MatPlus.Net

 Website founded by
Milan Velimirović
in 2006

15:30 UTC
ISC 2020
 
  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-Apr-2020

B P C F





 
 
MatPlus.Net Forum Promenade personal collection & Parry Series Hub
 
You can only view this page!
Page: [Previous] [Next] 1 2 3 4
(1) Posted by Dan Meinking [Tuesday, May 4, 2010 01:37]

personal collection & Parry Series Hub


I updated my personal collection with newer / better entries:

www.alphasleuth.com/pages/article-chesscollection.php

The PDF is free and worth every penny! :-)


Also added a Parry Series Hub repository page to track related developments:

www.alphasleuth.com/pages/parry-series.php

Hope to compile a complete collection of published works in the near future.


--Dan
 
(Read Only)pid=5297
(2) Posted by Dan Meinking [Wednesday, May 12, 2010 01:47]

Just added a pair of fun solver items to the Hub page. The latter is "C-", so cooks are possible!
 
 
(Read Only)pid=5380
(3) Posted by Per Olin [Thursday, May 13, 2010 18:43]

Dear Dan,

Thank you for making your problem collection available and thank you for the joke and understatement of the year: the PDF is free and worth every penny! Glanced through part of the problems and studied the proofgames with board and pieces; very enjoyable! This way of making public one's own works sets a good standard for all the rest of us! Once more: thanks!

Per
 
   
(Read Only)pid=5388
(4) Posted by Dan Meinking [Friday, May 14, 2010 00:03]

Thank you, Per, you are too kind! PDF is a good medium, although my collection could certainly be better organized. One of these days. :-)
 
   
(Read Only)pid=5389
(5) Posted by Dan Meinking [Friday, Jul 9, 2010 04:22]; edited by Dan Meinking [10-07-09]

For those who do not receive StrateGems: Petko Petkov published an instructive article on Parry Series in July-September 2010 issue (SG51). You can find a PDF version at the Hub link above (the PDF is at the bottom of that page). Although the article is very well written and meticulously organized, I do not agree with all of Petko's "theoretical" points. I will post some comments on the Hub page soon.
 
   
(Read Only)pid=5565
(6) Posted by Kevin Begley [Friday, Jul 9, 2010 08:14]; edited by Kevin Begley [10-07-09]

Thanks Dan.

[edit: upon further reflection, I retract my suggestion... self-helpmate seems already to be taken.]

Very interesting article.
Petko's proposal for naming conventions is interesting: pser.hs#n versus pserb.hs#n...

I believe I was the first to bring this peculiarity to your attention (long before your Parry article was published).
If indeed I was, it certainly does not entitle me to set naming conventions (the larger problem community must decide such matters, weighted by input from the inventor & programmers, of course); however, I would like to revoice my original proposal for naming conventions (which I believe I suggested upon disclosure of this curious anomaly).

My suggestion was: pser.hs#n versus pser.sh#n (read: parry-help-selfmate versus parry-self-helpmate).
To be honest, I primarily favor this because the reversal has real world meaning (ala self-help books).
In parry-help-selfmate, black has the initiative (like in ser.help...), whereas white does the heavy lifting in parry-self-helpmate (as in ser-self...).

Then again, I'll never understand why white goes first in a hs#n problem (since it is first, and foremost, a help-problem, black should go first).

On the other hand, Petko's proposal might be more clear for solvers, particularly if the b is underlined (underlines don't seem to work in this forum, but if they did, something like: pser.hs[u]b[/u]#n).

And, it may be that self-helpmate (sh#n) is a stipulation of entirely different meaning (hehe, my brain hurts just thinking about it).
[edit: correction, the logical meaning of "self-helpmate" is actually simply a semi-reflexmate (force a position where you can helpmate your own King in 1 move).]

The really important questions are:
1) do problemists have a preference (what form is most clear & concise)?
2) does the popeye team (or other fairy problem programmers) have a preference?
3) does the pser inventor have a preference?

That's my 2 cents. :)
I look forward to your commentary about Petko's article.
 
   
(Read Only)pid=5566
(7) Posted by Dan Meinking [Saturday, Jul 10, 2010 06:11]

Added my comments (on Petko's article) to the Hub page (see link above).
 
   
(Read Only)pid=5574
(8) Posted by Kevin Begley [Saturday, Jul 10, 2010 10:03]; edited by Kevin Begley [10-07-10]

Great comments Dan.
I agree almost completely, but, there are a few points which require further discussion...
I'll be, as always, long-winded. :(

point 1) Is pser- a Fairy Invention?

