working around MP3 patents, what features do specific patents refer to Options

[Prebsi]

I'm trying to get a handle on some objective measure of "really, really bad". I assume those are at the same bitrate. How much would you have to increase the bitrate on the hacked version for it to match (visually and/or audibly) the patented version? Is it even possible? How does that bitrate increase generalize to other types of audio such as speech or popular music?

Both examples above were compressed to 128 kbps. Increasing this to 320 did not create any visual or audibly difference.

The reason that the butterfly process can not be left out from the encoding is that the reverse butterfly process is performed in the decoding. I made a test where I removed the butterfly on encooding+decoding and ended up with a near-perfect result. No visually or audibly difference on the swept sine.

I think the reason for adding the butterfly process in the first place, is to handle the situations where one band is included but the next band is left out due to the perceptual filtering. I have not been able to create a synthetic signal where this is visible/audible though, expect at 32 kpbs where I did get some fuzziness on the graphics.