[ List Latest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]
Comments by CharlesGilman
Excellent, it works (as demonstrated in my update to Neutral Subject Chess) - although I notice that the "update" button converts the braces in the original list into %7B and %7D in the final full diagram code, and that is how I pasted into the submission. Is this expected behaviour?
Yes, that works on the whole - although it does mnot seem to work with the coloured images in this set. This is unfortunate as I am particularly keen to use the Diagram Designer on variants with more than 2 players. It also has the oddity, I noticed at the same time, that upper case gives black pieces and lower case white ones, even thouugh the image list gives the more usual reverse. This should be fixed one way or the other, to make it consistent and make usres sure that it will noit be changed again after a diagram is posted.
I notice that some piece sets in the Diagram Designer, such as Alfaerie - Many, have pieces identified by multiple letters. How are these deployed in a diagram? I tried putting the entire sequence in parentheses, which is how I am used to other systems doing it, but this did not work and instead treated the parentheses as "unrecognised piece" characters. I could not see the correct way described anywhere on the page.
Well I have no trouble editing Hourglass Hex Chess, which is a bigger file than Nearlydouble. Unfortunately that appears to demolish both our theories, as Hourglass Hex has comments as well!
When my attempt to edit fails I get error 403. When it is successful a garbled version of the variant appears on the next page - but with the "view submission" link at the bottom so that I can continue to the page as edited.
What it does not fix is my inability to edit certain pages. One that I did succeed with was Flyover Xiang Qi, on which I moved some text from the Notes to the introduction to move the array diagram lower down. I notice that it did not have any comments, whereas the pages that I have failed to edit do have some, so it looks like comments might be part of the issue.
You're quite right! I had to change the JavaScript setting for an entirely different reason and it has the side-effect of making the diagrams visible. Oh, well, that panic at least is over.
I am still having trouble posting updates, including to some variants as well as piece articles. The issue of updating variant pages has become more urgent as I have noticed that I cannot see ffen diagrams from the computer on which I am currently going online and was hoping to replace them. Even had there been no problem updating, it would take some time to replace them all, and I would be interested to know if this problem is due to my change of computer or the move of website, If it is the latter, then for all the shortcomings of these diagrams this needs fixing in the interim before I (a) am able to edit all my pages properly again and (b) I have the time to carry them all out.
For the record, in case an editor has time to make - and better success updating - the changes, the intention was to replace the three ffen diagrams on the Nearlydouble Chess page with the following virtual images respectively:
That's true now that I look at it more closely to check that I had not missed anything. At first I thought that (expressed in terms of just two of the players) 1PN3 NxP 2RPxN RxR would cover it, making the danger to left Rooks ultimately dangerous for right Rooks, but I had overlooked 2 BPxN, which does indeed leave the left Rook unable to escape. How about the following?
Both Castlings would take place the same side of the King, depending whether the inner Rook was still there.
Both Castlings would take place the same side of the King, depending whether the inner Rook was still there.
The combination of this thread and Undenary Chess makes me realise the possibility of the Chaturaji and Nearlydouble principles cancelling out to produce the following 4-player array retaining the FIDE pattern of each army's relative starting positions:
The layout of my Flyover Xiang Qi can be extended to 4-player versions of other nine-file-by-ten-rank variants, but Chaturaji suggests a very different kind of variant, with one aside of most pieces arranged on a square board. It is an idea that I myself applied to Shogi , but I declined to try it to variants where the two-player game itself uses a boards with differing numbers of ranks and files. If anyone has applied the Chaturaji principle to such games I would be equally interested to know.
I have realised since my last posting on this thread that Anglis Qi Modified has already gone, and I have decided to get rid of the following variants next: Pawn the Brain, Sextuple Besiege, VeCoTha. Gateway Chess I will retain as a page, but plan to change into a 3d, but nevertheless more Chess-like, variant.
Yesterday I got to the computer from which I had made successful updates, but unfortunately I could not try it out because the mouse was not working properly and there was not another one easily accessible that could be substituted. I still cannot post the updates that I want to from th computer that I am currently on.
