Check out Modern Chess, our featured variant for January, 2025.


[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Single Comment

[Subject Thread] [Add Response]
Rich Hutnik wrote on Sat, Oct 25, 2008 06:55 PM UTC:
I believe Mr. Muller meant to add this comment to this thread here:


'Game Courier got comparable results when enlarging, but it got better
results when reducing the size. Compare:'

Yes, indeed. The scaling routine seems to use some form of anti-aliasing.
I could achieve this in WinBoard by blowing up the bitmap by a factor 2 or 3, smoothen it, and then reduce to the desired size by sampling. But it seems a bit silly to do those first two step in WinBoard, every time you change the board size of piece set, as they are always the same. So it would be better to do them once and for all, and have WinBoard read in the smoothened large bitmaps, and perform the demagnifying StretchBlt on them to generate the desired piece size.

I must say I am not very pleased with either result. The anti-aliased
symbols look really ugly. Some black lines have a light-gray line running along them, which would have looked much better if it was background color. I think the true-type-font rendering is much more suitable for scaling pictures with sharp high-contrast edges and narrow lines, such as we are dealing with here.

I guess it would not be very difficult to create your Shogi pieces
through font-based rendering. The only thing that is needed is a more elaborate procedure to color-fill them. You could isolate the naked symbols printed on the 'chips' by flood-filling a number of times from the outside upto a specified color, to peel off the outer layers of the symbol until you reach the part you want to color, and then overprint with the symbol again to restore the outer boundary, and finally floodfill the outside to make it transparent.