bits and khz of your audio

gabriel g.

gabriel g.

Member
Sep 7, 2006
3,171
1
38
Hamburg
www.myspace.com
I wrote this at gearslutz.com in a thread about how to do best qualitiy recordings (khz and bit)
Maybe this helps some people to understand things better

1. Point for the bit thing:)

Signal to noise ratio of the PCM (pulse code modulation)

16Bit= 98,08dB
24Bit= 146,2dB

The Math behind this: M(bit you choose) x 6,02dB + 1,76dB (sine with complete approach)

2. KHZ thing:

Sample rate conversion (SRC)

48khz->44.1khz (44100/48000) ca. x 0984... this is interpolation with finite accuracy) so you maybe recieve jitter failures!!!

cheers
 
urmmm what is this function M ??? as in M(bit you choose)

and what do you mean by complete approach?

And there are many many tpes of interpolation used for resampling. Sometimes point is used, but more often linear is used, and i always go for the highest quality one in reaper.

You dont happen to know how much difference those different modes of interpolation make to the final waveform? Surely a worse method of interpolation is simply slower to reach the same accuracy......

or maybe you hit some of the failures of these methods
 
M stands for the bit range you choos (16,24 or 32)
If you go with the highest quality you maybe get jitter failures!!!
Or just go for 88.2kz and you´re fine because no problems in the math.

It is realy THAT simple!!!!
I learned this at collage from a very nice, book writing prof:

maybe you know his book. he´s called Thomas Görne
 
XGabrielX said:
I wrote this at gearslutz.com in a thread about how to do best qualitiy recordings (khz and bit)
Maybe this helps some people to understand things better

1. Point for the bit thing:)

Signal to noise ratio of the PCM (pulse code modulation)

16Bit= 98,08dB
24Bit= 146,2dB

The Math behind this: M(bit you choose) x 6,02dB + 1,76dB (sine with complete approach)

2. KHZ thing:

Sample rate conversion (SRC)

48khz->44.1khz (44100/48000) ca. x 0984... this is interpolation with finite accuracy) so you maybe recieve jitter failures!!!

cheers
Click to expand...

where does the 6.02dB and 1.76dB come from? and that 0984 (and is it supposed to be 0.984?)? 44.1/48 = 0.91875


edit: found this quote on wikipedia: "however, current digital audio converter technology is limited to dynamic ranges of ~120 dB because of 'real world' limitations in integrated circuit design.[1]"
 
There is an easy way to determine a file's bit rate when given sufficient information. In fact, as long as you know any three of the following four values, you can calculate the missing value.
Bit rate = (bit depth) x (sampling rate) x (number of channels)

(from Wiki - correct, strangely!).

Signal to Noise;

24-bit digital audio has a theoretical maximum dynamic range of 144 dB, compared to 96 dB for 16-bit.

The reason for the odd numbers is due to the Log scale, IIRC.
 
An important thing to remember is there is an analog noise floor -- we can talk all day about adding more bit depth but at some point the molecules in your microphone and cable and AD/DA converters are bouncing and making noise and using a higher bit depth than that does nothing for you. If I recall, 24-bit is enough if the converters are good, and 32-bit is way more than enough.
 
And always remember if you have shitty mic pres with a very low SNR (signal to noise ratio) there is nothing to fix with 24bit.

Also If your doing techno/blast-beat-triggered-to-death stuff there will be no dynamic left so why use 24bit over 16bit?

24x6 or 16x6 is the fast math to get your dynamic...it is close enough to comparing with your tech manual of the mic-pres.
 
the dynamic range of the human ear is around 20bits IIRC.

So, 24-bits is the maximum that theoretically AEs should need. However, high end plugins will usually keep data when multiplying, and therefore requirer a higher internal bit rate to deal with all the "extra" information. Working with really high end plugins will definitely require 32 or more bits to deal with the maths, but realistically - 24 bit recordings are more than enough for "high quality" as these plugins will usually loose a bunch of LSBs to give 24 bit on output anyway.
 
You must log in or register to reply here.
Top Bottom
Back