Home Propeller Head Plaza

Technical and scientific discussion of amps, cables and other topics.

Re: Goodness, gracious, and you're talking like an expert?

Alright, I buy the whole statistical treatment of dither and how NSD with the correct type of dither allows you to have a fixed mean and variance of error. The math don't lie.

I am still not 100% convinced that this still implies that with two separate sensors (ears) and two separate signals (stereo) where there is a correlation between these two signals that the sensors are quite sensitive to that even with a fixed mean psd from a statistical perspective that in a short enough period of measurement the error signal may still cause a change of perception.

This does not depend on the math but on the capabilities of the sensors and how the data from the sensors is processed. The only way to determine it would be via experiment so if you know of any that tackle this exact situation I would love to read it.

I am now conviced though that the differences we perceive between analog and digital are more to do with timing distortion (jitter) and filter distortions (and I'm sure in a lot of cases, substandard mastering)

This has made me thing more about analog and digital though. In essence our physical concept of analog is actually really digital. Continous signals are purely a mathematical construct. One of the major ideas in physics these days is that abosolutely everything is discrete (matter, energy have been known for a long time to be so, but now time and space are the latest things thought to be discrete... the works). Just the concept that energy is discrete means analog is actually digital with a large (but finite) bit depth and ADC is actually downsampling and DAC is actually upsampling which is why dither works for both ADC and downsampling since they are actually _literally_ the same thing.


This post is made possible by the generous support of people like you and our sponsors:
  The Cable Cooker  


Follow Ups Full Thread
Follow Ups


You can not post to an archived thread.