Dan: "I do not consider Parry Series to be any more "fairy" than normal series."
Petko: (roughly quoted) "according to the FIDE Album, normal series is considered fairy."

I find no dispute here. Series-movers are fairy, and parry-serries-movers *might be* no more fairy than series-movers.
Should they have their own special home in SG's S&S section?

I used to believe that SG's S&S section was a good idea, to help pull fairy forms more into the mainstream.
I assume this strategy helped make helpmates and selfmates more widely accepted.
However, the SG breakdown can be easily shown to be invalid.

If you publish your example (#3) as a pser.h#12, it goes into the Selfmates & Series section.
If you publish the same problem as a s#12 (White Must-Check), it goes into the fairies section.
And, if you pull Must-Checks into Stalemates & Series, another hybrid form may present itself -- it goes round and round.

Help-Selfmates are arguably much less fairy than series-movers or pseries-movers, yet they appear in fairies.

There are no simple dividing lines -- they exist only to play favorites, and serve only to hold back this art form.
There need be only be one section, if problems with Retro Play are properly designated.

point 2)
Petko: "When composing Pser Reflex problems... reflex tries are obligatory."
Dan: "Try-play is always a good thing to have. A very good thing, even. But one shouldn't view tries as an absolute necessity."

If you compose a reflex-mate (to use the old convention), without reflex-tries, it is best published as a semi-reflex-mate.
So, isn't it logical that a pser-r#n, without reflex-tries, should be published as a pser-semi-r#n?

point 3)
"Petko goes on to state that .. if X=Y (where X/Y = number of half-moves by the series-side / idle-side), is
true, the problem should be stipulated as White/Black Must-Check , or UltraSchachZwang (USZ). Sounds logical, but it's incorrect."

I mostly agree with you here, but things become complicated when X=Y and all moves (for one side) are parrying checks.
In this case, the solutions for pser- and Must-Check become completely equivalent, yes?
I do see your point: solving the problems may be an entirely different experience (pser- helps avoid triviality).

Your fine example (#3, DM, Mat Plus Forum, 2010) does function both as "pser-h#12" versus "s#12 White Must Check."
But, whereas the later is a trivial solve, the former offers the solver much more to consider.
Perhaps a better example can show thematic pser-tries, with a "Must Check" solution -- then your point would be clear.

However, Petko has a retort to play: problems which are equivalent in solutions, without thematic pser-tries, should be cast into the Must-Check category.
Petko's better argument -- which cannot be brushed off as dogma, because it would help keep database consistency -- is to default to the methodology of precedent.
And, if we can avoid a philosophical debate about imposed difficulty in compositions, I think I can show that Petko is correct.

Suppose I am working to restore several old, unsound Must-Check problems...
How am I to objectively determine which ones function better under the two possible forms?
Should I always impose the utmost difficulty possible (pser-), or default to the methodology of precedent (Must-Check)?
Suppose I make a new problem -- have I complete liberty to chose either form?

It is easy to see both sides of this -- neither view can be entirely debunked.
However, ultimately, I believe precedent favors Petko's view.

A r#n is deemed inappropriate when it could have been published, in simpler form, as semi-r#n.
So, shouldn't this precedent also hold for pser-problems, which could be published, in simpler form, as Must-Checks?

Maybe precedent is wrong in this case -- as it does reduce your very nice problem into a considerably trivial solve.
Or, perhaps the pser-composer has an obligation to employ some strategic elements (to avoid a Must-Check reduction).
All roads lead to a deeper philosophical debate, but if you aim to override precedent, the burden is yours to show its failure.
 
   
(Read Only)pid=5575
(9) Posted by Dan Meinking [Saturday, Jul 10, 2010 13:44]; edited by Dan Meinking [10-07-10]

@Kevin,

I'll briefly reply to (3). In this post...

http://www.matplus.net/pub/start.php?px=1278761756&app=forum&act=posts&fid=gen&tid=781

... you argue that precedence should be 'corrected', but here you argue that precedence should be strictly adhered to.

This isn't about precedent, but PREFERENCE. If I create a new Parry Series that just happens to work as USZ (like the example given), I would choose Parry Series every time. That's my preference.
 