I have just noticed that old-style pages are displaying comments, so it might be that mechanism that is causing PYOs to display them twice. If anything that suggests that it less likely to be connected with the issues in posting updates, but it still makes for a lot extra being displayed on the page.
I am adding <a href="../index/msdisplay.php?itemid=MSvecotha">VeCoTha</a> to the list of variants which I am considering discontinuing. It got hostile reviews from the start, and it would give me an excuse to eliminate the Halter from Man and Beast (assuming that I can edit the relevant pages). I do not recall anyone else taking up this method for limiting Nightrider and Rose strength.
<p>
The final straw was when the tabulation that I originally used to display the array diagram had stopped working properly by the time that I reviewed it. I did not fancy writing out 113 images required for a ffen-lookalike diagram, and so decided to resort to a 2kb single image. My attempt to post this failed so I had to resort to Ascii Art, and it then occurred to me that if the best way to represent a variant was by an actual single image it might be a sign that the array is too complex. If ffen could display Rivers, between files as well as between ranks, it might have been a different matter. If a virtual single image generator could display them it would be even better but it seems that the first such system cannot.
<p>
So my list of up-for-deletion variants currently stands at:
<br><a href='../index/msdisplay.php?itemid=MSanglisqimodifi'>Anglis Qi modified to add Cannons and Arrows</a>, which makes for quite a cramped 8x8 board.
<br><a href='../index/msdisplay.php?itemid=MSemperorsnobili'>Emperor's Nobility 3d Latrunculi</a>, a not very Chess-like 3d variant with a complex chain of promotions.
<br><a href='../index/msdisplay.php?itemid=MSgatewaychess'>Gateway Chess</a>, another one-off microregional like many of those that have already gone and with awkward not-very-Chess-like extra rules.
<br><a href='../index/msdisplay.php?itemid=MSpartnershipmit'>Partnership Mitregi</a>, an 8x8 promotion-free Shogi variant.
<br><a href='../index/msdisplay.php?itemid=MSpawnthebrain'>Pawn the Brain</a>, originally a showcase for divergent-piece names that I have replaced with a simple prefix for the original Take the Brain pieces.
<br><a href='../index/msdisplay.php?itemid=MSsextuplebesieg'>Sextuple Besiege Wellisch</a>, a hex version of my Quadruple Besiege variants and hard to illustrate.
<br><a href='../index/msdisplay.php?itemid=MSsquareversushe'>Square Versus Hex</a> and <a href='../index/msdisplay.php?itemid=MSxiangcata'>Xiangcata</a>, two variants whose mixture of geometries forces my Man and Beast to qualifying some pieces "only avaliable in 3d" with "usually".
<br><a href="../index/msdisplay.php?itemid=MSvecotha">VeCoTha</a>, for the reasons given above.
What I have just noticed is that the comments are appearing twice at the foot of the page. I do not know whether this is significant, but it is an anomaly and detracts from the page's appearance.
This seems the logical thread to continue the discussion started <a href="http://www.chessvariants.com/index/listcomments.php?subjectid=422706f7808776e0">here</a> as it exactly describes what I have encountered. I have not yet had an opportunity to try updating on a different computer, but I hope to do so soon.
The computer on which I am getting a problem has Internet Explorer 11, and all the ones from which I posted successfully had older versions of Internet Explorer. It is nothing to do with graphics, as the pages on which I identified the problem are ones without any images on them.
I had a word with someone who knows about the set-up on the computer that I'm using, and his first thought was that it was because the broadband was too slow - it's a temporary connection while the faster one is being fixed. However, more than a year ago I had no problem uploading even bigger amounts using dialup. Can anything have changed over 2015 that would make it harder to upload for a given capacity - for example, might a particularly dramatic increase in internet traffic be the issue?
<p>
One new feature that I noticed was a spellcheck. Could this be affecting things, particularly on pages with a lot of unrecognised words in them?
