Dither to 16 bit CD, and no need to dither MP3s.  Tho for finalized stuff I do mp3s from 24 bit masters.  You used to not could do that, but SF seems to do it now tho I haven't updated for years.

@

but sonar always creates a wave file (albeit temporary) THEN creates an mp3 from that, doesn't it?  at least it used to.  when I create an mp3 from sonar, I leave my dither settings on - for one thing I see no need to change them as it shouldn't make any difference one way or the other if it goes directly from project to mp3 and for the other, I thought sonar created a temporary wave first, then an mp3, so I thought the dither would be of some benefit.

Hi Beagle,

I export out of Sonar to a .wav that's the same as my project. I normally use 96kHz/32bit. Probably explains the wisps of smoke always coming from my CPU.  Then I load the .wav in to Sound Forge 9.0 and use it's MP3 codec. The manual says it's not Sony's but is licensed from Fraunhofer.

SF9.0 has a lot of MP3 options. I use 320/Hi-Q/Stereo. There's several different stereo options.

"Bud"
I export out of Sonar to a .wav that's the same as my project ... Then I load the .wav in to Sound Forge 9.0 and use it's MP3 codec.

I do the same thing with the exception that, before I convert to an MP3 (and yes it's the Fraunhofer codec - highly rated codec btw), I pull down the "Process" menu --> "Bit-Depth" --> "iZotope MBIT +Dither" and choose from bit depth, dither mode, dither amount and noise shaping to name a few available options.

"AT"
... and no need to dither MP3s.

Up until you've mentioned this, I was under the belief that any time I was changing the bit-depth, dithering was necessary to avoid having "good audible bits" removed instead of the desirable "bad inaudible bits" (abbreviated for simplicity's sake). I work at the same bit rate all the time and don't use dithering until I'm ready to convert to a lower bit rate. I've been wrong about this? 

You should dither once, Alegria.  Or what I've always heard/read.  Since it is introducing noise into the material that makes sense.

So I dither going down to 16 bit/CD.

@

"AT"
Dither to 16 bit CD, and no need to dither MP3s.

"AT"
You should dither once, Alegria.

 Ah, ok. I misunderstood you then. That's what I've been doing. Thanks for the confirmation. 

I work at the same bit rate all the time and don't use dithering until I'm ready to convert to a lower bit rate. I've been wrong about this?

You're doing it right, Alegria. (Except that the phrase you want is "word length" rather than "bit rate".) The confusion, I think, is whether or not you need to reduce the word length prior to MP3 encoding. You don't, as long as your encoder can handle 32-bit data (most, but not all, do).

I turned off dithering in SONAR as my workflow is now:

1. Export a 24/44.1 WAV mix.
2. Edit/clean up the WAV in Wavelab
3. Import edited WAV into a separate per song "mastering" project
4. Export dithered (set in my final stage mastering tool) 16/44.1 WAV or mp3. If I need a (rare) 24/44.1 master WAV I just deselect the dithering.

NOTE: I have the SONNOX tool bit was alluding to. I'm just now getting into learning it thanks to the groove3.com video.

"Bapu"
I turned off dithering in SONAR

Same here, but I do the mastering in SF 10 Pro with the help of Ozone 5. I haven't had the chance to publish to CD yet, and have only published to MP3s.

"bitflipper"
You're doing it right, Alegria. (Except that the phrase you want is "word length" rather than "bit rate".) The confusion, I think, is whether or not you need to reduce the word length prior to MP3 encoding. You don't, as long as your encoder can handle 32-bit data (most, but not all, do).

Thanks for dotting the "i" bit, I'm taking a closer look at this now. 

mattplaysguitar


How can the differences be 'dramatic', but only with $10,000 monitors AND brilliant ears. I wouldn't call that 'dramatic'.... Maybe 'subtle' would be a more appropriate term?... I assume were are talking 320 kbps VS wav here.

I agree, dramatic is a bad choice of word here.
 
But thanks for the replies guys.
I never use mp3 in a "serious" session, always trying to use the best quality possible. I'm going to keep up my "ear training lessons" because I know there's a lot to learn.
But I believed that I should be able to hear the difference between 128kbps .mp3 and .waw files, but  in this little personal blind test, I failed.