   
(Read Only)pid=5576
(10) Posted by Kevin Begley [Sunday, Jul 11, 2010 16:12]; edited by Kevin Begley [10-07-11]

@Dan

>"you argue [another thread] precedence should be 'corrected', but you argue [this thread] precedence should be strictly adhered to."

Incorrect.

In the other thread, I do make a case that our stipulation convention is flawed, and requires overhaul... yes.
I have discovered a remarkable method to unify, and hierarchically categorize stipulations, remedying numerous conventional flaws.
I do argue, in the other thread, for new convention, yes.

In this thread, I simply stated that precedent supports Petko's argument; and thus, the burden is yours to show it is in error.
And, without any objective evidence that one form is superior to the other, well, that is a major burden on your case...
...orders of magnitude more difficult to what I'm arguing in the other thread.

Your "comments" offer only a single problem (DM #3) to address this matter...only to leave the reader to infer why pser- is better.

I think I've inferred your meaning: while the solutions may be identical, the solver's experience is entirely different.
And, giving composers liberties to set the solver's experience may be as fundamental as having artists set the mood for the viewer.
If this is what you are implying, I do get it, and as I've said, I cannot debunk this...

Nevertheless...
Neither can I debunk Petko's very rational argument, which is grounded in precedent (established to preserve database consistency).

You still haven't clarified your position...
Would you allow each composer liberties to chose, or reclassify all USZ problems (with exactly matching pser- solutions) into parry-movers.

If you cannot objectively demonstrate that one such problem (your own) should be in pser-form, how will you demonstrate all should?
Will you address the issue of database overlap (inconsistency), which would result from allowing composers such liberties?
I just don't see how you make your case, without delving into deep philosophical waters... down to where few have surfaced victorious.

>This isn't about precedent, but PREFERENCE. If I create a new Parry Series that just happens to work as USZ ...,
>I would choose Parry Series every time. That's my preference.

I tend to agree with you there, but unfortunately, I don't think playing to the crowd's preferences is enough.
 
   
(Read Only)pid=5580
(11) Posted by Cornel Pacurar [Sunday, Jul 11, 2010 19:06]

Thanks to Dan, the members of the Parry Series Workshop (http://www.chessproblems.ca) had the opportunity to take a close look at and analyze Petko Petkov's impressive article three weeks in advance of its publication. It has been received with mixed feelings, and it generated numerous discussions, both in the workshop and behind the scenes.

Without a doubt, it is certainly refreshing and encouraging seeing this hot topic X-rayed and dissected in such a thorough manner by Petko Petkov, and some of the ideas exposed are very interesting.

At the same time, however, some ideas and conclusions are, in my humble opinion, questionable. Petko outlines that his conclusions are subjective, but this subjectivity, associated, in the author's own words, with “important theoretical and practical ideas”, could easily be misconstrued; having all these laid out in a very didactic manner by an author of this caliber, an IGM which, amongst other things, almost touches the sky with the number of his FIDE Album points, simply elevates that subjectivity to a different level! It is my belief that, to quote Israel Scheffler, "without objectivity as a guiding ideal there can be no science or, indeed, any rational deliberation whatever".

I do not want to dilute this post and get caught up in semantics and nomenclature (even though I will probably expose some of my thoughts about this as well, in another post), so i will get to the point, going beyond some of Dan's initial comments.

1) Petko concludes (or accentuates the following important theoretical and practical idea..): “Pser requires a series of half-moves by black or white.

In all fairness, and will all due respect, this is incorrect, and also begs the question: why should some Parry Series be more equal than others?! Not only that a series of (consecutive) half-moves is not required by any known and accepted conventions or definitions (the theoretical part), but, more importantly, this can easily be refuted in practice! Let's take a look at the following Parry Series scheme:

Scheme
(= 2+1 )

phser-=5

Solution: 1.g2-g3 + Kh4-h5 2.g3-g4 + Kh5-h6 3.g4-g5 + Kh6-h7 4.g5-g6 + Kh7-h8 5.Kf8-f7 =

The succession of half-moves is as follows: w-b-w-b-w-b-w-b-w and thus there are no white or black consecutive half-moves! Instead of simply stating “quod erat demonstrandum”, I will paraphrase a question: is this a real Parry Series?! Yes, it is, and it will always be!

2) “The following also applies: if the number of all half moves for side which realizes the series is “X”, and the number of parry half-moves from idle-side is “Y”, then X>Y (X is greater than Y).

