You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The .psf file seems to be twice as wide as it actually should be. The PSF header also reports cozette.psf to be of size 12x13 when it should be 6x13.
The glyphs look like this (I wrote a quick program that extracts the glyphs and prints them out in ASCII):
You can also easily confirm this by opening cozette.psf in a hex editor and looking at bytes 28-32, which is the width of each glyph as an unsigned 32-bit integer.
The problem seems to be that the input file gets misinterpreted somehow. This command reproduces the error: java -jar BitsNPicas.jar convertbitmap -f psf2 -o BAD.psf cozette.bdf
Same thing happens with the .sfd file. I couldn't test .otf because BitsNPicas doesn't seem to accept it. The .dfont file just straight up crashes.
However, if you use the .fnt file instead, it's at least a little better, with the font now being reported at 7x14: java -jar BitsNPicas.jar convertbitmap -f psf2 -o BETTER.psf cozette.fnt
If I understand correctly (from briefly looking at build.py), the .psf file is built from the .bdf file through fontforge. I unfortunately can't get fontforge to work at all, so I can't test anything related to that. But my guess is that fontforge is having the same problem as BitsNPicas.
The text was updated successfully, but these errors were encountered:
I generated bdf files from the sdf file using fontforge and still got the wrong widths. If I knew more about the file format, I'd try to fix the issue by hand, but alas I have no idea how any of these file formats work.
The .psf file seems to be twice as wide as it actually should be. The PSF header also reports
cozette.psf
to be of size 12x13 when it should be 6x13.The glyphs look like this (I wrote a quick program that extracts the glyphs and prints them out in ASCII):
You can also easily confirm this by opening
cozette.psf
in a hex editor and looking at bytes 28-32, which is the width of each glyph as an unsigned 32-bit integer.The problem seems to be that the input file gets misinterpreted somehow. This command reproduces the error:
java -jar BitsNPicas.jar convertbitmap -f psf2 -o BAD.psf cozette.bdf
Same thing happens with the .sfd file. I couldn't test .otf because BitsNPicas doesn't seem to accept it. The .dfont file just straight up crashes.
However, if you use the .fnt file instead, it's at least a little better, with the font now being reported at 7x14:
java -jar BitsNPicas.jar convertbitmap -f psf2 -o BETTER.psf cozette.fnt
If I understand correctly (from briefly looking at
build.py
), the .psf file is built from the .bdf file through fontforge. I unfortunately can't get fontforge to work at all, so I can't test anything related to that. But my guess is that fontforge is having the same problem as BitsNPicas.The text was updated successfully, but these errors were encountered: