So now I've actually had a chance to read your post 3f05Q:
Its quite interesting what your saying, and works in theory, but I don't think its true because what seems to be being affected based on those diagrams isn't actually dynamics, but harmonic content. The wave which was originally a sine is being turned into a square - I understand this to be aliasing. So the idea is that with the Nyquist-Shannon formula, you can interpolate to create a perfect copy of the original wave upto the nyquist frequency, but above this the sound will be increasingly aliased, and the wave will look more like a square due to lack of samples to make up the high freq sine wave etc... So I'm not sure why you think dynmics will be affected exactly?
I guess I'm just saying, based on those diagrams, the wave has just as much vertical freedom no matter how much it gets squared by aliasing, so the amount of aliasing might change, but the actual volume won't, leading to equal dynamics across the whole range...
Interesting discussion now...

___________________
New Mix: March 2010 Promo
Soundcloud|Facebook

quote:

Originally posted by kitphillips So now I've actually had a chance to read your post 3f05Q: Its quite interesting what your saying, and works in theory, but I don't think its true because what seems to be being affected based on those diagrams isn't actually dynamics, but harmonic content. The wave which was originally a sine is being turned into a square - I understand this to be aliasing. So the idea is that with the Nyquist-Shannon formula, you can interpolate to create a perfect copy of the original wave upto the nyquist frequency, but above this the sound will be increasingly aliased, and the wave will look more like a square due to lack of samples to make up the high freq sine wave etc... So I'm not sure why you think dynmics will be affected exactly? I guess I'm just saying, based on those diagrams, the wave has just as much vertical freedom no matter how much it gets squared by aliasing, so the amount of aliasing might change, but the actual volume won't, leading to equal dynamics across the whole range... Interesting discussion now...

Yeah, I think that in those cases (or most) that the output through a DAC would result in a sin wave of uniform magnitude. The difference is if you start applying some sort of processing in the digital domain. If you were to apply, for instance, a brickwall limiter on that 11kHz digital waveform it wouldn't be a uniform application. I'm not saying that this is critical stuff, but it does exist. Now, we've been focusing on a pure sin wave and we see inconsistencies at certain intervals. What's to say that behavior doesn't occur right on the attack of a snare hit recording? There's a chance that it can. What if we record a snare hit, or a guitar pluck and we want to retain as much of those dynamics in the recording and be able to manipulate them as close to the analog domain as possible? What if some of that dynamic range gets lost as a result of manipulating the audio in the digital domain? I think recording at a very high bitrate decreases the CHANCES of this happening. That's just my thought, at least.

quote:

Originally posted by 3F05Q Yeah, I think that in those cases (or most) that the output through a DAC would result in a sin wave of uniform magnitude. The difference is if you start applying some sort of processing in the digital domain. If you were to apply, for instance, a brickwall limiter on that 11kHz digital waveform it wouldn't be a uniform application. I'm not saying that this is critical stuff, but it does exist. Now, we've been focusing on a pure sin wave and we see inconsistencies at certain intervals. What's to say that behavior doesn't occur right on the attack of a snare hit recording? There's a chance that it can. What if we record a snare hit, or a guitar pluck and we want to retain as much of those dynamics in the recording and be able to manipulate them as close to the analog domain as possible? What if some of that dynamic range gets lost as a result of manipulating the audio in the digital domain? I think recording at a very high bitrate decreases the CHANCES of this happening. That's just my thought, at least.

i don't think you'll be able to hear those differences really...

___________________
Music----is----universaL

Pay a visit good fella:

Xetro's MySpace

Xetro's Facebook page

If someone can accurately hear the difference between 320kbps mp3 and WAV, I'll gladly pay them to master my tracks.

___________________
Youtube || Soundcloud || Synth Patch Banks

quote:

Originally posted by Vortex_SA i don't think you'll be able to hear those differences really...

Yeah, most likely... but it's an interesting discussion nonetheless.

quote:

Originally posted by 3F05Q Yeah, most likely... but it's an interesting discussion nonetheless.

but with live recordings you would know what could be the difference either, each hit of a snare is different from the other so is each pluck of a guitar, the only way i can think of doing this stuff is by hooking up the same mic to like 4 different recorders all same models and running them through different bitrates and samplerates but even then every device that comes out of the factory has slight differences from the other...

___________________
Music----is----universaL

Pay a visit good fella:

Xetro's MySpace

Xetro's Facebook page

Actually...what I really need to do is hook up an oscilliscope to prove the whole hard limiter on an aliased digital signal blah blah blah thing. So with that realization I retract any conclusion.