To Dither or Not

Dither of course. The video above shows you why. It’s been a while since my last album so in preparing the files for the new baby I realized I need to get things from 32bit to 16bit. As per some Gearslutz advice I’m going with the Pow-r type.

“Dither is an intentionally applied form of noise used to randomize quantization error, preventing large-scale patterns such as “banding” in images. Dither is routinely used in processing of both digital audio and digital video data, and is often one of the last stages of audio production to compact disc.” – Wikipedia

For more info:

Published by

Oliver Chesler

"Hello my name is Oliver and I'm going to tell you a story." I have been recording music since 1989 under the name The Horrorist. I have released over 60 singles and 4 full length albums. To hear my music please go to:

9 thoughts on “To Dither or Not”

  1. nowadays I never dither untill the very final stage before upload – I use r8brain samplerate convertor (free) by Voxengo – I find its quality way nicer than most other methods I’ve tried – reverbs dont go all weird and crashes/cymbal/HH dont tend to turn into (unpleasant) white noise bursts.

      1. -to say that it was more preferred with the sample material given might be more accurate. (also the mbit+ technology can be found in soundforge 10 and some other licencees) Different algorithms do better on certain types of program material- just because it shines on one piece doesn’t mean it’s the best choice for another. When doing Quick and dirty projects, of course we choose what sounds best and keep it moving. However, it is usually not a good idea to dither your projects if they’re going to be sent to a mastering engineer. POW-R has been around for quite some time (want to say since the 90’s) and IMO it’s age shows a bit when compared with more modern alternatives. Still probably the most licensed with PT, Logic, Waves, and others under it’s belt! Hey if it sounds good, do it I say.

  2. Interesting.

    I always wondered, if dithering was a good idea.

    By the way, Oliver, while you are at it. Why don’t you let us all participate in the process of having your album given the final finish. Starting from mixdown, mastering, artwork to delivering the final product.
    I bet we are all eager to see how you do it.
    Can’t wait for the album.

    Cheers. Philipp

  3. No harm in dithering in the final mixdown render, but it’s entirely not worth worrying about. Save that mental energy for focusing on the music itself.

    Highly recommend Ethan Winer’s very informative Audio Myths presentation:

    1. @Mark – a good stance to take, unless someone else is relying on you to be their final step before going to Disk. as a music producer, I would agree with that statement 100% .The Recording Engineer in me also agrees, but a mixing (sometimes called upon to be mastering) engineer definitely needs to at least have an Idea of what Dithering is and what options are out there so they can make an informed decision, or be informed enough to leave Dithering alone so that the next person in the chain has the most options to make the final product sound the best.
      –Rock On

    2. @Mark – I was going to post a link to that video, too! The video Oliver posted is totally unscientific. Sure, you can hear a difference when going to 8 bit, but you probably can’t when going to 16. The Weiner video demonstrates a null test between dithered and un-dithered files. The files almost completely null each other out, proving that dithering is not necessary. However, dithering can’t *hurt* either, so it is still recommended as a general practice.

Leave a Reply

Your email address will not be published. Required fields are marked *