RF & Audio

[jkeny]

Sorry, Marce but if you thought a bit about recovery of the clock you might have a eureka moment - forget about "bits is bits" - deal with the temporal issues. See the Hawksford paper I already referenced pg 5 if you want to find out more!

 

[wakibaki]

This paper is so flawed it's laughable that Waki would even cite it - he being a high speed digital designer & everything.

Evidently you find the content of the paper and my arguments persuasive, otherwise why would you find it necessary to resort to this kind of ad hominem approach? It's a sure sign to anybody reading this that you feel you're loosing the argument.

w

 

[jkeny]

Waki, I just can't take you seriously & I tend to want to prick pomposity & posturing when I come across it. You really do make some extraordinary claims about yourself - DMM measurement of characteristic impedance 🙂, high speed digital designer🙂r, citing a paper that is laughable 🙂. Sorry, but it's just my way.

If, however, you wish to defend the paper you cited, I'm all ears.

 

[wakibaki]

Keep digging.

w

 

[marce]

A bit more reading material, I have been looking for for info regarding some of the stuff we are discussion that is later than the early 90's.

http://www.wolfsonmicro.com/documents/uploads/misc/en/Specifying_Jitter_Performance.pdf

http://www.wolfsonmicro.com/documen...e_of_spdif_digital_interface_transceivers.pdf

Jkeny, cant find the reference on Pg 5!

 

[jkeny]

Thanks Marce,
I'll have a read of those papers when I get a chance.
Look for the section in the Hawksford paper section 3 Jitter in the Digital Interface/3.1 Interface Bandwith Limitation

Here's how JosephK expresses it (again I hope he doesn't mind me quoting him?):

This clock ideally should not be influenced by the actual data transferred, but because of imperfections, it gets modulated, in a correlated way to the original audio data.
This correlation was inspected by Hawksford in his article, and was found that the recovered clock will be modulated such that the original audio signal will show up in the phase noise spectra of that recovered clock, though strongly distorted.
Also this distortion had been shown, and had been found consisting of a series of harmonics, from the third upward.

 

[Telstar]

I'd like to point out that any spdif receiver has a minimum jitter of 50ps, which makes it flawed to start with.
Either integrated cdp (with signal taken from the transport via i2s) or from proprietary means (read fiber optics such as used in the TEAC P70/D70 combo) or from a soundcard also via i2s.

USB transceivers are also at least in the 50ps range.

Now, can we get back to the RFI topic? 🙂

 

[jkeny]

I'd like to point out that any spdif receiver has a minimum jitter of 50ps, which makes it flawed to start with.

This is not a limitation of SPDIF, just the existing lower limit of current SPDIF receivers from Wolfson

Either integrated cdp (with signal taken from the transport via i2s) or from proprietary means (read fiber optics such as used in the TEAC P70/D70 combo) or from a soundcard also via i2s.

Agreed SPDIF should be avoided if at all possible

USB transceivers are also at least in the 50ps range.

Do you have a ref?

Now, can we get back to the RFI topic? 🙂

Sure, this should be hived off to another thread but it does impinge on RF considerations too as there is an idea that sonic improvements noted with lower jitter may be down to RFI reduction to some extent!