Dan has already covered this in great detail in his comments and in the post above, but I would also like to add the following: another example of a problem that “happens to work” as BUSZ is my own phser A->B in 37 moves, 4th Honorable Mention, Good Companions Quick Composing Tourney, 2010 (http://www.alphasleuth.com/qctXweb6.pdf). I previously outlined the motives and the rationale behind composing this problem in this very forum – http://www.matplus.net/pub/start.php?px=1278822538&app=forum&act=posts&fid=gen&tid=703. The aim being the highest number of consecutive check-and-parry moves, it was certainly appealing to me to show this without additional regular moves. At that time I did not have BUSZ in mind at all, and even now, after all this, I do not feel the urge to add introductory moves just to have “X>Y”. Should this prevent this problem from achieving (actually, maintaining) the status of “real Parry Series”? I beg to differ.

3) “[..] from the idle-side [..] we must have two half-moves: a parry half-move, which parries the check and the last, final half-move, which gives mate, stalemate, etc.

Something that might, eventually, be true for a certain subset, is not necessarily true for the whole set, and one should avoid such generalizations. Two half-moves of the idle-side are certainly not a must! Here is basic example:

Scheme
(= 2+2 )

phser-!=5

Solution: 1.h2-h3 + Kg4-f5 2.Kf2-g3 3.Kg3-h4 4.Kh4-h5 5.h3-h4 !=

Parry On!
 
   
(Read Only)pid=5582
(12) Posted by Dan Meinking [Sunday, Jul 11, 2010 20:26]; edited by Dan Meinking [10-07-11]

@Cornel -- Thanks for the further elucidations. See my comments below on related discussion.

@Kevin --

"You still haven't clarified your position...
Would you allow each composer liberties to chose, or reclassify all USZ problems (with exactly matching pser- solutions) into parry-movers."

OK, let's take your approach. If we MUST classify an all-checks Parry Series as USZ (just because all moves happen to be checks), then that must apply to all other genres too. For example:

DM, The Problemist 1997
(= 6+3 )

h#3 b) bBe5=>g1 (6+3) [ White Must Check??? ]

a) 1.Ke3 Re2+ 2.Kxf3 Re3+ 3.Kxf4 Rf3#
b) 1.Kf1 Be2+ 2.Kg2 Bf1+ 3.Kh1 Bg2#

Surely there must be HUNDREDS of published problems (esp. longer selfmates) where one side or the other just happens to give check at every turn.

The declaration of "must-check", or even "semi-r#", is a TOOL to be used at the composer's discretion. If an USZ is unsound as a Parry Series, then USZ it must be. If a semi-r# is unsound as a r#, then semi-r# it must be.

But the composer shouldn't HAVE to declare "all checks" or "no tries" in their stipulation.
 
   
(Read Only)pid=5583
(13) Posted by Kevin Begley [Sunday, Jul 11, 2010 20:51]; edited by Kevin Begley [10-07-11]

@Dan,

>"If a semi-r# is unsound as a r#, then semi-r# it must be."

I did not say unsound.
If a problem functions as both (r#2 and semi-r#2), but has no "reflex-tries" (as Petko likes to call them), it is considered to be incorrect (if not worse) to publish this as r#2.

The same way it would be disgraceful to add this unnecessarry checking condition to your witty helmpate.

Composers do not have such liberties here.
Reflexmate is not a "TOOL" that works in this case (they are responsible to use the "CORRECT TOOL" for the job!)... why should parry be different?
 
   
(Read Only)pid=5584
(14) Posted by Dan Meinking [Sunday, Jul 11, 2010 21:32]; edited by Dan Meinking [10-07-11]

With direct-mates and self-mates, composers declare tries by adding check-marks. But with reflex-mates, they declare tries by virtue of the stipulation? That makes no sense.

As Cornel alluded to: a "must-check" problem happens by design, not by chance. If it is incumbent upon the composer to declare "all checks" when NON-checks are permitted (as in Parry Series), then that must be applied to every other genre. Why should the h#3 above be any different?
 
   
(Read Only)pid=5585
(15) Posted by Dan Meinking [Sunday, Jul 25, 2010 19:51]

Added a new article, "Parry Series: Flashback", and updated my 'Parry Series vs. UltraSchachZwang' arguments, on the Hub page:

www.alphasleuth.com/pages/parry-series.php
 
   
(Read Only)pid=5692
(16) Posted by Ian Shanahan [Friday, Jul 30, 2010 12:08]

Forgive me throwing a cat among the pigeons, fellows, but the Parry idea really is as old as the hills: it dates back to checking zigzags and the somewhat simpler 'Black moves only to check' condition.

