Weird question about MP3s and Linux

[elfs]

We all know that MP3 encoders have quality controls: that when you encode an MP3, you choose how much of the original to keep, how much to throw away, and how much to "model," that is, how much can be restored to the outgoing audio stream by inserting modeled sounds into the psychoacoustic profile of the stored MP3

Comments

[tamino]

I don't think this is the case. My impression is there are very strict rules about what constitutes a correct decoding of any given mp3.

Any way you can turn off real time priority for a comparison?

[shockwave77598]

I think you are simply listening more acutely now that you've made the change. The data has not changed. The processing has not changed. You are simply paying closer attention is all.

The differences you site occur at encoding, not decoding. During decode, the algorithm is the same regardless. Each block says how many bits and what bitrate the data is in and the quality settings; the decoder then renders a new block of wave data and feeds it to the sound card.

Unless the decoder is for some reason not decoding properly due to the priority (perhaps it is being prempted) I can think of no explanation for this except you are simply listening more keenly. Set the priority back and listen just as keenly to create a control for your test.

[atheorist]

It might be related to scheduling jitter - though doing a blind/randomized test to see whether you can tell one configuration from the other would be a neat idea.

[wookie_cd]

Back when my desktop was a pentium, I think its cpu was clocked at about 90MHz. It fell just short of 44khz stereo playback, however winamp could be configured to drop its CPU consumption considerably by limiting to mono or a lower khz output.

Today, the in_mp3 input module in my winamp 2.91 still has options to disallow stereo or downsample to half or quarter output.