http://www.digido.com/

articles --> dither

Quite math intensive, but if you skip that and try to understand the underlying principles instead, it's quite a good read.

The good stuff starts at Part II

Daniel,
Whenever you need to reduce bit lenght, you must use a dither algorithm so as to avoid distortion and artifacts.
For example, transferring 24bit tracks from PT to ADATs (16/20bits depending on model), mixing down to older DAT machines, MDs (16 bits), etc.
There is a little difference between when using dither and power dither.
Dither shall be selected applied to individual track fades inside PT environment, once you are a PT LE user.
Use Power dither Level 3 (which is the one I recommend) when you are doing the FINAL bounce of your MIX (24 bit/44k to 16bit/44k).
A very common mistake inside PT environment is to make the bounce of a 48k/24 file directly to 44k/16. We must first do the frequency conversion to 44k/24 and then 44k/16 with a nice (power) dither plugin.
It is a pain in the ass doing all thsese bounces, lots of additional audio files and so. Besides, trusting in PT for all these bounces...
You might have seen a few weeks ago, I did a poll and around 64% of guys still work at 44k/24.
Opus and other guys are big fans of Apogee dither and so. Select the one that best fits your needs.
I myself do not enjoy very much the Waves and Master X dither.
Hope it helped ya
Nice monday

Thanks falkon2 for the article......still working through all of that, but what a great learning device.

Alecio, thank you, too for the info. May I ask another dumb question?

Should I dither before using the "Bounce to disk" option in Pro Tools? Does this not do the same as dithering? Thank you so much for helping me clear up my misconceptions!

-Daniel

Hi Daniel! No questions are silly on here friend!
what are your parameters of your daily recordings sessions? 44k/24? 48k/24?96k?

are ya importing tracks from higher sample rate/ longer bit sessions?
NO need to dither inside, except for the option of applying dither to fades.

I have question. If you have used the waves mastering bundle you know that they have dither options. L2 obviosely using IDR, but the other give you a dither on or off option. I use a HD system and the processing is at 48 bit. I know that when you put a plugin on a track it is going to take your 24bit data (givin your workin at 24 bits) and do the processing at 48 bit and then dither it back down to 24 bits. Am I right or am I off? Well the question is if you leave the dither set to off on the Waves plugs, say I have a LinLB, LinBrB, and then an L2 at the end of the chain is it going to keep the data at 48 bits untill it hits the L2 where it is dithered using IDR to 16bits, without the need for each plug to go 48bits back down to 24?

Alecio-

I almost always work @48/24, but some times need to bring in 48/16 or 44/16 files into a 48/24 session.

Daniel

what is your main stuff? post? dvd? CD audio?

Post and CD Audio

Video guys do 48k/16.
I you plan archiving for future DVD release, 48k/24 is nice but for CD Audio, in my opinion, I would stick to 44k;24 bits.

MIssilanious, I´ve got something for you, after doing my homework(!!!) and checking with Gurus...

It will take your 24 bit data, process at 48 bits, then truncate it to 24 bits. If the manufacturer applies dither within the plugin it will add dither as a final stage, so that when it is truncated that is done so with dither applied.

Quote:

" Am I right or am I off? Well the question is if you leave the dither set to off on the Waves plugs, say I have a LinLB, LinBrB, and then an L2 at the end of the chain is it going to keep the data at 48 bits untill it hits the L2 where it is dithered using IDR to 16bits, without the need for each plug to go 48bits back down to 24? Supose using Pro Tools HD".

Answer:
After EACH plugin the data is reduced to 24 bits. It has to, because each of the Motorola chips can only receive and can only send 24 bit data, so sending the data between the chips (like from the mixer chip to the plugin chip, back to the mixer chip, on to the next plugin chip, etc.) has to happen at 24 bits.

Hope it helped ya dude!

(Many thanks to Nika A., my PT Guru)

Also remember that you do not want to dither during recording the tracks...too much dither can indeed cause mastering engineers to scream, curse and forever remember you for giving them a nightmare of a time dealing with your final product!

There are some dithering applications in which have a certain setting in which allows you to set it for when you plan on using dither in multiple passes.

For example the UV22 has a normal and a low setting. The Low setting is for when you do plan on using multiple dither passes. Although it is typically recommended to only dither once there may arise on occasion where you will need to.

What that occasion is I don't know..but it's there! lol

Opus :D

Ii'm aware of putting dither such as IDR Pow-r dither or whatever during your last and final stage of mixdown, thanks for the answer to my question Alecio.

Ya wellcomed!