• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Digital Audio Explained (from an engineer's standpoint)

Status
Not open for further replies.
Register to hide this ad
"Engineer" means a lot of things. This is a bit like asking a guitarist about capacitors. Higher sampling rates and bit depth do have huge advantages during production, though.
(edit: well, not really "though" - the advantages during production serve to even more strongly bias the perceptions of folks like this when it comes to simple playback)
 
Last edited:
P.S. I'm actually in favor of higher quality digital as standard formats, but in order to allow for higher quality digital manipulation before ever being analog, as part of playback. I suppose I should post that in both threads...
 
I've heard producers and audio engineers tell some of the most incredible whoppers ever.

I've also had to correct the actual mechanical and electrical engineers for some of the contract manufacturing companies I've worked with on things I should never have had to correct them on. Again, the more you take for granted the more you become at the mercy of the authority of who is providing you with the information...
 
I love the part where he says:

"You'll get more resolution at that frequency".

:no:

Um. Well, there's another person who doesn't understand sinusoids <sigh>. Herein lies the danger of someone who is technically inclined, but only to a point and tries to leverage that knowledge in ways that stop adding up very quickly. Hearing this guy talk about digital copies becoming degraded is where I stopped listening. This is the kind of reason there is so much bullshit everywhere. People with a certain form of technical background think their knowledge gives them this magical extrapolation power which provides them this incredible wealth of understanding in areas that have zero correlation (viz. actual electronic engineering). He sounds seriously straight out of a marketing class. He is literally all anecdotal data.

"Yeah I totally like stood behind John Williams while he was conducting orchestral scores for super famous stuff so obviously that means I know just everything there possibly is to know about audio and all that".
 
Last edited:
rifftrax said:
Hearing this guy talk about digital copies becoming degraded is where I stopped listening.
Click to expand...
And yet, experiencing that phenomena ticked off Mariah Carey such that half a million of her CDs were scrapped because they sounded bad as compared with the master. :)
 
He made a general comment that lots of things can go wrong and then stated the production house used a double speed process for the glass masters and she didn't like it - no direct reference to digital copy degradation, or what actually caused the issue.

Gotta run - I copied a word doc 5 times so now I have to go correct the spelling. Just the vowels though. :)
 
jjones3 said:
He made a general comment that lots of things can go wrong and then stated the production house used a double speed process for the glass masters and she didn't like it - no direct reference to digital copy degradation, or what actually caused the issue.
Click to expand...
Exactly!

Bits are not bits and there's far more to go wrong in the real world.
 
Any modern piece of digital equipment that has jitter, be it computer, audio or communications, should be tossed. Jitter is easily fixed and has been for decades.

Timing issues are, in reality, not a big issue. With modern PLL and high speed stable, glitch-free clocking circuitry there should be no timing issues unless you are thinking of the state of digital electronics back in the late 60s, or you got one badly designed piece of equipment.

And, bits really are bits. That's the whole premise of digital. :)
 
E-Stat said:
Until of course you consider jitter and other timing issues when you evaluate what we actually hear.
Click to expand...

Let's stick to the topic being discussed, which is digital copies becoming degraded. Jitter and timing issues, which are solved problems, do not apply. For a copy of a bucket of bits to be degraded, there has to be a difference. Discounting corruption and errors which are easily detected, there is simply no difference in a bit perfect copy.
 
jjones3 said:
Let's stick to the topic being discussed, which is digital copies becoming degraded. Jitter and timing issues, which are solved problems, do not apply.
Click to expand...
Jitter is solved? If you say so! By timing, I refer to systems that must decode lossless codecs like FLAC on the fly and phase issues caused by the brickwall filters required for the crippled Redbook standard.

jjones3 said:
For a copy of a bucket of bits to be degraded, there has to be a difference. Discounting corruption and errors which are easily detected, there is simply no difference in a bit perfect copy.
Click to expand...
That's the theory, isn't it? :)
 
E-Stat said:
That's the theory, isn't it? :)
Click to expand...

You crack me up. Look, if you want to have a discussion, I'm all for it. I'm always open to learning new things and testing what I believe I know. Hell, I even enjoy a good debate now and then. But, you have to drop the vague statements, generalities, troll bait, and actually... you know... discuss something.
 
jjones3 said:
You crack me up. Look, if you want to have a discussion, I'm all for it.
Click to expand...
I present the viewpoint from a recording engineer who has worked for decades with a wide range of gear and formats. If you watched the video, you'll remember his uneasiness over observing audible differences he cannot measure.

I'll take the opinion of guys like that (and another who records for Disney who posts on another board) who have compared live feeds with various recording formats and finds differences in all of them. And with generational copies. Especially when it concurs with my experience and that of audio reviewer friends who I trust.

Yogi Berra says it best:

"In theory there is no difference between theory and practice. In practice there is."

YMMV
 
At 14:05, Allen Sides makes a statement to the effect that sampling at 192khz is superior to 44.1, not because it will include frequency content above human hearing range, but because higher hearable frquencies (he uses 15khz as his example) will be sampled more and so will have "better resolution". This is a common misunderstanding of the sampling theorem and a misconception I was a little shocked to hear him utter.
 
RobInONT said:
At 14:05, Allen Sides makes a statement to the effect that sampling at 192khz is superior to 44.1, not because it will include frequency content above human hearing range, but because higher hearable frquencies (he uses 15khz as his example) will be sampled more and so will have "better resolution". This is a common misunderstanding of the sampling theorem and a misconception I was a little shocked to hear him utter.
Click to expand...
Fine. We all know the underlying problem is the unavoidable ringing and phase errors introduced by a brickwall filter that has to completely suppress all content in the space of one-tenth of an octave.

That's the real problem. Higher sample rates don't require that crutch. You won't find a single speaker on the planet that uses a brickwall filter for obvious reasons. Many in fact use 6 db/octave single order crossovers to ensure phase integrity.
 
I'm happy to trust (or at least take very seriously) his ability to identify what sounds good and to make things sound good. He's demonstrated those abilities, extensively.

I don't trust his undemonstrated ability to identify, non-blind, potentially inaudible differences in formats and equipment any more than anyone else's. I simply don't think one follows from the other, because the bias toward always hearing a difference overwhelms the skill, and then some.

That's not to diminish the skill, only to say that his explanations and anecdotes related to these things do not carry any special weight for me, because to me accuracy is neither demonstrated nor expected. If I were assembling some sort of dream team to listen to something, evaluate adjustments to something, etc., I'd certainly be happy to have him on it.
 
The development of oversampling and digital filters have rendered those "problems" insignificant. I have seen no dbt that demonstrate that redbook standard is inadequate for playback files. Do you have any empirical evidence, i.e. double blind or abx comparisons, that show that redbook is "crippled"?
 
Status
Not open for further replies.
Back
Top Bottom