Website founded by
Milan Velimirović
in 2006

15:42 UTC
ISC 2022



Remember me

Forgot your
Click here!
to create your account if you don't already have one.

Rating lists


MatPlus.Net Forum Threemovers More Nth degree black corrections in 3#
You can only view this page!
Page: [Previous] [Next] 1 2 3 4 5
(1) Posted by Jean-Marc Loustau [Tuesday, Dec 8, 2009 13:09]

More Nth degree black corrections in 3#

Miodrag Mladenovic has published in summer issue of MPR 2007 a very interesting article about Nth degree black correction in 3#; in fact I did not know it until recently, but as Milan quoted it in his award of 3# MPR 2007, I asked to a friend to show me the article and I warmly advise anybody who is interested in black correction to read it (many very good problems in it).

The purpose of this post is just to show 2 more examples of this theme: Miodrag could not know them, the first one because the solution has been published at the end of 2007 (and by the way its thematical content, a 4th degree, has never be mentionned, nor in the solution, nor later, and variations were published in wrong order), and the 2nd one because it is a very recent publication.

These 2 problems have a little story; the 1st one has been conceived in 1985 during the spring meeting in Fontenay-sous-bois: the theme of a quick tourney was black correction in 3#; we did not succed to show a sound version in time, but some days later we finally got a first good position; nevertheless the 3 authors were not able to agree on it, and even after working some time, it has been impossible to find a “final common position”… So the problem was almost forgetten. Fortunately, searching in very old papers, it has been unearthed in 2006, and worked again, and then we succeed to find a version which was really improved and which, I think, each of us likes. The story of the 2nd problem is quite the same: I have conceived it I would say between 1973 and 1975, when I was a beginner. I did not succeed to build it and it was also forgotten… and also found recently in very old papers, so, as I had some time to loose, I tried to achieve what I was unable to do as a beginner…

Michel CAILLAUD, Jean-Marc LOUSTAU and Jacques ROTENBERG
Phénix 2006
(= 11+13 )

1 Bc6! threats 2 Bxd5+ Kxd5 3 Qb7#
1… Rc4 ad libitum (ex: 1… Rxc5), 2 Sf3! threats 3 Sxg5#; 2… Bxd2 3 Sxd2#
The opening a6-d3 is used here (Somov).
1… Rd4! 2 Sd7! threats 3 Sf6#
and not: 2 Sf3? Bxd2! (3 Sxd2+, Rxd2!)
The new harmful effet is the fact 2… Bc3?? is no more efficient.
1… Rc3!! 2 Sf7! threats 3 Sxg5#
and not: 2 Sf3? Rxf3! nor 2 Sd7? Rg3 or Rxh3!)
The new harmful effect is the fact 2… Bd2 is no more possible!
1… Rb4!!! 2 Sc3+! Kd4 3 Sf3#; 2… Kf4 3 Rf3# (not 2 Sg3+? Kf4!)
and not: 2 Sf3? nor Sd7? nor Sf7? bxc5+!)
The new harmful effect here is a new interference of Ba5: 2… Bxc3 is no more possible!

So we have a 4th degree black degree correction after a black Rook cross. There is also a kind of duel between Rc4 and Se5. I think the fact the harmful effects from 2nd to 4th degree are interferences to the same black unit (Ba5) is valuable.

Another interesting correction variation in by-play:
1… Rc2! 2 d3+ Sxd3 3 Qxd3#
and not: 2 Sf3? Bxd2! (3 Sxd2+ Rxd2!)

An interesting try is: 1 Be6? (2 Bxd5+) But 1… Rxc5! 2 Sf3 Rc8+!
So the key (1 Bc6), which is a pre-interference of the black Rook on c file, can be seen as a white correction; this try justifies the fact to consider the variation 1… Rxc5 as a specific primary variation, making the Rook cross clear (by the way it would be possible to build the problem with Bb1 on c2 with an accurate cross, but may be the 4th degree is clearer with a random move, and also the variation 1… Rc2 is quite valuable).

The second problem is lighter; it shows that harmful black effects are not necessarily interferences.

Springaren 2009
(= 10+5 )

Here the primary variation is accurate; but first some tries:
1 Bc7 ~? (2 Qc2#)
1… Bxc7 2 Kb2! Block Bc7~ 3 Qc2#
But 1… Bc5!! (black correction: 2 Kb2? Bxd4+!)
So 1 Bxd6!?? (white correction) stalemate!

