|
What the hell is dithering ? (pg. 2)
|
View this Thread in Original format
| Derivative |
thats a good philosophy. megalomaniacal in a way. but good.
i dont feel as if i fully understand something unless i can explain it to somebody who doesnt know anything about it. and make them understand it. so its good that there are now all these people around that know about dither and end up experimenting with it. forces me to get my arse in gear and try my best to out do them all. heh. so it benefits everyone.
im not into keeping knowledge to myself either. i think really all that is disseminated in this forum is actually very very simple when you break it down and digest it in the right order. the hard part is finding out. and the time it takes to cram all that knowledge into your brain and keep it there. |
|
|
| harriz |
| quote: | | Originally posted by Derivative |
| quote: | | [i][b]Originally posted by Djshy[i][b] |
Cheers guys.Thanks for sharing:)
Makes perfect sense.:)
It acts like a mask on the recording
smothing/curving the sound by shading the ''harshness'' of the 16 bit file.
Where you originally had a file at 96/24,
the same file now has less space sounding a bit more ''square''.
By masking the slight ''harshness'' you make the recording sound closer to 96/24.
But by doing it more than one time it goes from sounding ''warmer'' to sounding ''muddy'.
So I will not add that because the file should be dithered & mastered properly already.
Thank you very much for explaing. :)
Are the loops/samples/drum sounds availabe on sample cds/dvds and on the net dithered or undithered? |
|
|
| harriz |
| quote: |
thy are you anengineer ? |
Very interesting stuff !!:eyes: :gsmile:
This is what I want to study too!
Doesnt have to be about music either, just anything to do with sound and recording.
How do you like it ?
How hard it is to learn/understand recording/shaping sound ? |
|
|
| DigiNut |
| quote: | Originally posted by Derivative
dithering is the process of adding low level noise to a recording.
|
That's a major oversimplification considering that there are so many dithering algorithms.
I suppose it depends on how you define "noise". Quantization noise sounds like harmonic distortion because that's essentially what it is - the result you get (not the process itself) is similar to a "sample-and-hold", which makes the waveform squarish and adds the undesirable odd harmonics that sound gritty to a listener.
Dithering can be as simple as taking averages of the waveform, which will result in kind of a white noise; or it can be something far more complex like UV22-HR (the algorithm itself isn't public domain, but the result is a DC bias near the 22 kHz mark, which is *NOT* considered "noise" by most engineers), or Waves IDR, which interpolates the signal up to 48 bits and does some sort of waveshaping before re-quantizing it back down to 16 bits (or whatever the desired bit rate is).
So let's all be clear on this - dithering is not the process of adding noise, that just happens to be a side effect of certain quantization methods.
If you want the "long story" on what dithering is, rather than the abridged version in my first post, then I'd recommend looking up "dithering algorithms" or "decimation algorithms". There are literally hundreds of different ways to do it, all suited for different tasks. |
|
|
| Icone |
So, in the end, is it always best to dither when converting from 24bit to 16bit?
I mean, will a 'regular' listener possibly hear the difference from a sample that has been dithered and one that hasn't been? Or is this just some sort of 'beauty trick' for engineers? ;) |
|
|
| DigiNut |
| quote: | Originally posted by Icone
So, in the end, is it always best to dither when converting from 24bit to 16bit?
I mean, will a 'regular' listener possibly hear the difference from a sample that has been dithered and one that hasn't been? Or is this just some sort of 'beauty trick' for engineers? ;) |
Given the number of releases you've got, it obviously can't be that important. :p
It's like cologne - it has its purposes, but lots of people do without it and dousing on too much can make things a whole lot worse. :p |
|
|
| MaxC |
Here's a dithering conundrum for you:
In my frequent travels to audio production forums, there seem to be two recurring rules to dithering. Those rules are as follows:
1. Always dither when going from a higher bit rate to a lower bit rate.
2. Never dither more than once.
My question is, if you mix down a track to 24 or 16 bits with the intent of importing it into a composition (which will itself be dithered eventually), do you dither? If you do, you will violate Rule #2; if you don't, you violate Rule #1. I've never had this matter cleared up to my satisfaction... |
|
|
| DigiNut |
| quote: | Originally posted by MaxC
Here's a dithering conundrum for you:
In my frequent travels to audio production forums, there seem to be two recurring rules to dithering. Those rules are as follows:
1. Always dither when going from a higher bit rate to a lower bit rate.
2. Never dither more than once.
My question is, if you mix down a track to 24 or 16 bits with the intent of importing it into a composition (which will itself be dithered eventually), do you dither? If you do, you will violate Rule #2; if you don't, you violate Rule #1. I've never had this matter cleared up to my satisfaction... |
You don't mix anything down to a lower bitrate until the final master. |
|
|
| Storyteller |
| quote: | Originally posted by Icone
So, in the end, is it always best to dither when converting from 24bit to 16bit?
|
Any decent sequencer works 32bit floating point internally when it comes to sound processing. in the end sound wil be translated back to the sound properties of the track settings. Any studio god-like guy taking himslef seriously works in the bitrate/hz range which is most likely to be the final processing rate.
So if you're producing for a dvd it's likely your whole project would be 96khz/24bit, in this case you don't need to dither at all, overcoming all these seemingly notable (by uber leet pro's :D) problems..
So if you're working for a cd this would mean you'd work on 44.1khz 16 bit the entire project. |
|
|
| J.L. |
have a question:
Let's say you were to convert a 32 bit wav file into a 192 kbs mp3 file. Would dithering still apply, or is that only for 16 bit? Or am I not making any sense??
Btw, thanks for you audio people sharing your knowledge. (BTW, I'm an engineer too... just not of the audio kind :( ) |
|
|
| Storyteller |
| mp3's are always made of 16bit wav files, so your 32bit file will be resampled and dithered into 16bit wav and then it will be converted to mp3. |
|
|
| Icone |
| quote: | Originally posted by DigiNut
Given the number of releases you've got, it obviously can't be that important. :p
It's like cologne - it has its purposes, but lots of people do without it and dousing on too much can make things a whole lot worse. :p |
Hehe, T-Racks does do the dithering to 16bit, so I kind of always use it automatically... |
|
|
|
|
