Website founded by
Milan Velimirović
in 2006

7:12 UTC
ISC 2021



Remember me

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

Rating lists


MatPlus.Net Forum General The history of A-to-B-Chess
You can only view this page!
Page: [Previous] [Next] 1 2
(1) Posted by Per Olin [Wednesday, Oct 12, 2011 19:50]

The history of A-to-B-Chess

Among us there are many persons good at finding historical facts. It would be interesting to find historical problems, where the stipulation is to find the play from position A to position B. In the case of proofgames, it means that A is not the standard initial position.

Two historical facts are known to me:

1) In PDB can be found probably the oldest example of the genre by Arthur Cyril Pearson published in Chess Monthly in 1879 (PDB number 12-P1124557). The position was wKb5 wSa3 wSc3 - bKc1 bSb2 bSd2 and the second diagram had the colour of all pieces changed. Stipulation: "Change one of these positions into the other in as few moves as possible, playing White and Black alternately, and not moving either of the Kings into check." The solution by the author needed 10 moves (1. Sc2 Sb3 2. Ka3 Kd2 3. Ka2 Kd3 4. Kb1 Kc4 5. Sd4 Sa1 6. Kc1 Kb4 7. Sb3 Sc2 8. Sa4 Sd1 9. Sb2 Sc3 10. Sd2 Sa3), while the shortest found is in 7,5 moves, not unique: 1. Sa4 Sdc4 2. Kb3 Kd2 3. Ka2 Sd1 4. Sb2 Kc3 5. Kb1 Kb4 6. Kc1 Sc3 7. Sb1 Sa3 8. Sd2

2) Dittmann in his book Der Blick zurück mentions a special case of this form, where A=B and the position is to be reached in an uneven number of halfmoves. By Ceriani there are several problems, quite amusing, in the PDB with the stipulation 'orthoreconstruction'.

Having myself published my first proofgame A-to-B in 1993, the question is: can more examples of A-to-B-Chess dated last millennium be found?
(Read Only)pid=7465
(2) Posted by Cornel Pacurar [Saturday, Oct 15, 2011 01:10]

Actually both a=>b and ser-a=>b have much older mediaeval roots.

Civis Bononiae 236, Ar. 562
(= 2+2 )

To interchange the Red and Wh. Kts in XVI moves. The Kts are confined to the square of 9 squares in the corner a8.
(Murray, H. J. R., A History of Chess, page 673)

Civis Bononiae 288
(= 1+11 )

The Wh. Kt. takes all the other pieces, finishing with Rh1 in XXV exactly.
(Murray, H. J. R., A History of Chess, page 677)

Of course, there are also more recent examples. For instance, even Shinkman's famous 782 Pittsburgh Gazette Times 29/xii/1912 ( could be seen as a (conditional) ser-a=>b ("In how many moves can the King and Queen be transposed without moving the King to "g2". At the conclusion the other pieces to be in same position as at the start.")
(Read Only)pid=7467
(3) Posted by Cornel Pacurar [Saturday, Oct 15, 2011 15:28]

And here is the move-length record for the A=B subset:

Jean-Michel Trillon
Special Prize Rex Multiplex 1982
Dedicated to Jacques Rotenberg

(= 3+58 )

Return to the diagram position in (exactly) 32551.5 moves
(Read Only)pid=7468
(4) Posted by seetharaman kalyan [Saturday, Oct 15, 2011 21:11]

How many GBs. required to write the solution ? :)
(Read Only)pid=7469
(5) Posted by Bojan Basic [Sunday, Oct 16, 2011 12:50]

Here is a related article I found somewhere on the Internet. I cannot remember where I found it (and Google search shows nothing), but I uploaded it to my page and it will be there for some time.
(Read Only)pid=7472
(6) Posted by Hauke Reddmann [Sunday, Oct 16, 2011 14:56]

@seetharaman: ~33000*4 byte (in dumb mode).
Since I am in dumb mode too at the moment, what hinders
W&B to play pendulum with Qb3-a2 and Kf6-f5?