Solution: 1 e7! Block.
1… Be5 2 Bd6! and 3 Qc2#
1… Bf4! 2 Be5! and 3 Qc2#
and not 2 Bd6? Bxe3!
1… Bg3!! 2 Bf4! Bxe1 3 Qc2#
and not 2 Bd6? or 2 Be5? Bxe1!
1… Bxh2!!! 2 Bg3! and 3 Qc2#
and not 2 Bd6? or 2 Be5? or 2 Bf4? Kd2!

Of course we have 4 Bishop variations with Loshinsky theme, but also 4th degree black correction, harmful effects being that at each degree the black Bishop allows the white Bishop to play to an additional square (the black Bishop cannot be captured because of the stalemate).
The following variation is interesting, because it offers matters for discussion:

1… Bc5 (!!!!) 2 Rd1+! Kxe3 3 Qh3# or 2… Kc3 3 Rc2#
and not 2 Bd6? or 2 Be5? or 2 Bf4? or 2 Bg3 Kc3!

Some analysts could see there a 5th degree, because in this variation there are the same 4 harmful effects as in 1… Bxh2, the new harmful effect here being the fact the black Bishop does not control any more Qh3#.
Nevertheless it can be said that the previous variations use also the fact the black Bishop cannot play any more on c5, so the fact it uses this possibility at 1st move makes somewhat fallacious the reason it has lost access to c5 !
Nevertheless Nevertheless it could be answered that in this special case, because the position is a Block, the fact to be on c5 is, at first sight, less good than the possibility to play on it (so could be meaningful to say the “Bishop has lost access to c5” as a defending move), because after the block tries 2 Bd6/Be5/Bf4/Bg3? the Bishop will precisely have to leave c5 and allow Qc2#; the refutation of these tries 2… Kc3! is due to a new correction effect (the flight on c3 is now still active after white Bishop move).
Well, well… I do not say it is actually a 5th degree; I have no definite opinion about that, I think the 2 answers (it is and it is’nt) can be defended… And by the way it is not so important, it is what it is…

Also a last variation, as previously: 1… Bxc7 2 Kb2! Block Bc7 ~ 3 Qc2#

I would not like to hide a flaw which to my opinion is of a minor importance; after moves 1… Bxe7/Kc3 there is a dual (2 Bd6(+)/Rd1): 1… Bxe7 can be seen as a “bad primary variation”, as it occurs often (does not correct the primary error), so it is almost without importance; the dual after 1… Kc3 is a little bit more damageable, although minor because out of the main theme, forgivable I hope the position being quite light…

There are also several other tries with changed variations and various refutations, but being out of the main play I prefer not to quote theme here. I have to say also that my first purpose, in the 70’s, was to show the Loshinsky theme (to my knowledge it does not exist a lot of examples with 4 black Bishop variations): as a beginner I did not know very well what is a Nth degree!

Well, I hope you will like these problems, and as Miodrag wrote, "there are plenty of possibilities still to be discovered".
(Read Only)pid=4416
(2) Posted by Miodrag Mladenović [Tuesday, Dec 8, 2009 15:47]

I like both problems a lot. They are excellent. Of course I did not know this problem (the first one) at the time when I wrote the article. However it’s one of the best problems showing this theme. Since I composed one problem with bR correction I know how hard is to find a scheme where bR intercepts bB. It’s great that bR completes cross in this problem. Amazing combination.

By the way I composed one more problem showing this theme and it’s already published online (MP36 - #1430). You can find a link to the PDF document showing originals that are going to be published in MP36. I managed to find a matrix where bB intercepts bR. I think it’s a new combination. I found a matrix couple of years ago but I could not manage to create correct problem until recently. Even this version does have some duals but I think that generally it’s OK problem.
I am very happy to see new problems showing this theme.
(Read Only)pid=4417
(3) Posted by Jean-Marc Loustau [Tuesday, Dec 8, 2009 18:12]

A very good problem too (I like particularly the 1st harmful effect, quite hidden)...
I am sure many good problems remain to compose with this theme!
(Read Only)pid=4420
(4) Posted by Miodrag Mladenović [Tuesday, Dec 8, 2009 19:36]

The construction of first threemover is of course excellent. However I noticed that it's possible to put wPa3 instead of bPa4. In my opinion it's a slightly better position because it does allow one more move by thematic bR. Also it does not prevent 1...Qxa3 move so it does not decrease the number of moves of bQ. And finally there are no double black powns on the "a" vertical line. Also by making this change there are no captures of powns at all in resolving position.

Of course since it's a #3 where white has to checkmate black it may be logical that it's better that black does have more pieces than white but in this case I think that there is enough justification to place wPa3 instead of bPa4.

