Talk:Noise shaping

Does anyone out there know of any example of noise shaping in digital video?203.214.75.127

MS/s?
What does MS/s mean (section "Noise shaping and 1 bit converters")? MegaSamples per second? The link points to http://en.wikipedia.org/wiki/Sampling_frequency, but there is no such unit explained. —Preceding unsigned comment added by 87.160.145.159 (talk) 21:08, 3 October 2007 (UTC)


 * Yes.


 * Atlant 22:42, 3 October 2007 (UTC)

"Dither effectively reduces quantization error ..."?
This article says that "Dither effectively reduces quantization error by adding noise prior to the quantization process." I'm no expect (still a student), but this doesn't seem accurate to me. From my understanding, all dithering does is add noise so that the quantization error be uncorrelated, and thus less offensive to human perception. If I'm correct then the article should be fixed. Whaa? (talk) 16:08, 27 February 2008 (UTC)

"Article filled with errors and misconceptions"
Noise shaping does not reduce quantization error, it just moves it to desired parts of the spectrum.

Dither does not reduce quantization error, it just decorrelates quantization error from the signal, transforming quantization distortion (correlated and dependent on the signal) into quantization noise (constant, independent of the signal) and thus totally linearizes the digital sampling process.

Noise shaping can be applied independently of dither int order to just spectrally reshape quantization distortion.

According to its Wikipedia article, POW-R is not noise-shaping, it's just colored dither.

I've added several dubious tags to the incorrect assertions.--80.24.18.223 (talk) 07:36, 27 May 2008 (UTC)

I agree with everything except ... what's the difference between shaping and coloring? Both refer to EQ, so I don't see how you can say that noise-shaping is not colored dither without further explanation. Granted, uncolored dither is not noise shaping, but you have to find some way to define "colored" that doesn't involve EQ in order to show evidence that colored dither is anything but noise shaping under a different name. BrianWilloughby (talk) 20:09, 2 December 2010 (UTC)

Outdated reference link
In the references, the link should probably be instead, but I seem to be unable to edit that part of the page.

(MT-001: Taking the Mystery out of the Infamous Formula, "SNR=6.02N + 1.76dB," and Why You Should Care) —Preceding unsigned comment added by Wmattis (talk • contribs) 21:40, 28 October 2008 (UTC)

conservation of power?
> The effect is that the quantization error is low-pass filtered by a 2-sample boxcar filter (also known as a simple moving average filter). As a result, compared to before, the quantization error has lower power at higher frequencies and higher power at lower frequencies.

There is a suggestion that filters inherently shuffle power around, when in reality they can attenuate or accentuate power at different bands independently. Intellec7 (talk) 15:59, 19 May 2022 (UTC)