BenMMusTech


You always want to have the last word

Oh the irony 

mattplaysguitar

BenMMusTech


You always want to have the last word

Oh the irony 

It was suppose to be!

For those interested in the twist introduced with the accumulation of noise across several channels...

Music signals sum additively (as you would expect) whereas noise signals do not - that is uncorrelated noise signals do not. In this sense uncorrelated means random and not in any way correlated to the music signal on the channel. This kind of noise is typically caused by thermal noise in resistive components and is usually the type of noise that comprises the noise floor of well designed circuits.

In a way uncorrelated noise is benign in that adding together two channels does not double its level but increases it by square root of 2 - about 1.4, and adding together 4 channels increases it by only 2 times, 16 channels - only 4 times. That is, its result is the square root of the sum of its squares. 

This is why some very low noise pre-amp circuits use parallel signal paths as you get a halving of noise to signal for each 4 parallel paths. It also explains why wide tape is quieter than narrow tape.

On the other hand correlated noise does sum normally. Correlated noise being noise related to the music in the channel eg intermodulation distortion, harmonic distortion, clipping, aliasing noise, over and undershoots, ringing etc.

On the main topic, imho the sensible approach is to be aware of the compromise necessary to make sure your signal is large enough to have a usable SNR, and small enough to guarantee no input clipping. That will depend on source material, won't it? 

Rock guitar has such a low dynamic range that it can be recorded at almost any digital level, any recording noise being swamped by amp and pick-up noise, and the loudest parts limited by the self compression and limiting of speakers and amp at full output. On the other hand, classical music demands the best from both recording equipment in terms of self noise, and in the choice of nominal recording level so that there is not a chance of input clipping. Also bear in mind that classical music is often released on CD at low levels in any case. We are talking 16 bit here and the music is quite capable of being achingly beautiful at levels far less than -20dbfs

The great thing about 24 bit recording is that you don't need to sweat too much to get as close to 0dbfs as possible. 

Can any of you say that pieces you have recorded at -20dbfs or less have suffered - except from when you got the gain-staging in the analogue world wrong? 

BenMMusTech


 

Have you even read any of that Paul Frindle stuff I mean you are quoting most of his stuff and he def sugests the posibility of harmonic distortion,  who know's where it is happening??

Pretty much everyone else in the thread knows. But as you usually do, you're assuming that you not understanding something means it's somehow ineffable. It isn't, it's just that YOU don't understand it. But being so aggressively determined not to listen to what people are telling you, I do wonder what the point of these threads is. After all this verbiage, you are still as clueless as you were before. When was that "digital warmth" one? Something like six months ago? That's a long time to have learned nothing.  

There was a guy who used to post regularly over at GS by the name of OldAnalogeGuy and he used to come out with similar BS, I wonder if they're related