Once again this problem deserves prize regardless of this change. This is more my personal taste but I am wondering what is opinion of other problemists about this question.
(Read Only)pid=4421
(5) Posted by Jean-Marc Loustau [Tuesday, Dec 8, 2009 20:16]; edited by Jean-Marc Loustau [09-12-09]

I understand your reasons, why you prefer a WPa3; this is the kind of reason which makes the style of each composer. I must say that I prefer a BPa4, for other reasons: first I prefer to avoid white Pawns when it is possible (this is also a personnal taste), and also I prefer specifically in this case if Rook has not too many primary variations, because the cross is more visible I suppose (OK that's very subjective).
There is also another possible change (and surely several others, but I don't remind very well all of them): the BRg1 can be replaced by BPf2; this is lighter, but we prefer BRg1 because the Rook gives more use to BSg2 and WPd2 and also we get (not thematic) tries:
1 f6? Rf1! so 1 Rf3? (2f6 and 2 Sg3+ Kd4 3 Sc6#) Rxc5! because after 2 Sg3+ Kd4 square f3 is obstructed (3 Sf3#); in the actual play after 1... Rb4 2 Sc3+ Kd4 then 3 Sf3#, and 3 Sc6# is impossible as square c6 is obstructed: a little echo... Of course this has a very very minor importance (it was not even mentionned when we sent the problem for publication), and may be it does not deserve a Rook for a Pawn, but again a matter of tastes...
(Read Only)pid=4422
(6) Posted by David Knezevic [Saturday, Dec 12, 2009 15:45]

Thank you Jean-Marc for the wonderful post, and to both you and Misha for giving us a chance to follow a short but sharp and educative grandmasters' discussion.

Regarding Loshinski theme, I never viewed at it from your perspective, as a black correction, but now I realize that it does have the corrective nature.

In mid-seventies there was a series of articles on this theme in Soviet chess magazines (64 and Shakhmaty v SSSR, as far as I remember). The stalemate-avoidance mechanism cought my attention and I remember that, seeing a (record) 4-variation rendering by Y. Vladimirov, I wondered why he didn't let the black rook go all the way down to the first row. So I constructed the position quoted below, but didn't put "after Vladimirov" above it simply because already he had used a much older idea (I beleive that it was invented by C. J. R. Samelius in early 1960s).

M.Velimirovic Backo Petrovo Selo JT 1977
(= 15+3 )
1.f7 Rg6,Rg5,Rg4,Rg3,Rg2,Rg1 2.Rg7,Rg6,Rg5,Rg4,Rg3,Rg2
(1... Rxg8 2.Bb1)
(1... Rxf7 2.Rg3/Rg2/Rg1 = triple)

You won't find this position neither in FIDE Album, nor in WinChloe or other databases. Even friends to whom I showed the problem didn't accept it with the enthusiasm. Nevertheless, I am convinced that this is an important task despite the triple in a side variation which I view as the most harmless one, especially with following description of the action by white: "rook goes down and then two squares left" :-)

Here are some examples for reference:

Yakov Vladimirov
Bulletin C.Š.K SSSR 1972
(= 15+3 )
1.Rf6 Rb5,Rb4,Rb3,Rxb2 2.Rb6,Rb5,Rb4,Rb3

Carel J.R. Sammelius Schakend Nederland 1963
(= 11+6 )
1.Rd7 (2.Bxe5) Bg3,Bxh2,Bd4,Bxd6 2.Bf4,Bg3,Rxd6,Bc7
(prototype problem?)

Carel J.R. Sammelius
hm Probleemblad 1964
(= 12+6 )
1.Rg1 Rg3,Rg4,Rg5,Rxg6 2.Rg2,Rg3,Rg4,Rg5
(Read Only)pid=4427
(7) Posted by Jean-Marc Loustau [Sunday, Dec 13, 2009 21:29]; edited by Jean-Marc Loustau [09-12-14]

Thanks to you too Milan for these very good problems! Yours is an impressive task, with the theoretical maximum number of Loshinsky variations, and also a 6th degree. Just a little point, to be accurate, and excuse me for the obviousness: if with this stalemate mechanism the Loshinsky theme is actually linked to black correction, it is of course generally not the case with other mechanisms (the first famous Loshinsky problems for example use a much more difficult mechanism with interferences, but without correction).

