Wait wait wait. Are you guys seriously comparing the quality of burned disks, not disk vs file?

If so, then like I said, you're listening to the burner and media. And considering the lowest-common-denominator burners with pennies-per-piece media, yeah, sure, they might verify fine...but cause all sorts of problems in a "real" CD player. And yes, that will be audible.

But, yeah, bits are bits, bytes are bytes, code is code, storage is storage, it's either there or it isn't, it's either correct or not. Otherwise, like I said, *all of computing* would not work *at all.*

FTW

A program is much more sensitive than audio. Bad bits may induce a noticeable effect in audio. Bad bits will stop a program dead in its tracks (program bits, not data bits).

Nearly everything in a computer is checksum to some extent, even your RAM. =D Expensive RAM even does error correction for those applications where a 64 digit number has to be perfect or else.


EDIT: This thread moves fast! 8 replies since I started.

A program is much more sensitive than audio. Bad bits may induce a noticeable effect in audio. Bad bits will stop a program dead in its tracks (program bits, not data bits).

Nearly everything in a computer is checksum to some extent, even your RAM. =D Expensive RAM even does error correction for those applications where a 64 digit number has to be perfect or else.

Exactly! Guess how many so-called 'engineers' I've met don't understand the difference? Oops, I only have ten fingers...

By the way, if the goal is just to avoid USB for some reason, a network renderer pulling audio from a NAS via Ethernet like the Auralic Aries will do that. The problem is, these things generally cost more than home built local storage music servers, and they usually sound worse. Naim's NDS sounds lovely, but it costs literally all the money.

USB, with output via a dedicated card, and going into a well designed asynchronous interface I think is quite good.

I'm critiquing the concept of over-reliance and over simplification of checksums wrt both. Bear in mind, any audible discrepancies can only be determined at the point of playback. So that makes it more difficult to point to where exactly one might be hearing something if anything at all.

What I'm saying is, if you're comparing a pressed CD (usually with good reflectivity, well-shaped pits, consistent track spacing) to a burned CD (with less reflectivity due to the nature of the media, unknown-shaped pits depending on laser, unknown track spacing depending on laser), then you're hearing the media and burner, because the lower-quality copy (which verifies fine on a computer) will cause the CD's tracking servo and error correction to go nuts. It will produce an output, but it may be audibly different due to the machinations of the servo and error correction--the first of which can cause jitter to go nuts, the second of which will interpolate missing stuff.

A much better comparison would be good disk to ripped file.