On Mon, 07 Jul 2003 08:15:21 +0100, Jim Lesurf
wrote:

In article , Chris Isbell
wrote:
[snip]

[snip]

Not sure what you mean above. (Mind you, I'm not sure what their ad is
supposed to mean, either! :- ) The wideband SNR is usually poorer than
the narrowband. e.g. 32k point FFTs where a 90dB wideband SNR converts to
around 120dB per frequency bin. (Forget the precise value.)

What I think this means is that the measuring equipment needs to be
this good to make the measurements. (Hence the narrowband figure is
'poorer' than the wideband one.)


Anyway, what about the theory?

Assuming a resistance of the order of 0.5 ohms, a temperature of 25
degrees C and a pass band of 100kHz.

Let's read two or three references, like "Thermal agitation of
electricity in conductors", J. B. Johnson, Physical Review, vol. 32, pp.
97-109 and "Thermal agitation of electric charge in conductors", H.
Nyquist, Physical Review, vol. 32, pp. 110-113, both of them from 1928.

Let's take the accepted formulas and calculate the intrinsic noise level
of the said cables.

Is this what they mean?

I don't know. My view of their site is that the technical information
is better at obscuration than enlightenment and it likely to be either
suspect or misleading.

Translating from another post in the same thread of
fr.rec.son-image.hifi:

François Yves Le Gal: I know of nothing that has an effective
resolution exceeding 24-bits, i.e. roughly 144dB wideband. It is very
difficult [approximate translation!] to go below -120dB, except by
fanatical isolation of everything.

Nicolas Le Goff: I calculate 29nV RMS of Nyquist noise under these
conditions which at a typical consumer level of 0.707V RMS gives a
signal to noise ratio of 148dB.

FYG: Correct.


Original message:

Je ne connais rien qui dépasse les 24 bits de résolution
effective, soit 144 dB en bande large à la louche. Et on a tout plein de
merdes à partir de - 120 dB, sauf à tout isoler comme un fou.

Je trouve 29nVrms de bruit de Nyquist dans ces conditions
soit au niveau de tension consumer trypique de 0.707 Vrms
un rapport S/N de 148dB

Vala.



--
Chris Isbell
Southampton
UK