Comments/Ratings for a Single Item
💡📝H. G. Muller wrote on Sun, May 22, 2022 06:08 PM UTC in reply to Aurelian Florea from 03:13 PM:Because you press the Piece1 or Piece2 buttons only once?
Yes! That was it!
I see another problem. The computer moves twice!
💡📝H. G. Muller wrote on Sun, May 22, 2022 07:38 PM UTC in reply to Aurelian Florea from 07:19 PM:Good catch. This problem surfaced everywhere after a change in the general Diagram script. I did fix it in most places, but apparently forgot to fix it here. It should work now.
(The reason was that this applet abuses the routine WeirdPromotion, which would be called for every move, to invoke the routine that fetches the best move from the end-game table and play it. But at some point the Diagram script started to call WeirdPromotion twice, also once to figure out whether it should allow the user to choose a piece. This to also support unexpected promotions that involve a choice. So the WeirdPromotion routine included in this applet should be more careful now, and only invoke the routine for playing a move when it has not already been done.)
💡📝H. G. Muller wrote on Mon, Oct 3, 2022 02:38 PM UTC in reply to H. G. Muller from Sun May 22 07:38 PM:The checkmating script is now also available in a version that can handle Pao-like hoppers and some bent riders (namely those that change direction after the first step). I have not updated the page on CVP yet to do this, becase the current interface on that page has no mechanism to specify such pieces. On my website there is a version where you can select pieces through buttons; there I also have buttons for the Cannon, Vao, W-then-B (Manticore), D-then-B (Osprey) and vW-then-B (Snake Tongue).
http://hgm.nubati.net/applets/mate31.html?files=8&ranks=8
The Vao can be very useful in combination with the W-then-B pieces, or even the Marquis (NW). Because these pieces can be used as a mount for activating the Vao in a way that cooperates extremely well with their own move.
Nice work, HG
Very insightful and your apps are really helpful and instructive. Thank you!
💡📝H. G. Muller wrote on Tue, Feb 6 05:23 PM UTC:I have now added checkmating paragraphs to the Piececlopedia pages for Ferz, Wazir, Knight, Dabbabah, Camel, Zebra and Phoenix, with links to a 3-vs-1 Applet configured for using these.
💡📝H. G. Muller wrote on Thu, Feb 8 03:29 PM UTC in reply to H. G. Muller from Tue Feb 6 05:23 PM:I am now using a standard sentence as Checkmating paragraph for most pieces, like
The XXX cannot inflict checkmate on a rectangular board with only assistance of its own King, and is thus a minor piece. Even with a pair of XXXs you cannot force checkmate on a bare King, but paired with another minor this is sometimes possible. Try it!
What exactly the role of the XXX is, is then explained on the Applet page, as well as what property the other piece would need. Obviously the second sentence would be different if a pair of XXX can checkmate, and in that case I would preferably use the pair as an example.
Only if there is something interesting to remark that is specific for the piece, the paragraph is longer.
Concerning weak pieces: it turns out Ferz + Wazir can force checkmate on 9x9, provided the Ferz is on the shade of the corners. (And not at all otherwise.) On 8x8 this end-game is generally a draw, but that is because the checkmate can only be performed in two of the corners, so that in general the distance to a deadly corner is much larger than on 9x9. On 6x6 the deadly corners are again close enough to shephard the bare King there when it has taken refuge in a safe corner.
💡📝H. G. Muller wrote on Fri, Feb 9 08:09 AM UTC in reply to H. G. Muller from Thu Feb 8 03:29 PM:I have now re-instated the buttons for selecting the second piece on the configurable 3-vs-1 Applet page. The fist piece will always the one specified by the URL, which is supposed to be the topic of the page. But the URL can specify the second piece through other=X in the query string, with a limited set of minors X. (And it will default to using a pair of identical pieces.) The user can then select another partner for the piece the page is about, from the repertoire offered by the buttons. (Which is pretty large.)
This pretty much finishes this configurable Applet page; I don't see any reason for letting the user change the board size. (Which can be configured through the URL, and would not be allowed to go above 10x10 anyway for capacity reasons.)
While adding checkmating info for most pieces in the Piececlopedia, I encountered some new mating patterns. (E.g. how an FAD can be used to checkmate in corners of the shade it is not on, or how lameness/sliding sometimes can allow mates that would be impossible with the corresponding unrestrained leaper. I could add these to the main 3-vs-1 Applet page, which already discusses the more common cases. But I wonder if it wouldn't be better to split up that page, to leave it purely for the 3-vs-1 Applet, and put the general theory in an independent article (to which it then refers). As it is the theory is already a bit hidden far down on the page, where people might never scroll.
The Applet page could use some enhancement anyway; the current one cannot do hoppers and bent (or ski) sliders. While the EGT code has already been upgraded to handle those. This is an interface issue, though. In the general 3-vs-1 Applet you must specify the pieces through the move-definition aid. I have no ideas on how this could be enhanced to allow specification of bent trajectories or hoppers. Perhaps there should be buttons in addition to the move-definition aid, similar to those on the configurable page, which could be used to select the most common hoppers and bent riders directly. (E.g. for the minors the Manticore, Osprey and Ski-Bishop moves.) Or perhaps better, to add such moves to the move set currently specified in the aid panel.
12 comments displayed
Permalink to the exact comments currently displayed.
Why trying to define pieces I always get capture only pieces?!