So far, so good. I have devised the array diagram<br><IMG SRC="http://play.chessvariants.com/pbm/drawdiagram.php?code=1sruut-----2sxumj----3siljj---4ssixr--6ssss-11-SSSS6--RXISS4---JJLIS3----JMUXS2-----TUURS1&cols=11&font=Optima&set=small&shape=vhex&bcolor=FFFFF0&board=201.012.120.&colors=olive+darkkhaki+darkolivegreen&tcolor=000000"><br>for the Mishling form of <a href="../index/msdisplay.php?itemid=MS4linepiecehexc">4 Linepiece Hexes</a> (and similar for variants with subsets of the pieces). For other versions I can use C for the Waitress, D for the Shammes, and E and V for their Contra- pieces.
<p>
Unfortunately I am not sure that an attempt to update the page would work well as I have started encountering a 403 error when I have entered my changes and click to get to the next page of the update process. I am posting from a dfferent computer, but I cannot see any difference in the settings. It seems to happen only when the box edited has a large amount of text, as I have managed to post with less text, even from the new computer. Has anyone any idea why this might be happening? Has some limit been reached that has come in recently, or that includes past edits, or even comments? If one of the latter two, is there any way that editors could "cleanse" it of any past material that might be affecting it?
It is good to see a virtual single image generator up and running <a href="http://play.chessvariants.com/pbm/diagram-designer.php">here</a>, but I have a few questions.
<p>
Firstly, is there any way of arranging cells other than in a parallelogram? Some square-cell variants have multiple concavities, generally arranged symmetrically, and most hex variants are arranged on boards that are themselves hexagonals. Can it be domne with a symbol for an absence of cell, perhaps?
<p>
Still on the subject of hex variants in particular, when I select a hex geometry I get cells of just two colours rather than the three that I would expect - despite there being three colours specified by default.
<p>
This moves me on to the colours of pieces themselves. Can <i>they</i> be displayed in more than two colouyrs, to represent lrager numbers of players?
<p>
Finally, can Rivers be shown, and if so are they shown only between two ranks or can more complex arrangements be shown?
Sorry, that link went wrong/ It should be Cyclohex.
Single-images arrays have always been a very last resort for me, as they are so inefficient. The advanmtage of a separate image for each cell is that just 28 distinct images can be used to illustrate any possible FIDE game from end to end. A stored single image of the same resolution requires the equivalent of more than twice that for just the array position, The Cyclohex array image takes up 104 kilobytes, far larger than the text of any variant, and I would hate to illustrate a whole game of that. Another problem with single-image array files is that the mechanism for loading them does not always work. It did not, for example, when I tried to post such a file for VeCoTha in place of tabulated multiple images, and instead I had to resort to an emergency Ascii Art diagram. In case anyone asks, there is never any error warning, it just shows as an invalid image on-screen after I edit the page to incorporate the image.
Of course doing multi-image diagrams raw has its costs as well, which is why I emphasise stored. That is where ffen diagrams come in, as regards traditional computing. As far as the HTML document is concerned it is a calculated image, taking up not much more space in the text of the page than a single image but without an image file's use of memory outside the text of the page either. As far as a large enough screen is concerned they also behave like a single image, at least if nothing else is on the same line, and desktop monitors have been getting bigger all century. The problem is how they behave on the mobile devices to which this thread's title refers - and I am guessing that Ascii Art is not very mobile-friendly either.
Now there are various facilities for playing many of the variants on these pages, and I would be surprised (and indeed disappointed at the capabilities of programming) if they generated an image file for every position that ever came up. In terms of more typical programming, It would be the equivalent of every control on a Visual Studio form replicating the entire definition of tha control. What I do not know is how the playing facilities stand up to use on a tiny screen. If they can be made to behave as single images even on-screen, surely something can be done somewhere along the line that can be done to make ffen diagrams do so as well. It would be a shame to lose such a useful shorthand for want of it functioning on mobile devices.