The full explanation of the problem in human-intelligible format
MAY take the one or other Gb, though :-)

(Read Only)pid=7473
(7) Posted by Kevin Begley [Sunday, Oct 16, 2011 16:29]; edited by Kevin Begley [11-10-16]


As already noted, there are A->B problems, with a unique solution, where diagram A = diagram B.

Of these, there are perhaps four important sub-types -- either:
1) diagram A = diagram B (no other attributes matter),
2) A=B + some attribute is apparently stipulated for alteration (e.g., author stipulates two likes units must platzwechsel),
3) A=B + some "hidden" attribute is implied to have been stipulated for alteration (e.g., en passant/castling rights must alter to meet a secondary stipulation), and
4) A=B where A and B represent complete POSITIONS, not merely diagrams (which must be fully equivalent).

Clearly, the author's stipulation is intended to imply a specific sub-type (#3).
It conceals that an "exact" number of moves is required (read: A=B + alter player on the move) -- otherwise, it suffers a glaring design flaw.
I highly doubt the author would have missed this -- but, you make a good point: you should not have to guess.

This is also found in unique Proofgames (sometimes getting to the diagram a half-move sooner is a cook, other times it constitutes an invalid solution).
Unfortunately, the simple A->B stipulation does not clearly distinguish what is intended.
Many stipulations assume that the solver will make the correct "educated guess" -- nobody seems to care that this is the definition of stipulation failure.
[Besides, even our good stipulations -- those which are stipulations, and unambiguous about significant details -- are too informal, non-universal, and unmemorable.]

I believe the correction depends upon a new, more developed understanding of problemist culture.
Problemists have an uncontrolled desire to conceal information.
We don't tell who is on the move (figure it out), we conceal castling rights, we conceal the possibility of en passant capture, we conceal the board coordinates, we don't state that the diagram is illegal (maybe you should remove one of those 9 black pawns!), etc, etc, etc...
It is a culturre of -- as Nabakov called it -- "Splendid Insincerity."

Except, we tend to perpetuate insincerity, even when it produces no punch-line -- even when it clearly infringes upon what the audience is entitled to expect.
It's all the elements of a tragic-comedy, except absent the thought-provoking humor... in fact, entirely devoid any element of humor.

The rules should be universal and unambiguously clear.
Stipulations should clearly, logically, and universally convey what the author intended as the complete mission for the solver.
Fairy conditions should not be hiding within aims/stipulations.
You should not have to guess which rules a particular author has implied.

In cases where these are deliberately concealed, the audience expects to laugh.

Yet some -- as if fooled into believing the insincerity itself is always a sign of good humor (and high culture) -- fight to perpetrate a carelessly bad humor.
That's how a joke played to enrich the audience has evolved into several unfunny clowns on the stage.

This problem is quite interesting -- like a carefully choreographed "SOKOLY" (Polish Wedding dance).
It should not be diminished with the "record" label.
Records are a fantasy concocted to distort the appreciation of art (generally, to allow childish comparisons of non-comparable items).
I have a world record for being the only one uniquely me, but this says nothing about why I may be interesting.
(Read Only)pid=7474
(8) Posted by Cornel Pacurar [Sunday, Oct 16, 2011 18:49]

Biased opinions like "Records are a fantasy concocted to distort the appreciation of art (generally, to allow childish comparisons of non-comparable items)" should better be avoided. The answer to this was given by A. C. White (in the "Les tours de force sur l'échiquer" introduction) more than one hundred years ago: "task problems require no timid justification from their supporters; they deserve full recognition for what they profess to be and are; they point the way along which the development of the chess-problem, if there is to be any further such development, must be sought; they convict the critics, who call them monstrosities and pitfalls, of being themselves bigoted partisans and sterile conservatives". As to the "record" label "diminishing" the 32,551.5 (!) moves problem: Jean-Michel also published a similar problem just a few half-moves shorter, and the length of the solution was important to him; if anything, this label enhances the problem - we are looking at a problem which solves in 32,551.5 moves, which is the A=B move-length record, like it or not.
(Read Only)pid=7475
(9) Posted by Kevin Begley [Sunday, Oct 16, 2011 19:42]; edited by Kevin Begley [11-10-16]

Oh, I do like the problem...
I don't like your label.

