against !!
It seems that each new tool raises interrogations, and fears.
The best answer, I think, is to ignore it. It's like a cold in the head, just wait, and you'll forget it.
In the meanwhile, I must say, all these moral accusations are unpleasant and tiring, in any direction they go.
| | | (71) Posted by Jean-Marc Loustau [Sunday, Mar 21, 2010 09:48] |
@Jacques: I have not interpreted the questions asked as a duty of justification, just... let us say, as "interest on my approach". I don't think Sergiy had an aggressive intent. From this point of view I had no reason not to give these "explanations", and I did not feel compelled to do so.
By the way, if I understand well, it seems that about the matter of the topic, the EGTB, you, Sergiy and me have (almost) the same ideas (even if we have different tastes, what is normal, and perhaps different ideas about how to manage the "differences of opinion").
| | | (72) Posted by Marek Kwiatkowski [Sunday, Mar 21, 2010 14:11]; edited by Marek Kwiatkowski [10-03-21] |
All the discussion exposed, probably against anyone’s will, a huge change in possibilities of “building” of studies. To get or to make, it turn out to be no crucial question, though. It hides another one. I have experience in both groups problems and studies. Thus, believe me that there should be no difference (Fide Album) between problems and studies. Even if engines don’t find solutions in layouts with fortress or “crazy piece”, they can precisely show the lack of other solutions.
By the way, the last Fide Album contains incorrect studies. Let’s hope the last time.
[example]
E. Dobrescu & V. Nestorescu
1st PR Szachy 1969
correction EBUR 1998
Fide Album 1998-2000
 (= 4+6 )
1.h7+ Kh8 2.Kh6 Rf8 3.Rf7 Ra8 4.Ra3 .... 8.Re3 Rd8 9.Rfe7 Rd6+ 10.R3e6 +-
Analysis by Stockfish 1.6.3 JA:
1.h7+ Kh8 2.Rxe8+ Kxh7 3.Re7+ Kg8 4.Kg6 Kf8 5.Rf7+ Ke8 6.Rxg2 h1Q 7.Re2+ Kd8 8.Ra2 Ke8 9.Rg7 Qc6+ 10.Kh7 h3 11.Raa7 Qd5 12.Rg8+ Qxg8+ 13.Kxg8 Kd8 14.Rh7 Ke8 15.Rb7 f4 16.Rh7 Kd8
+- (6.43) Depth: 21 00:01:12 29214kN
| | | (73) Posted by Georgy Evseev [Monday, Mar 22, 2010 11:49] |
@Jean-Marc Loustau
Thanks for your posts
This thread really needs more info from composers who have really used EGTB as a composing aid. My wrong assumption about your work was based on two thoughts. First, I did not imagine how much time you devoted to this endgame. Second, I understand how this work can be automated if proper tools are available. So, my apologies.
But in pinciple you only confirm my main idea. Appearance of new tools allows to achieve things not achieved before OR to do the same things as before but much easier. It is the second possibility which should worry us.
For example, similar tools already exist in problem chess. Vaclav Kotesovec definitely uses some software to generate fairy problems with small number of pieces. I myself also have several computer-generated problems.
In problem world there is real threat of a "problem spam", which at some time can also appear in a study world. For example, Vaclav Kotesovec has sent about 200 fairy problems with 4 pieces to FIDE Album 1995-1997. Separately, they were quite printable (two long solutions with echo finals), but the judges got fed up with them and so no one of these made it into Album.
It is quite possible to make similar searches on EGTB. How will anyone react if, for example, a big series of endgames with two or three mate or stalemate finals will appear in different magazines under the name of some unknown author? This thing can easily be programmed, though it is difficult for me to estimate time needed to compute some result.
At the same time, I know that Yakov Konoval has already automatically found all possible endgames with multiple promotions (probably only in some material combinations), including some known classics.
-----------------
And now, as promised, the positively commented solution of the position from my post 30. In a sense this position is a boundary case: it still can be solved, analyzed and understood by human. The position was found during a search for long wins with Rook against two Knights (with the help of Wilhelm program) and analyzed for some hours to check if the result is worth anything. This is not exactly the "longest win", but the position which I liked most.
So, here goes the commented solution.
 (= 2+3 )
"It is interesting that in initial position black formally have a material advantage. White have to capture both black knights to win, and so they should use the fact that black forces are dispersed. If one knight is captured while another is unable to unite with the king, then white win.
"So, it is natural that white start by attacking both knights 1.Rd1 Sf3! (1...Se2 loses after 2. Rxa1). Now white cannot win by capturing the knight a1 and so they have to pursue another knight. 2. Kf3 (first capture refusal) Sg5+ 3. Kf5 Sh7. Black had to accurately select the squares for the knight, but now they seem to consolidate their forces. But at the same time they limited the mobility of the king. And now white start to play for checkmate. 4. Rg1!, preventing black king from leaving the corner. But now black has time to move the second knight nearer 4... Sb3 5. Ke6! (5. Kg6? Kg8=) Sc5 6. Kf7!, threatening 7 Rg8#. But the second black knight is now near enough and black can play 6...Sf8!. White again (third time) cannot capture a knight - 7. Kxf8? Kh7!=.
"But let us look at current position more attentively. It has become quite static. Black Kh8 and Sf8 cannot move because of immediate mate. But to keep the mating net white king also should not move and the rook cannot leave the g-file. Second black is also not as free as it looks. It cannot move too far away from black king, because white will be able to capture Sf8, but at other squares its movement can be limited by the rook.
"But now white have to move the rook along the g-file. Where?
"After 7. Rg4? Sce6! white are in zugzwang for the first time.
"After 7. Rg2?! Se4! white are again in zugzwang.
"After 7. Rg3?! Scd7! we see a third zugzwang for white, because seemingly "killing" move 8. Rg5? is refuted by Sf6! and knight is defended by fork.
"But black have only three good squares for the knight and white have fourth: 7. Rg5!!. Now 7...Sce6 8. Rg4! (black in zugzwang!); 7...Se4 8. Rg2! (black in zugzwang!); 7...Scd7 8. Rg3! - black again in zugzwang and after 7...Se5+ 8. Kxf8 one knight is captured and the mobility of second is quite limited and it soon perishes, for example 9...Kh7 10. Ke7 Kh6 11. Kf6 Sd7 12. Ke6 Sf8 13. Kf7 and so on.
| | | (74) Posted by Jean-Marc Loustau [Monday, Mar 22, 2010 14:55]; edited by Jean-Marc Loustau [10-03-22] |
Georgy : you wrote : « I understand how this work can be automated if proper tools are available”. I must say in my mind this would have changed nothing: the great number of hours I spent was not for finding or building the study, but mainly for understanding; by the way in this particular case there were 2 levels of understanding:
1) “Understanding” of the material: I don’t say I have a 100% understanding of the material, of course not, but the very long time I spent (I said 90%, it’s approximate) helped me to understand or identify some generic win and draw principles or strategy, or tracks; this was quite difficult because there was almost nothing written.
2) “Understanding” of the position itself: I mean of course finding the threats, what happens if black plays other moves, but also what happens if white play some other attractive moves (tries or attempts).
This work of human understanding was mandatory to me. I think I would not have published a position without a “very good understanding” (OK this is subjective), and it’s the case with all the positions I have published. I can thus affirm that, for example, contrary to what has been written somewhere, my study of the Gurgenidze JT can be “analyzed and understood by human” (“solved”, I don’t know, probably not, but I am not a solver, and sincerely this is not important to me). And as I said before, it is not necessary to spend the same deal of time I needed to get a satisfactory 1st level understanding (but I think that a “well commented solution” is mandatory).
So, surely, as you say, “Appearance of new tools allows to achieve things not achieved before”; but the work for understanding these things has to be done (this is my opinion) and can be heavy (often the case), and it will remain a human work; at least in the study field. From this point of view, if this work is actually done, the EGTB should allow great progress in the knowledge of chess. That’s the reason why I hope I will find some time, some day, to write something somewhere (and which allow by the way to make my studies more easily understandable) :-)
A last word about Kotesovec case: I always have thought that if he had sent to the album only the 4 or 5 best among his 200 problems, probably most of them would have been selected!
(PS: I like you study KR/KSS!)
| | | (75) Posted by Hauke Reddmann [Monday, Mar 22, 2010 16:07] |
Newsflash! Leonardo da Vinci revealed to be an android
from 35th century! Louvre immediately trashed the Mona Lisa!
Hauke
P.S. As EGTB reveal, Rg5! is only the shortest win -
only Rg4 spoils it, Rg2/Rg3 just waste time (White must
play Rg1 in the next move and restart). Interstingly,
in that position Black would draw with Sd3.
| | | (76) Posted by [Monday, Mar 22, 2010 18:02]; edited by [10-03-22] |
Hauke writes:
>Newsflash! Leonardo da Vinci revealed to be an android
>from 35th century! Louvre immediately trashed the Mona Lisa!
Well, one fairly well identified threat in the earlier postings
is not that of quality but that of assessing it. If someone
decides to 'mine', and then sends the result (in five thick binders)
to a tourney, or a chess column, or, heavens forbid, FIDE album
judges, the recipients are pretty sure to feel as if they were under
attack.
As long as they have a choice to scrap it all, fine. But if they
have agreed to assess and report on every contribution, things
probably look a bit different.
So perhaps it's more a question of 'Louvre uncovers 258 other
Mona Lisa paintings found in their vaults. Announces that it will
now close for five years while Louvre experts assess which Mona Lisa
is the real one.'
or perhaps even
'Louvre uncovers 258 other Mona Lisa paintings found in their vaults.
Removes all Leonardo's paintings from display, as they are most likely
to be mass-produced work instead of carefully handcrafted art.'
or possibly even
' ... Louvre experts agree that it is impossible to create so many
paintings of the same subject, and are re-assessing evidence that da Vinci invented
some kind of camera.'
| |
No more posts | Page: [Previous] [Next] 1 2 3 4
MatPlus.Net Forum General SPAM in MPR |
|
|
|
ISC 2024
