Ok, so this time, I just made one very large collection gig font. This should make it easier to distribute, and use, and I think the compression benefits from it (less headers, more data to run through? I dunno.) It has all the soundfonts I could convert. A couple gave the error "no sound files" or some such, so I just skipped them (Millo_MultiLayered2 & some other, don't remember)
Also, a few of these crash linuxsampler. they are probably bad files, either due to my python script, or to awave studio. I'd love to get some testing going on this, and see if I can fix it. I don't have the time to do it by myself, right now.
I would really like to be converting sf2 to gig, purely in python, or even creating gigs directly from hydrogen xml files, so if anyone knows of a lib that I can do this (make gig files), I would much appreciate it. Even a linux-command-line program, written in anything, would be a great improvement.
I'd also like to improve the SF2 creation part. Since I made the script, I think I have a found a few libs that seem a lot more feature-rich for this purpose, and I've learned some good tricks that might make it better. Again, when I have more time, I would be willing to port the hydro2sf2 to something leaner, and better. Maybe in that version I could include using a more python way of converting the samples (currently uses linux command-line flac and sox.)