Orthodox problems maybe (and only maybe!) deserve some recognition, when the records are significant.
For example, the longest helpmate has proven to be significant -- but, its worth could not be immediately recognized -- only after a long time trying to surpass it have we come to realize this.

The problem you are ascribing a record to is a) relatively short-lived & untested, and b) highly unorthodox -- thus, your quote does not apply here.
I know of nobody who considers it highly significant as a record...
I expect plenty do find this an interesting work... but, quite apart from orthodox records of established significance, I do not expect to find any serious attempt to surpass this.
Please, don't assume a lack of trying is due to some apparent difficulty in overcoming the present problem.

Furthermore, the ground rules which must be observed to surpass such a problem are entirely unclear -- maybe you think you know them!?.
The trouble here is, you attempt to establish a record to fit a problem, rather than establishing records for problems to chase.

We have had this discussion before -- concerning long fairy series-movers.
As I stated then, one need only concoct a simple fairy condition to obtain an additional half-move.
This can be done indefinitely -- read: bye bye sweet record.

I have seen the results of tourneys cast specifically to achieve some distorted quantity fixation (e.g., with instantaneous unorthodox records, cyclone records, Bohemian records, etc)...
And, I can confidently reiterate: it is an infantile wish to reduce non-comparable art into a form where counting pretends to supplant a truly artistic appreciation.
Surely, we can agree, unorthodox chess problems have so much more to offer.
(Read Only)pid=7476
(10) Posted by Kevin Begley [Sunday, Oct 16, 2011 20:29]

If the world record for being like me were established prior to my birth, and many had attempted to produce such a being, I would hold a record of great significance: just for being me.
If people who knew me were to attempt to establish such a record, say after my birth, it would constitute a complete sham.

The test of time will determine what records are significant.
We need not establish instantaneous records to appreciate every interesting problem.

The merit of a good problem (like the merit of art, like the merit of being me) can not be justified by claim to a concocted record.
(Read Only)pid=7477
(11) Posted by Cornel Pacurar [Sunday, Oct 16, 2011 20:38]

@ Kevin Begley: It is quite unfortunate that you are hijacking this thread. You have also edited your post about 10 times in 30 minutes, making any attempt to properly answer it an ineffective one. So, to end this: your opinions are irrelevant to me, the fact is that this is the longest A=B rendering and I couldn't care less if you like the label or not!
(Read Only)pid=7478
(12) Posted by Kevin Begley [Sunday, Oct 16, 2011 21:15]

Hold on a second, there, Mr.Cornel Pacurar...

I don't make edits to prevent you from responding -- just trying to be more clear.
I'll try to avoid posting too soon... meanwhile, I appreciate your patience.

I am truly sorry you don't care for my opinion... but then, I did not post it specifically for your benefit.
I expect only that you be tolerant of my opinions.

As for hijacking this thread...
First, I deeply resent that you would liken my opinion to such a heinous and despicable criminal act.
As an American, I find that remark particularly offensive.

Second, the claim is dubious....
You introduced this problem to this thread -- a thread concerning the history of a stipulation.
Not because it was historical, but as a side-line -- to declare it as some record (apparently of no established ground rules).

I do appreciate the problem, I'm glad you shared it....
Just be aware that comments you make in this thread open the door for others to reply.
I gave a only very brief comment -- a small aside, in post script, of my opinion concerning the practice of declaring fairy "records."
I think I'm entitled to that.

