Hi, I have finally come to a state in the development of the new audio filter layer where I can show my work. The patch is attached to this mail. I still have some problems with the buffering. If you want to help me have a look at line 179 in dec_audio.c. Sometimes I have to throw away data in order to avoid buffer over flow in buf this only happens when forward/backward but is still annoying and makes the aflayer unsuitable to use with mencoder. The new and fun stuff is: 1. Resampling from any to any sample rate. 2. Auto configuration 3. Changing the number of channels to any number of channels. Anyway it as you can see the patch is huge. Have a look and give me feedback. I hope to apply it to CVS soon. Cheers, //Anders