To Dither or Not To Dither...

[alphacat]

See, this thread is why I still love this forum, no matter what anybody else says.

Thanks a heap everybody. Since this is a sample pack contest on another forum that brought this up I'm gonna have to take a look at the particulars of the samples - rates, depths, etc. - and do some dithering experiments and A/B them with the lowest and highest quality samples respectively using a flat/TPDF/Hamming type algorithm.

Big Up The Audio Wiz Massive....

[macc]

True, but my understanding was that the dither just makes the quantisation error random, rather than directly correlated to the signal. Don't you only need 1 effective bit of additive noise in order to do this?

Yes, but the noise has to average correctly (0001001111 or whatever in the LSB) in order to be effective, ie to be dither. Normal noise from hardware gear doesn't behave that way. All dither is noise, but not all noise is dither

I'm totally with you, I was just being finnickity for the sake of it. It is just a tickbox after all; I never said you shouldn't...

No sir, that wasn't directed at you at all, sorry!

It was at the 'I don't hear the difference so I don't bother' crew. It's not like it takes a lot of work, so why not?

[macc]

The guy sounds like a bit of a douche, but it's cool

I reckon anyone who cares that much about dither is bound to be a douche.

I master music all day long and care very greatly about audio quality, but you have to draw the line somewhere. Dithering is good practice, and being careful with it when appropriate is also. But it's a miniscule piece of the puzzle and obsessing over it for each and every tune (at least, in loud music) just seems like overkill to me. Even my geekiness has its limits.

Now I am off to the pub to get shitfaced

[FSTZ]

MACC: don't drink and dither

[staticcast]

Gotcha. Man, if the production forum were a house party we'd probably be the guys drinking orange juice and arguing over Fermat's last theorem...

[staticcast]

Haha, thought I posted that before macc's last reply

[macc]

MACC: don't drink and dither

@ static _cast - that thought keeps me awake at night! I may be a geek but I am from Essex too

[abZ]

True, but my understanding was that the dither just makes the quantisation error random, rather than directly correlated to the signal. Don't you only need 1 effective bit of additive noise in order to do this?

Yes, but the noise has to average correctly (0001001111 or whatever in the LSB) in order to be effective, ie to be dither. Normal noise from hardware gear doesn't behave that way. All dither is noise, but not all noise is dither

I'm totally with you, I was just being finnickity for the sake of it. It is just a tickbox after all; I never said you shouldn't...

No sir, that wasn't directed at you at all, sorry!

It was at the 'I don't hear the difference so I don't bother' crew. It's not like it takes a lot of work, so why not?

Let me clarify something tho Macc, everything that goes out from me or my label is mastered professionally. Pretty sure my guy dithers but whatever he thinks is best. When I do some self masters for test purposes I master from 24 bit and then straight to 320. No dither. I usually use Sound Forge I think it is slightly more than checking a box. It is enough that I don't bother usually. Point is I leave dithering up to you guys.

[naus]

Basically, when you convert from 24 bit to 16 bit the 8 'least significant bits' (a.k.a. the bits with the least amount of data on it) is removed (which is why the headroom tends to be a problem when downsampling) and something called 'quantisation error' can occur, which presents itself in the form of nasty artifacts; by reducing bit rate, errors can occur when a sample exists on a value that doesn't exist in the new bit rate (for the sake of argument i'm going to give the example of 4.5; it has to change to either 4 or 5 upon resample, both of which are wrong).

Dither is essentially white noise intentionally placed on top of the track to make the error (and resulting artifacts) less noticeable, and to help make it less audible the noise can be sculpted to avoid certain frequency ranges. As a rule, if you look at a dithered piece of audio through a frequency analyser (this is easier when you do it when the track is silent or at it's quietest) you'll see that there's a 'blanket' of noise throughout most frequencies with big spikes in noise in the 0-40Hz and 18-20KHz ranges that's dither!

Hope this sheds some light on dither peeps :3


Dub Fiend

sure listen to track in my sig. uploaded today. there are subtle low end pops im sure of it.


Is this why when i upload 320's to soundcloud I get these popping type artifacts on playback?

Do you have an example? Id suspectzero crossings, I never notice if there in my tunes until I upload...

[macc]

Let me clarify something tho Macc, everything that goes out from me or my label is mastered professionally. Pretty sure my guy dithers but whatever he thinks is best. When I do some self masters for test purposes I master from 24 bit and then straight to 320. No dither. I usually use Sound Forge I think it is slightly more than checking a box. It is enough that I don't bother usually. Point is I leave dithering up to you guys.

Gercha

I wasn't quite with you. As it happens some people report better results going 24 > mp3, though it is, I admit, still on my list of things to test (I rarely make mp3s).

Nevertheless, if you are going 24 > 16 in wav/aiff format then you may as well add dither - most limiters (which, is presumably the last thing you'll be doing) have a dither included.

Anyway, noooo wuzzas

P.S> not checked out that USB error thing since we spoke, will check when I get a minute

[abZ]

Let me clarify something tho Macc, everything that goes out from me or my label is mastered professionally. Pretty sure my guy dithers but whatever he thinks is best. When I do some self masters for test purposes I master from 24 bit and then straight to 320. No dither. I usually use Sound Forge I think it is slightly more than checking a box. It is enough that I don't bother usually. Point is I leave dithering up to you guys.

Gercha

I wasn't quite with you. As it happens some people report better results going 24 > mp3, though it is, I admit, still on my list of things to test (I rarely make mp3s).

Nevertheless, if you are going 24 > 16 in wav/aiff format then you may as well add dither - most limiters (which, is presumably the last thing you'll be doing) have a dither included.

Yeah by the way that patch didnt do shit. Still looking for a solution to me problem.

What limited do you use that includes dither? I have just been using classic limited which is pretty featureless.
Anyway, noooo wuzzas

P.S> not checked out that USB error thing since we spoke, will check when I get a minute

[abZ]

Ahh I screwed up that quote but on mobile so it would be a pain to fix it right now srylul