If you want to challenge that opinion, that's fine... I'll be happy to consider your opinion more carefully... but, please, do so in another thread...
Better still, off-line -- you have my email -- I much prefer to avoid the spectacle which often results here, from the slightest difference of opinion.

Just don't blame my edits for preventing your reply.
And, don't you go accusing me of hijacking this thread.
(Read Only)pid=7479
(13) Posted by Dan Meinking [Sunday, Oct 16, 2011 22:10]

I have zero interest in this discussion, but I am duty-bound to point out the following:

KB to CP in this thread:

"As for hijacking this thread... First, I deeply resent that you would liken my opinion to such a heinous and despicable criminal act. As an American, I find that remark particularly offensive."

KB to DM in another thread:

"You want to hijack this thread -- this forum -- as a platform to hurl insults at a fellow composer (your "good companion")? I will not be a party to it."
(Read Only)pid=7480
(14) Posted by Kevin Begley [Monday, Oct 17, 2011 00:58]

I don't believe I was the first to use that term, Dan.
But, fair enough -- I regret having used that expression, and I am sorry for any offense it may have caused.
(Read Only)pid=7481
(15) Posted by Cornel Pacurar [Monday, Oct 17, 2011 01:26]

Obviously, "hijacking" is not a taboo word and "thread hijacking" has nothing to do with airplanes, or, for that matter, with Americans in particular, thus there is no need to play that card here (it seems quite pathetic, especially for someone living north of the border). The therm "thread hijacking" simply describes a radical departure from the topic of a thread - the original content contained in the post becomes moot, and people now respond to the thread jacker's input and that becomes the focus of the tread.

Of course, I introduced this problem to the thread because it concerns the history of a particular segment of the "A-to-B-Chess" - it is from 1982 and answers to the original question ("can more examples of A-to-B-Chess dated last millennium be found?"). The fact that I selected the longest such example was simply a bonus, and it reflects, indeed, my interests. This does not make it less interesting, and does not diminish its merits - the author's intention is obvious as soon as one sees the total number of moves (without a doubt, the same idea could have been implemented in many thousand moves less). The fact that you do not "know of nobody who considers it highly significant as a record" does not mean much - probably you did not even know that it exists, otherwise you could have posted it here yourself, thus really contributing to this thread. But you did not "like the label", and thus have considered relevant to the thread to ("merely", right?!) embark in a series of remarks (including, but not limited to, "Records are a fantasy concocted to distort the appreciation of art (generally, to allow childish comparisons of non-comparable items)"), that could easily be seen as either reflecting a dubious intent or just plain ignorance. Therefore, as you said, "just be aware that comments you make in this thread open the door for others to reply". In any case, this is my last post on this thread, and out of respect for our colleagues I will ignore any further posts of yours, regardless of what you have to say or how you say it.
(Read Only)pid=7483
(16) Posted by Kevin Begley [Monday, Oct 17, 2011 01:44]

>especially for someone living north of the border.

Where I live does not concern you.
Do you really want to make this a personal matter, Cornel?
I don't consider this the venue for this kind of talk... but, if you want, you and I can have a personal discussion offline.

>...and people now respond to the thread jacker's input and that becomes the focus of the tread.

I told you, Cornel, that term is offensive.
Please discontinue using that term.
There is no justification for it.

I focused on YOUR comment, regarding records (which was an out of place comment, in this thread).
I gave my opinion, in a small post-script...
(Read Only)pid=7484
(17) Posted by Ian Shanahan [Monday, Oct 17, 2011 03:20]

This disagreement just goes to show that there are radically different - and opposing - views on the subject of task records in chess composition.
(Read Only)pid=7485
(18) Posted by Kevin Begley [Monday, Oct 17, 2011 09:07]; edited by Kevin Begley [11-10-17]


This is really not about orthodox tasks & records -- some of these can be quite significant, some are great accomplishments, a few hit the rare jackpot of artistic merit.