Some years ago, I published in "The Problemist" a R#18 Circe, Black moves only to check - it was a complex logical wenigsteiner, as I recall. Around the same time, I began planning a comprehensive anthology of Australian series-movers (with luck and better health, this will see the light of day eventually...). I decided to include this problem therein because it really was a hybrid series-mover.

Invoking mathematics, strict move-alternation (by analogy, two-dimensional) is 'orthogonal' to pure series-play (which is therefore monodimensional). So parry-series, zig-zags et al. are 'fractals' whose dimensionality is between 1 and 2.
 
   
(Read Only)pid=5721
(17) Posted by Dan Meinking [Saturday, Jul 31, 2010 02:19]

@Ian:

I hope you're feeling better SOON!! Looking forward to your series treatise. :-)

The form "black plays only when in check" (BPOWIC) is in fact discussed in the aforementioned article. Those who participate in Cornel's series workshops know that the idea of series problems and/or Kings moving when checked dates much, MUCH further back. (related article forthcoming)

While it's certainly true that some Parry Series ideas (mostly selfmates with a checking finale) could be expressed as BPOWIC, the parry series-mover offers room for growth that BPOWIC, must-check, and zigzag forms lack. That's because check is NOT a requirement on the final move in Parry Series.

This allows for nearly all pser-h* forms (which normally do not end in check + counter-check), and even some pser-s* forms which use zugzwang in the finale:

DM, StrateGems 2009
(= 6+8 )

pser-s#11 (6+8) C+
1.Bb1 2.Be3+ Kh5 3.Bc1 5.Bxe2+ Kh4 6.Bd1 8.Bf2+ Kh3 9.Bg1 10.Bh5 11.Rg4(zz) Rxg1#

To your health, sir!
 
   
(Read Only)pid=5724
(18) Posted by Ian Shanahan [Saturday, Jul 31, 2010 21:36]; edited by Ian Shanahan [10-07-31]

Many thanks for your good wishes, Dan. (I have type-II diabetes which has led to stage-4 renal failure: I'm fighting both now with naturopathic as well as Western medical methods ... but it will be a long struggle to regain my former energies while circumventing dialysis and kidney transplants.)

Anyway, there seems to be a point of confusion: "Black moves only to check" is an utterly distinct condition from the one you describe. Here, white plays moves AVOIDING check from Black until it is advantageous to attract a check, which Black is OBLIGED to play. The only thing in common with Parry etc. is the intermittent, 'fractalous' alternation.

My point is simply that zigzags and BMOTC are additional precursors to your excellent idea - though quite different in spirit, if not in move-structure.
 
   
(Read Only)pid=5725
(19) Posted by Dan Meinking [Sunday, Aug 1, 2010 06:05]

@Ian -- My goodness. I hope you're able to PARRY all that ails you in short order!
 
   
(Read Only)pid=5726
(20) Posted by Dan Meinking [Saturday, Sep 25, 2010 20:32]

For those who did not receive SG52 (StrateGems, Oct-Dec 2010), I wanted to share a new variant, Legal Parry Series, invented by Valery Liskovets of Belarus! Legal Parry Series (denoted legal-pser-*) functions the same as normal parry-series except that the idle-side moves not only in reply to check, but also when they have no retro-move available.

Here's a light original (posted with Valery's permission, of course) to demonstrate:

Valery Liskovets, original
(= 2+5 )

legal-pser-h#5 (2+5) C-

Intent hidden:
1.Ba7 2.Rc5 3.Rg5 h7! (white has no retro-move, so must be on-move!) 4.Rc5 5.Rc7 h8=Q/R#

It'll be interesting to see how composer's exploit this clever Parry Series twist! Valery and I, in fact, have a couple of joints (slated for SG53) that nicely expand on the above matrix. :-)

Here's a simple selfmate application:

DM original
(= 6+2 )

legal-pser-s#13 (6+2) C-

Intent hidden:
1.Bc5 2.f8Q 3.Qf3 4.Qa3 5.Rd3 10.g8R 11.Rgd8 c6! (black has no prior move, so must be on-move) 12.Rb8 13.Rb5 cxb5#

I thank Valery (now, and retro-actively, of course!) for this new inspiration. :-)
 
   
(Read Only)pid=5980

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

MatPlus.Net Forum Promenade personal collection & Parry Series Hub