I add also a problem which was not in the Miodrag’s article, another 4th degree by Rook with the “antiBristol” mechanism:

The Problemist 1964 1st Prize
(= 10+13 )


1 exf3! (2 fxg3+ Sg2 3 Qxg2/Bxd3#)
1… Rg3 plays 2 Qe5# Qg7 3 Rxc3#
1… Rg5! 2 Qf4 and 3 3 Qd2# (2… Qg5?? Not 2 Qe5? Rxe5!)
1… Rg6!! 2 Rxe1 and 3 Bxd3# (2… Qg6?? Not 2 Qe5?/Qf4? Rxc6!)
1… Rg7!!! 2 Qe7 and 2 Rxc3# (2… Qg7?? Not 2 Qe5?/Qf4?/Rxe1? f7 plays!)
(Read Only)pid=4433
(8) Posted by Miodrag Mladenović [Monday, Dec 14, 2009 07:30]; edited by Miodrag Mladenović [09-12-14]

I am assuming that the correct position is wBe3->e4 and +wRc6. This is another very good example. I did not know this problem. It's been composed in 1964 when I was born. And I do not recall it seeing this problem before. Thank you very much for posting this proble. I'd like to know as much examples as possible.

About Milan's task. Yes it's amazing task. I knew this problem from before but when I wrote article I did not think about stalemate avoiding magnet theme problems as thematic problems. I still prefer problems without magnet theme. Usually they are more complex with more different white mating moves and play is more interesting. Also in Milan's problem there is no clear random move by bR. Yes it's not required but I still prefer prolems where black piece has a random move before it starts correcting play. I think that magnet theme problems (showing Nth degree black corrections) should be a separate category within this theme.
(Read Only)pid=4434
(9) Posted by Jean-Marc Loustau [Monday, Dec 14, 2009 09:51]

Diagram corrected, my apologies.

I agree with you Miodrag; if we agree the main theme is "Nth degree", I prefer too the problems without Loshinsky theme, which are generally richer and more difficult.
Of course that's does not mean that "this problem is better than this one", it is just a "thematical" point of view.
The correction seems may be clearer when there is an actual random move, but to me it is not so important and I like the 2.
(Read Only)pid=4435
(10) Posted by Hauke Reddmann [Wednesday, Dec 16, 2009 11:49]

Ha! I even begin to like 3# :-)
@Milan: What about +wPc5, so you can change the key to 1.Rf8-g8?
I think it's a small thematic improvement to consider.

(Read Only)pid=4439
(11) Posted by David Knezevic [Friday, Dec 25, 2009 20:35]

Hauke: I even begin to like 3# :-)

What took you so long? :-)

Hauke: What about +wPc5...?

Well, it may contribute to a less expected key, but I don't think it justifies the added pawn.
(Read Only)pid=4473
(12) Posted by Hauke Reddmann [Saturday, Dec 26, 2009 13:37]

I thought it was a neat addition that also in the key
the white R follows the black one. Largely a matter
of taste.

(Read Only)pid=4480
(13) Posted by Miodrag Mladenović [Sunday, Feb 7, 2010 12:01]; edited by Miodrag Mladenović [10-02-07]

Here is another problem showing 4th degree correction. I sent this problem to OT and I could not show it here prior to competion of award.

Miodrag Mladenović
4th HM Olympic Tourney 2010
(= 8+6 )


1.Ra5! [2.Rxa4+ Sb4 3.Rxb4#]
1...B~ 2.Qc2+ Kb4/Kxd4 3.Rxa4/Sf5# (a)
1...Bc6!? 2.Sc8 ~/Rb1 3.Sb6/Be2# (ab)
1...Be6!!? 2.Sxe8 ~/Sc5 3.Sd6/Rxc5# (abc)
1...Be4!!!? 2.Be6+ Bd5 3.Bxe5# (abcd)

Black errors:
(a) - opening of line a5-e5
(b) - closing of breaked line e1-e6-b6
(c) - closing of breaked line e1-e6-d6 and line e1-e8
(d) - clsoing of line e1-e6

This is so far the most economical example showing this theme where bB intercepts bR three times. Although problem won only 4th HM I think this is one of the best #3 that I ever composed.
(Read Only)pid=4697
(14) Posted by Andrey Selivanov [Sunday, Feb 7, 2010 21:03]

Very nice mat in three! Picture position, good key and correction!
(Read Only)pid=4703
(15) Posted by Jean-Marc Loustau [Monday, Feb 8, 2010 11:19]

An excellent problem indeed!
(Read Only)pid=4716
(16) Posted by Juraj Lörinc [Monday, Feb 8, 2010 11:50]

What do you think then about judge's commentary? My dilettante translation:

"Black correction in the bishop's play. Unfortunately, the third variation is clearly unsuccessful: firstly it has clearly twomover character and secondly it has the checking 2nd move, while the second and third variations were quiet ones."

