Re: A partial answer for norman

On Mon, 01 Oct 2007 19:05:57 -0000, Raving <raving.loonie@xxxxxxxxx>
wrote:

On Oct 1, 1:51 pm, r norman <r_s_norman@xxxxxxxxxxxx> wrote:
On Mon, 01 Oct 2007 17:19:50 -0000, Raving <raving.loo...@xxxxxxxxx>
wrote:



On Oct 1, 12:55 pm, dkomo <dkomo...@xxxxxxxxxxx> wrote:
In the last decade, it was discovered that adding noise to D/A
conversion phase.

Yes, this has improved things greatly.

What sticks in my craw is that every time that I have heard this
injection of noise explained by the engineers who incorporate it, the
justification for it's effectiveness is... "shrug".

EMPIRICAL EVIDENCE alone. It just sounds better. No explanation.

What you have provided here seems to provide an exception for what
norman delclares, as pasted in below:

I have searched my previous posts for naught, looking for some
reference there that is even tenuously related to being able to hear
the artifacts introduced by digital sampling at 44 kHz. Since I am
essentially totally deaf above 2 or 3 kHz (I can't hear the highest
notes on a piano) I can't feel your pain.

[End of norman paste in]

To answer the last point first, my comment about nothing in my posts
relating to this subject of audition refers specifically to the
history of one particular thread, where you raised the issue totally
out of the blue (although you have a history of obsessing on this
points).

Here is how Raving thinks. It is my POV. It is a single
categorization that describes the entire set of all Raving's "focal
awareness". ~~> http://tinyurl.com/2g4lg6

======================================
norman subdivides into disjoint subsets. That is your POV.

If I am correct in my observation, everything that norman says is of
the form:
~~~> "placing in a disjoint (sub)set".
======================================

Cordially,

Raving


To answer the first point last, the process of adding random noise to
a signal to be digitized is called "dithering". http://en.wikipedia.org/wiki/Dithering

The Wikipedia article mentions use of this technique in WW II, the
early to mid 40's but mentions the use of the word 'dither' only from
the early 60's. I first learned of the technique in the late 50's
when I worked at Hewlett-Packard's R&D center on digital circuit
design in the counter-time division I seem to recall that the word
'dithering' was in use at that time but I can't be sure. The
technique means that you can make a digital measurement more accurate
than the quantal step size (i.e. better than +/- 1 bit). In that
way, measuring time intervals with a digital counter and a 1 MHz
clock could produce accuracy better than 1 microsecond, even to 0.1
microsecond, giving an extra significant figure to the result. (At
the time I seem to recall they were using a 100 MHz clock to get up to
1 nsec resolution).

The fact that some engineers may be ignorant of half-century old
standard methodology and technique should not lead you to conclude
that there is no explanation for how it improves digitization.

I find it curious and very informative that you have many comments
about "how Norman's mind works" but absolutely none about the factual
content of the information I provide. But never mind, that is just
part of the way my mind works.

.

Relevant Pages

  • Re: A partial answer for norman
    ... it was discovered that adding noise to D/A ... The Wikipedia article mentions use of this technique in WW II, ... 'dithering' was in use at that time but I can't be sure. ... The fact that some engineers may be ignorant of half-century old ...
    (talk.origins)
  • Re: Advice - overnight track works
    ... Then of course there is noise and noise. ... You have to bear in mind that generally speaking the railways don't ...
    (uk.transport.london)
  • Re: inspiration
    ... interpret as meaning their unconscious mind, ... How does all this differ from the idea of a "muse"? ... I tend to write during times when it's very quiet, ... White noise is nearly better than quiet sometimes. ...
    (rec.arts.sf.composition)
  • Re: A partial answer for norman
    ... it was discovered that adding noise to D/A ... The Wikipedia article mentions use of this technique in WW II, ... 'dithering' was in use at that time but I can't be sure. ... microsecond, giving an extra significant figure to the result. ...
    (talk.origins)
  • Re: ANC with wrong reference signal
    ... There have been three diploma thesises on this particular algorithm (not ... notch, which will do best at removing a tonal signal. ... exact technique that you describe, so I'm not sure how it's supposed ... works best in white noise, ...
    (comp.dsp)