Dithering in Sonar

http://www.youtube.com/watch?v=Sl0U_L3tb_M watch this , this is pretty good, and yeah you can just use the dither on the L3

Chregg


ps just as a matter of curiosity, which dither are you using in sonar rec, tri or one of the pow-r ???

I use the pow-r (mostly 2 or 3)
 

ambusher

Chregg


ps just as a matter of curiosity, which dither are you using in sonar rec, tri or one of the pow-r ???

I use the pow-r (mostly 2 or 3)
 

Scott Garrigus in his Sonar X1 Power book recommends Pow-r 3 "because it provides the most accurate processing...also requires the most computing power.."
 
John

"Hi all,you must be careful to use one dither on export". i think he is only using one dither pal, one of the pow-r dithers

Izotope published this great guide that helped me understand dithering. It's a good read whether or not you use Ozone

When I use any of the Sonar's dither algorithms the result is a noticeable noise floor (-50dB) in the exported wav file.

If the resulting noise floor is -50db, that's way too high and definitely not the result of choosing one dithering algorithm over another.

Something else is amiss, perhaps inadvertently applying dither more than once, e.g. enabling dither in a mastering limiter as well as in the SONAR export options.

dlesaux


Izotope published this great guide that helped me understand dithering. It's a good read whether or not you use Ozone

I think link should be:
http://www.izotope.com/products/audio/ozone/OzoneDitheringGuide.pdf
 
John

Ok I have an excellent book on this but cannot find the chapter on this.  Pow-r 2 is suppose to be for less dynamic music and Pow-r 3 is suppose to be for dynamic music.  This is too simple though, if you don't have any mid frequency problems then I use Pow-r3 and if I do then Pow-r2.

"Since noise audibility depends on the nature of the audio being processed, the POW-R algorithm is made available in three variants, optimized respectively for simple program such as spoken word, limited dynamic range program such as rock music, and wide dynamic range program such as orchestral music[5]. Each algorithm moves the noise to those frequencies where its audible effect for the particular audio type is minimized".   Sorry copy from wiki but I have books that back this claim up

This just confirms what I have said.

What nobody has mentioned and do not seem to understand is dither is only important for time based effects,  so lets say you want to make a premaster file ready for mastering.  I would dither here rather than when you go from 24 bit to 16 bit.  This is becuase it is unlikely you are going to use time based effects on the master.  All the good stuff was process at 32bit internally but you have bounced down to a 24 bit file, you have just lost 8 bits of reverb tails and delays and the like.

Any questions???

Peace Ben

BenMMusTech

 All the good stuff was process at 32bit internally but you have bounced down to a 24 bit file, you have just lost 8 bits of reverb tails and delays and the like.

This statement is wrong as 24 bit files are integer files whereas the 32 bit files are floating point which has 25 effective bits when converted to 24 bit integer.  Hence when going from 32 bit floats to 24 bit integers you lose one whole bit of reverb tail.

So is dithering bad if you handle everything at the sample sample frequency and bit depth, sounds like it? Is it like Antailiasing in RGB Graphics?

Actually, graphics anti-aliasing is a pretty good analogy. The principle is similar.  And like in the graphics world, the higher the resolution the less necessary dithering is - get high enough resolution and it becomes completely unnecessary. Also like in the graphics world, once you add dither there's no taking it back - you have permanently modified the original data. And in both graphics and audio, it's best to do it only once, and only when converting to a lower-resolution format.