Using the static binary on Case's page, the encoder adds a TON more bass tones to any sample I feed to it. Anyone know what's up here?

I assume that it's rather a play back related problem than something with the encoder -> It's the same Encoder for Windows and Linux. Did you test the encoded files on Windows and GNU/Linux?

I jumped the gun.....it is the xmms input plugin. When played through mppdec it sounds like it should. Anyone know of anything to try with the plugin? Xmms version 1.2.7 - xmms-musepack-0.98.so (On Frank's page).

I use the same plugin but everything sounds as it should.

I also use that xmms plugin and I didn't notice anything weird about the sound. I had also a few other (older) versions and they all played fine. IIRC the major difference in v0.98 is about tags, not the sound anyway.

Which sound architecture is everyone using on Linux....I am on ALSA and I can say without a doubt the decoder is doing something it shouldn't be. I used the binary, I compiled my own, same thing with both. Used the same flac with mp3, ogg, and faac, and be damned if they didn't sound like they should. The only thing I can try next is too decrease the gcc flags (I am on gentoo), if this doesn't do it I will be at a loss. If you have any ideas whatsoever please put them down I WILL try anything.

I'm using ALSA too.

At first redhat set arts, but then I chose oss. Can't remember why exactly, but it was definately not about sound quality - it was fine with both oss and arts.