Version: 1.0 (using KDE KDE 3.95.0)
Installed from: Debian testing/unstable Packages
As differnet boards have differnet levels of difficulty, it would be nice to have the score based on a 'par' value, the minimum
number of clicks needed to solve the board.
Also, considering bugs #141195 and #150386, how about a bonus for using less cabling than the computers solution.
(and for key clicks, how about a third click type for flipping a tile 180 degrees)
I agree with you. My plan for kde 4.1 is to change the scoring to take extra moves (how many exceeding the minimum) and time into account.
About using less cabling than neccesary I don't like the proposed solution very much. Maybe I will try to only generate levels for wich all cables have to be used to connect all terminals.
And I was planning for one move to mean moving one cell, no matter how many degrees. This also helps for example when accidentaly pressing the wrong mouse button.
Thanks for the report.
Fela Winkelmolen wrote:
[bugs.kde.org quoted mail]
Actually, I like having to be precise about which key I use to rotate
the cells. Perhaps there could be a user skill setting for counting keys or
What about undoing clicks when it is obvious the user accidentally clicked the wrong button (like left_click/*woops* right_click/right_click counting as one click), this would still penalize people always using the same direction.
It don't really seems something worth making an option.
SVN commit 785910 by fela:
* New scoring algorithm (now the score is based on the elapsed time as well as the number of moves)
* The cells now are always of the same size (it wasn't always the case, before).
M +4 -1 TODO
M +10 -2 src/abstractgrid.cpp
M +4 -1 src/abstractgrid.h
M +58 -51 src/mainwindow.cpp
M +19 -9 src/mainwindow.h
M +2 -1 src/renderer.cpp
M +5 -1 src/view.h
WebSVN link: http://websvn.kde.org/?view=rev&revision=785910