Home Digital Drive

Upsamplers, DACs, jitter, shakes and analogue withdrawals, this is it.

RE: Null Test Difference - test to measure real fidelity of a device

To time align is simple: just move to the first transient (select one of adequate width), then delete the part of the samples before the peak. I always use the same song track as I know by heart the point at which to operate the alignment.

Finally, make acquisitions micrometrically of the same length as the original file.

I got the same doubts too, but with due reasons. In fact, before starting to carry out such measures, I have tested various procedures for about 6 months.

The interpolation of some editing software works very well. I tried to cut two acquisitions of the same piece of music, as well as change its width several times, in order to implement as mnay interpolations as possible. Well...in the end from these files (when put in antiphase) it came out only thermal noise (-130 dB). This is a sign that interpolation does not introduce phase and/or amplitude errors. Indeed, a simple deviation of 0.1 dB at a single frequency makes a peak at -90 dB appear. (Something you can try by introducing equalization on one of the two files) Same thing happens in the time domain within the deviation of a single sample which for the 44.1 kHz corresponds to 22.67 microseconds.



What it can be seen from the above null test is:
two heavily processed signals in order to have high interpolation, then one of the files receives microequalization with an amplitude of 0.1 dB and 0.2 dB at 1 kHz and 2.5 kHz.

Red Line 0.1 dB @ 1 and 2.5 khz
Green Line 0.2 dB @ 1 and 2.5 kHz
Blue line tests is the null result after interpolation on the two files with editing software.

You can see perfect zero signal down to noise floor (blue line). Furthermore, compared to a deviation of 0.1 dB, the level rises of 6 dB. Therfore there is a further amplified phenomenon which on one side it makes it very sensitive, on the other it becomes very visible for the necessary assessments.

Obviously this was just a test because, when you go to the real acquisition, it is used a calibration within a maximum range of 0.01 db, without any interpolation.

There are many things to consider. Six months of testing can not be told in a few posts...and I have dome many tests before...



This post is made possible by the generous support of people like you and our sponsors:
  Amplified Parts  


Follow Ups Full Thread
Follow Ups

FAQ

Post a Message!

Forgot Password?
Moniker (Username):
Password (Optional):
  Remember my Moniker & Password  (What's this?)    Eat Me
E-Mail (Optional):
Subject:
Message:   (Posts are subject to Content Rules)
Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:  Automagically notify you when someone responds.