I've released v1.1 which includes more options:
Code:
Description:
Converts a BMP image to convfont format
Usage:
BmpToConvfnt <imagefile> [options]
Arguments:
<imagefile> Path to the BMP image file
Options:
-w, --width <w> (REQUIRED) Width of each glyph in pixels
-h, --height <h> (REQUIRED) Height of each glyph in pixels
-s, --startcodepoint <s> Starting code point (e.g., 32 for space, 65 for 'A'; defaults to 0)
-n, --numglyphs <n> Number of glyphs to read from the image (optional, defaults to all)
--xpad <xpad> Left padding, in pixels, before the glyph grid starts [default: 0]
--ypad <ypad> Top padding, in pixels, before the glyph grid starts [default: 0]
--xcellpad <xcellpad> Horizontal spacing, in pixels, between glyph cells [default: 0]
--ycellpad <ycellpad> Vertical spacing, in pixels, between glyph cells [default: 0]
-o, --output <o> Output file path (optional, defaults to input filename with .txt extension)
-c, --charsfile <c> Path to a text file containing characters to map to glyphs (optional)
-help, /? Show help and usage information
--version Show version information
See the readme for more details on how to use it.
Notably, this version allows you to point to a "chars file", which correlates codepoints for you, making it easier to use an image file laid out in a non-sequential manner.
Download:
https://github.com/merthsoft/BmpToConvfnt/releases/tag/v1.1
There are single-executable files built for windows, mac, and linux, or you can use the .net 10 version which isn't combined into a single executable (it's a smaller download because it doesn't include the framework shtuff needed to make it work, should work on any OS with .net 10 installed, so windows, linux, and mac are all supported). Up to you, really.
I'm considering making a web tool as well that can take in the same chars file, as well as adding support to output actual .FNT files for ease of use in other situations as well. 
