Hey Everyone, I have a question about dithering ... I know the basics about dithering and understand what it does but what I want to know is what is the difference between the dithering options when exporting ? ... triangular, rectangular, power 1,2 and 3 ..... what is the criteria for using one over the other ?
Any help would be appreciated .....

I am prepared for arguments, but I don't believe any human will hear the difference among those dithering options and I have seen some scholarly articles supporting this position.  I usually use Power 2 but only because I just did that once and never changed.

This might help http://www.cakewalk.com/Documentation?product=SONAR%20X2&language=3&help=Mixing.61.html

konradh
I am prepared for arguments, but I don't believe any human will hear the difference among those dithering options and I have seen some scholarly articles supporting this position.  I usually use Power 2 but only because I just did that once and never changed.

Not going to argue, but I did a classical project once where the harpsichordist thought there was something "wrong" with a track but couldn't put her finger on it. Turned out dithering had been turned off accidentally, which it hadn't been for the other tracks. When dithering was turned on, then to her it sounded "right."
 
So...if someone can hear the difference between dithered and non-dithered, they may be able to hear the difference between different types of dithering. Although in this case, it was more something she "sensed" than "heard." (And if there's a car going by outside, you're not going to hear any difference, no matter what!)

gitarman4u
Hey Everyone, I have a question about dithering ... I know the basics about dithering and understand what it does but what I want to know is what is the difference between the dithering options when exporting ? ... triangular, rectangular, power 1,2 and 3 ..... what is the criteria for using one over the other ?
Any help would be appreciated .....

Processing time measured in nanoseconds?  
 
I set mine and forget it but I never dither unless changing bit depth. 

As the documentation says, the different dithering patterns have different levels of CPU usage.
These dithering patterns have been around a long time in the Sonar export feature.
I also think they go back to the days when CPU power was much inferior to the systems available today.
These days, why wouldn't you always use the Pow-r 3 dither pattern ? . . . unless your CPU wasn't up to it.
 

I set up a test project which had an EXTREMELY long reverb tail with an automation curve set up to rasie the volume higher & higher as the tail faded out.
 
Then listening carefully on headphones I tried all the different dithering options and to be fair, there were some very subtle differences between them, but I couldn't really say one was "better" than another, except that rectangular did seem noisier than the others.
 
I might try this again with Splat, just for fun.

The 'Digital Audio' section of http://en.wikipedia.org/wiki/Dither explains the dithering process and the different types really rather well.

Educational thread.  Thanks.
 
I recently read others on this topic and was reading Isotope's article on dithering last weekend.  (It was hoppin' at my place, let me tell 'ya.)  One thing the author recommends is to only dither once.  I've been dithering twice, once at mix-down, when I export a multiple-track 24 bit project to a single 24 bit stereo wav, and then again when I export that 24 bit wav from its own 'master' project down to a 16 bit mp3.
 
I've been using Sonar's Triangular and/or Pow-r 1 - just because - but Isotope naturally recommends Ozone's MBIT dithering options.  I've read in these forums that the Pow-r 3 is equally good.
 
As I'm typing this I'm thinking also that I need to research and look at the literature regarding the samples I'm using as to whether they're 24 bit or 16 bit to start with, and also what the mp3 encoder options really mean.  For mp3's I usually export at 192 kbps and set the Sonar encoder slider to '1' for high quality.  I don't check the 'enable high/low pass' options.  I did have a rogue random rumble in a few mp3's when I first started using Sonar Xi.  Can't remember now if it was dither or mp3 settings or both that took care of that.
 
I don't know that I can detect any difference in the dithering algorithms, but it's good to have a basic understanding.  Thanks for the discussion.

I might be doing it wrong, but I only dither with Ozone 5.