View Single Post
Old 26th Feb 2021, 5:34 pm   #2089
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,901
Default Re: The Audiophoolery Thread.

Straight analogue audio over domestic distances isn't difficult to connect. If you start getting audible differences with non-extreme cables, then something somewhere is wrong and I'd be digging hard to find just what and to understand it. Only in this way can you sort out a problem and then know that it ought to stay sorted.

Things are different with digital connections. With serial data sent at a reasonable bit-rate to fit it all in along with checksums and framing, the bandwidth goes up dramatically and the very-short-compared-to-a-wavelength thing goes out the window. You need clean edges and a good opening on an eye diagram to get low error rates. This means lines must be terminated properly to avoid triple transit echoes and higher order bounces. You really need to look at the waveform delivered at the receiving end and look for rings and ISI.

Seemingly minor changes can move the timing of echoes and position spikes right at the decision time of your eye diagram and the error rate can jump dramatically.

Digital things with error correction can appear perfect right up to the point and then there is nothing graceful about their degradation.

Peter Scott on this forum and myself were involved for a while in measuring the error rates of data links for telecomms systems and their vulnerability to jitter and other impairments. I got involved in the vulnerability of digital radio systems to signal to noise and interference. These things are all quite understandable, but when something seems to go wrong, it needs investigating so that you can fix it with confidence that it wasn't just luck.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote