skamp
Aka: HeadFiend, BatFi
- Joined
- Oct 8, 2008
- Posts
- 1,334
- Likes
- 94
Any kind of interference with an effect on USB data transmission would affect bits (zeros and ones), indiscriminately.
I don't think I've ever heard errors in USB data transmission, but I suppose they would sound like glitches (like clicks and pops or scrambled audio), because of the randomness of the errors, thereby breaking the coherency of the data. They definitely wouldn't selectively choose which bits to change so as to keep the audio signal's coherency (i.e. no glitches) and alter some property of it (warmth, soundstage and whatnot). You would need some pretty smart "alteration" with lots of coherent (i.e. selective) changes for that to happen.
In other words: if you change a single bit in a packet of structured (meaningful) data, you break it. It turns into gibberish. It's not a slight or subtle alteration. It can turn the number 127 into 255 [*], double the original value. In contrast, if you superimpose a 60Hz hum on an analog signal, well, you still get the original music, with an added bass hum.
P.S. I just thought of an example of digital transmission errors that you might have had the occasion to witness: Over The Air digital TV. Every now and then, you get errors. What do you see and hear? Warmer colors? Constricted soundstage? Nope. What you see is blocks of pixels that look completely wrong. When the transmission is really bad, the image is completely garbled. Sound also gets affected, and it sounds like, well, glitches. Nothing subtle about it.
If that doesn't convince you, or at least makes you doubt your beliefs, I don't know what will.
[*] 127 (base 10) is 01111111 in binary (base 2). Change the 0 into 1 and you get 11111111, which is 255.
I don't think I've ever heard errors in USB data transmission, but I suppose they would sound like glitches (like clicks and pops or scrambled audio), because of the randomness of the errors, thereby breaking the coherency of the data. They definitely wouldn't selectively choose which bits to change so as to keep the audio signal's coherency (i.e. no glitches) and alter some property of it (warmth, soundstage and whatnot). You would need some pretty smart "alteration" with lots of coherent (i.e. selective) changes for that to happen.
In other words: if you change a single bit in a packet of structured (meaningful) data, you break it. It turns into gibberish. It's not a slight or subtle alteration. It can turn the number 127 into 255 [*], double the original value. In contrast, if you superimpose a 60Hz hum on an analog signal, well, you still get the original music, with an added bass hum.
P.S. I just thought of an example of digital transmission errors that you might have had the occasion to witness: Over The Air digital TV. Every now and then, you get errors. What do you see and hear? Warmer colors? Constricted soundstage? Nope. What you see is blocks of pixels that look completely wrong. When the transmission is really bad, the image is completely garbled. Sound also gets affected, and it sounds like, well, glitches. Nothing subtle about it.
If that doesn't convince you, or at least makes you doubt your beliefs, I don't know what will.
[*] 127 (base 10) is 01111111 in binary (base 2). Change the 0 into 1 and you get 11111111, which is 255.