Was it really unfortunate or rather the best way how to pursue the economy of means in the difficult thematical area?

Further nitpick, more funny than serious - I did not know Tula town had separated from Russia :-)
(Read Only)pid=4717
(17) Posted by Miodrag Mladenović [Monday, Feb 8, 2010 20:26]

Juraj wrote:
What do you think then about judge's commentary?
Was it really unfortunate or rather the best way how to pursue the economy of means in the difficult thematical area?

Well, I wish the third variation was a quite one. However I could not find a quite variation within this matrix. Personally I think that the main point in this problem is black correction play. It's not easy to find different harmful effects and at the same time useful effects that are preventing previous white variations. However I think that this is a very difficult theme where this check can be tolerated. Of course I do respect judge’s comment. It’s a valid opinion. It’s very interesting definition of this variation: “twomover character”. I like this although I was never thinking this way.

I can confirm that I did not pursue the economy. If there was a way to find quite variation with several pieces more I would definitely go with that position. When I compose problems showing some complex theme I do not pay too much attention to the economy. First I try to find position that works (matching some basic criteria for correctness). Then I try to improve position. I try to economize it but without taking any content. It just happened here that matrix worked so well so that I was able to cut down position to 14 pieces. So far I composed about 350 problems. But I can count on my fingers how many of them made me happy right away when I composed them. This one is certainly one of them.
(Read Only)pid=4722
(18) Posted by Vlaicu Crisan [Monday, Feb 8, 2010 20:47]

Let's hope the judge of OT #3 section, mr Yakov Vladimirov, will be informed about Juraj's question above and will provide an in-depth answer!

In my opinion, this problem has everything:
- all white pieces, except the King, are playing an active role during the solution
- the 4th degree correction is clearly shown, with bBd5 interfering bRe1 on 3 different squares
- the last variation convincingly exploits the closing of line e1-e6, even if it is a checking move
Just compare this with the first problem opening the thread, having a similar thematic (and also a check in the 4th correction), but using 10 more pieces.
By the way: that one is again an excellent composition and I won't be too surprised if it will win a prize in Phenix 2006 informal tourney!

Congratulations, Misha, and just forget about the [rather low] place received! I am absolutely sure we will admire again your composition in the FIDE Album.
(Read Only)pid=4723
(19) Posted by Juraj Lörinc [Monday, Feb 8, 2010 21:11]

I was not trying to suggest the judge's comment is in any way inappropriate. Rather, as the comment was shedding completely different light at the issues related to composing and assessing such a threemover, I thought it would be interesting to confront it with author's opinion.

It happens very often that author's and judge's opinions differ considerably, we (almost) all know this situation so well. But then, general public has the most often only the access to the judge's one, the opinion of the author himself being often unknown.
(Read Only)pid=4724
(20) Posted by Miodrag Mladenović [Monday, Feb 8, 2010 21:28]

Vlaicu wrote:
Congratulations, Misha, and just forget about the [rather low] place received! I am absolutely sure we will admire again your composition in the FIDE Album.

Thanks for your opinion about my problem. By the way I did not want my post to be understood as questioning the quality of the award itself. I posted my problem to complete collection of threemovers showing this theme. Actually I think that award is very good. I did not yet have time to go through the all problems but I saw several top ones. They are all excellent problems. It's not easy to compare problems showing different themes. And also I am sure that if there are 10 different problemists making award all of them would be different. There is always a personal taste. If this was thematic tournament with some specific theme given in advance then it would make sense to comment award.

Also I think that there is a general rule that judges are favoring problems showing themes that they do prefer. That’s somehow logical to me. If you did compose yourself #3 showing some theme then you do have much better feeling how hard is to compose problem showing that specific theme. However if you never composed problem showing some other theme you may not understand completely how big effort is put into that problem. For a while I was composing mainly problems showing some modern theme. I thought it’s more complicated to do than to compose one phase problem. However when I started working on this series I learned that it’s more difficult to compose problem showing this theme and I started appreciate more one phase threemovers. I think that I composed about 30 or more #3s showing reciprocal changes and Lacny theme. However I managed to compose only about 10 problems showing quaternary black correction play.
(Read Only)pid=4725

Page: [Previous] [Next] 1 2 3 4 5

MatPlus.Net Forum Threemovers More Nth degree black corrections in 3#