But, in Fairy chess -- a relative infant on the scene -- the records belong on a different shelf altogether.
The genre has yet to crawl, while we all want to pretend it's outpacing ghost particles.

It is natural for a chess player to wonder, "what's the longest directmate? helpmate? selfmate? unique game? etc"
But, your average variant gamer does not ponder, "what is the longest number of moves to go from a diagram position to the same diagram position, with an alternate player on the move, using only an 8x8 board, but any number of orthodox chess men, placed anywhere, and no fairy conditions, etc etc etc.?"
That's why advocates can only label such things "record" -- the moment they try to pin down specifics, everybody begins to cringe at the headline buried in fine print.

Already, the constraints here would lead the Guiness Committee to reject such a record (it appears too contrived).
Remember that old book -- the one which dubbed the Winchester Mystery House, "the world's largest 'haunted' house."
Well, the ghost story was known to have been concocted hoax, in the interest of luring tourists to view what may be the worst architectural planning in history.
Sarah Winchester is given distinction, when all she did was ignore real professionals, who tried to advise her how to avoid living in an absurd maze.
And, here is a committee which carefully investigates, and provides detailed information on what you need to do to break a record!
Whereas, in fairy chess, the record-buffs can't even begin to detail the constraints.

It is well known that Cornel has been pursuing and promoting records in fairy chess, for some time...
As such, any expression of a contrary view -- which I know is shared by many problemists (perhaps to varying degrees) -- might be taken more personally than warranted.
It's not meant to be taken personally... it's a shame we have to tip-toe lightly around such issues, to avoid incivility (a larger shame, in this forum).

As for ignoring me -- quite sad to see an aspiring artist avert their eyes... especially to that which shakes their core beliefs.
He suggests I might be ignorant...
We all have a long journey, to reach enlightenment, but, what is the path toward ignorance, if not to ignore other opinions.
That's not my path, brother, that's the wrong way.
(Read Only)pid=7486
(19) Posted by Juraj Lörinc [Monday, Oct 17, 2011 10:05]

Now the thread seems to be really diverted from the original topic...
(Read Only)pid=7487
(20) Posted by Kevin Begley [Monday, Oct 17, 2011 17:03]; edited by Kevin Begley [11-10-17]

Quite right... enough about records in unorthodox A->B problems.
Returning to something closer to the scheduled flight plan...

Consider the history of A->B problems, as recorded by problem databases.

PDB has barriers to entry of A->B problems -- it's easy to add a PG, not so with A->B problems (except in cases where B=A).
Problems which require more than one diagram generally require special entry.
I count only [mod: ~50] A->B problems in the PDB database -- B=A in [mod: ~46], leaving only 4 entries where B!=A (2 of which are cooked).
Either I am overlooking a great many in my search (problems may be entered in a scatter), or a significant history of A->B is not reflected in this database.
And, given that the entrants in this database tend to gravitate more toward the retro genre, this is particularly noticeable.

search: [mod: see improved search in next post]

In Win Chloe, these problems can be entered easily enough, but money is a barrier to access.
I count only 161 A->B problems in the Win Chloe database -- B=A in 64, leaving 97 problems where B!=A (2 of which are cooked).
My filter may fail to account for some, but searches are generally more reliable here, due to a more formal storage structure.
Problem coverage here is generally much better in all other (non-retro) genres.

filter: aim = Attendre la 2nd position a partir de la 1st... -OR- stipulation = Retour a la position du diagramme

Any improved criteria to locate such problems -- for either database -- would be appreciated.
Obviously, I am missing a number of problems which could be reclassified as A-to-B (such as the clear historical example given by Cornel).
It's also worth noting that, except in cases where B=A, you can not step through A->B solutions on either database.

Difficult to estimate how much A->B history goes unrecorded in databases... it would seem little is preserved (electronically).
(Read Only)pid=7489

Page: [Previous] [Next] 1 2

MatPlus.Net Forum General The history of A-to-B-Chess