I have reached the point where I particularly wish to decide what to do with Anglis Qi modified to add Cannons and Arrows, and therefore seek the guidance of other contributors. I could de-tabulate the page leaving the variants as they are, drop some variants and keep the rest at the same time, or just discontinue the entire page. Enlarging the board would encroach too much on other contributors' variants. As it might be difficult to judge before the page has been edited, I include the arrays below in alphabetic form, where A=Arrow and C=Cannon:
Anglis Taijitu rnbqkbnr ac----ca -------- -------- -------- -------- AC----CA RNBQKBNR Yanglis rnbqkbnr pcpaapcp -p-pp-p- -------- -------- -P-PP-P- PCPAAPCP RNBQKBNR Anglis Eurasian rc-aa-cr -nbqkbn- pppppppp -------- -------- PPPPPPPP -NBQKBN- RC-AA-CR Anglis Hindcannon -c-aa-c- rnbqkbnr pppppppp -------- -------- PPPPPPPP RNBQKBNR -C-AA-C- Anglis Midcannon rnbqkbnr c--aa--c pppppppp -------- -------- PPPPPPPP C--AA--C RNBQKBNR Anglis Forecannon rnbqkbnr pppppppp -c-aa-c- -------- -------- -C-AA-C- PPPPPPPP RNBQKBNR
I eventually decided to get rid of tables altogether and just show diagrams above and below each other and text. This not only (hopefully) makes the pages better suited to mobile devices, it addresses display issues that I found on traditional computers, and it reduces total memory used, whereas a more complex solution would increase it. The variants that I have carried it out on so far are:
12 Sharp Chess;
16 Seasons;
2 Jewels;
2 Level Guru Mahachaturaji;
3 Level 4 Player variants;
3 Player Honeycomb;
4 Faces;
4 Linepiece Fusion;
Armies of Faith 1/2/3/4/5/6;
BacCanCat;
Brookschach;
Chaturanga with minor changes;
Commedia dell'Arte Chess;
Compact Hex;
Cornucopia;
Courier Leapale;
Crossover-piece Dual-dircetion variants;
Crouching Stepper... ;
Diamond Ring Chess;
Double Cross Besiege;
Fimbriated Chess;
Flyover Xiang Qi/Shogi;
Gutenschach;
Half Nearlydouble;
2 and 3 dimensional Herichess;
Hourglass Hex Chess;
Irwell;
Larger Wildeurasian Variants;
Lengthleaper Chess;
Mini Fivequarters;
Mitred Framing 1/2/3;
Nested Chess/Xiang Qi/Shogi;
Notchess;
OctHex 146;
Pass Variants;
Proto Prelates;
SerPent;
Shoxiang 108;
Small Game Nearlydoubles;
Stock Goes East 49 Files;
Taijitu Qi;
Tardis Taijitu;
Tee Garden Shogi;
Tetrahedral Shogi;
Turn Qi;
Twin-board Ecumenical Chess;
Weltschach;
Westfield Chess;
Yo[n]o Shogi;
Yoto.
12 Sharp Chess;
16 Seasons;
2 Jewels;
2 Level Guru Mahachaturaji;
3 Level 4 Player variants;
3 Player Honeycomb;
4 Faces;
4 Linepiece Fusion;
Armies of Faith 1/2/3/4/5/6;
BacCanCat;
Brookschach;
Chaturanga with minor changes;
Commedia dell'Arte Chess;
Compact Hex;
Cornucopia;
Courier Leapale;
Crossover-piece Dual-dircetion variants;
Crouching Stepper... ;
Diamond Ring Chess;
Double Cross Besiege;
Fimbriated Chess;
Flyover Xiang Qi/Shogi;
Gutenschach;
Half Nearlydouble;
2 and 3 dimensional Herichess;
Hourglass Hex Chess;
Irwell;
Larger Wildeurasian Variants;
Lengthleaper Chess;
Mini Fivequarters;
Mitred Framing 1/2/3;
Nested Chess/Xiang Qi/Shogi;
Notchess;
OctHex 146;
Pass Variants;
Proto Prelates;
SerPent;
Shoxiang 108;
Small Game Nearlydoubles;
Stock Goes East 49 Files;
Taijitu Qi;
Tardis Taijitu;
Tee Garden Shogi;
Tetrahedral Shogi;
Turn Qi;
Twin-board Ecumenical Chess;
Weltschach;
Westfield Chess;
Yo[n]o Shogi;
Yoto.
Any of you who have seen these in the old form and dismissed them as badly designed might wish to revisit them now that the tabulation issue has been ironed out.
Sorry, that should have been 10x10, not 100x10. The latter size would be too large even for me - including in the 3d sense of a cube of side 10.
25 comments displayed
Permalink to the exact comments currently displayed.