I know you are on vacation and I would not expect or want a response until you return to work fully revitalized!
I have read the “24 (32) to 16 bit” entry to this forum and I am somewhat confused by what was stated so I will attempt to ask a simple question. But before that, I would first of all like to thank you, Aleksey, for your excellent software and your attention to the associated forums you provide to us. I would guess you are a busy person with your company and the fact that you participate directly on these forums is something that motivated me to try your products. That was a mere week or so ago. I have r8brain and Elephant2 as well as your free software. I heard about your software via the Steinberg forum. By the way, where in the world are you located?
For simplicity sake my question is also assuming I do not need or want to use a plug-in such as Elephant2. After I hear your thoughts on this question I’ll get to inserting Elephant2 into the equation (or mix, if you will).
I ask this question strictly from a theoretical standpoint but your subjective experience, of course, would be greatly appreciated too.
To my question:
Assuming I record in 48k 24 Bit (I use Nuendo 3) and my goal is to get the file (music) as accurately (and as sonically pleasing to the human hearing system) as possible to the lowly CD – 44.1k 16 Bit format -
A.) Export Audio Mix from Nuendo as 48k 24 Bit with Apogee UV22HR inserted to dither to 16 bit. Then r8brain to 44.1k 16 Bit.
B.) Export Audio Mix from Nuendo as 48k 24 Bit without Apogee UV22HR inserted to dither to 16 bit. Then r8brain to 44.1k 16 Bit.
C.) Export Audio Mix from Nuendo as 48k 32 Bit FP with Apogee UV22HR inserted to dither to 16 bit. Then r8brain to 44.1k 16 Bit.
D.) Export Audio Mix from Nuendo as 48k 32 Bit FP without Apogee UV22HR inserted to dither to 16 bit. Then r8brain to 44.1k 16 Bit.
E.) Other and if so please describe.
Thank you kindly and sincerely,
My listening tests indicate the same. It is nice when theory and subjective results crossfade with each other.
My next question is, theoretically, if I am wanting/needing to use Elephant2, what are the "correct" steps to accomplish the same ultimate goal (CD - 16 Bit 44.1k). I realize it adds another step (export from Nuendo) , but what are the (theoretically) correct steps to achienve the CD format from the original 48k 24 Bit using a plug=in such as
I think I was wrong because I remember now Aleksey told me the only way to not dither with r8brain is to convert to 32 bits.
So, if you follow my instructions you'll be dithering twice! And this is NOT a good practice at all...
Remember! Dithering is needed only for bit depth reduction (ie. 24/32 to 16 bits), not for SRC (Sample Rate Conversion: ie. 48 to 44.1 KHz). And must be the last step at all.
Since Elephant is a Mastering Limiter (and mastering is the last step in production workflow), the choice of using dithering option on Elephant is up to you. Following this idea, you have to set Elephant as last plugin in the chain and set its dithering option to 16 (bits).
In that case you have a 48 KHz/24 bits file in your Nuendo platform but already dithered to 16 bits. Next step will be that to simply truncate to 16 bits. Maybe this is possible inside Nuendo exporting to 44.1 KHz/16 bits without using Nuendo dithering option (ie. this is possible in Adobe Audition/Cool Edit Pro)
Please note, I did not mention your Apogee plugin because we're assuming you are dithering with Elephant (Dither must be applied only once). Remember also that Elephant has only the TP Dithering mode; if you prefer other dithering algorythms you have to bypass Elephant dithering option.
In my book, export as is first. No change of bit depth !
Then use r8brainpro to samplerateconvert.
Then import and redither/noiseshape to 16 bit.
The last two stages of mastering should be samplerateconversion and then dithering/noiseshaping - in that order ! Never samplerateconvert the noiseshaping ...
(Of course, all this AFTER the work on the master is finished)
I agree with you, Yannick!
Nuendo: Export my 48k 24 bit original file.
r8brain: SRC 48k to 44.1k and leave as 24 Bit.
Nuendo: Insert Apogee UV22HR 16 Bit dither plug-in OR Elephant, if desired or needed. Export as 44.1k 16 Bit.
It is correct to use Elephant prior to Nuendo converting from 24 to 16 Bit?
If you want to use Elephant with the Dithering option switched ON yes, it is correct. It is absolutely useless to dither an already converted 16 bits file (in this case you are only adding noise without removing quantization distorsion occurring when you reduce the bit depth)
Thanks Andrea. I hope this is my final question: Would I be better off in step one, the first export from Nuendo, to export my 48k 24 Bit original file as a 48k 32 FP file?
Then SRC to 44.1k 32 bit (no dither applied by r8brain only at 32 B FP).
Reimport to Nuendo, insert Elephant with 16 Bit dither, export again as 44.1k 16 bit?
Sorry to be so thick.
Yes, I think this is the right order. As a matter of fact, for me Audition works directly at 32 FP bits. I don't know yet about Nuendo (maybe next month...)
Don't worry about that William! Dithering is a subtle subject very often presented in an esotheric fashion. It took me a day or two to search, read, try and understand the pragmathic truth...
This topic was last updated 180 days ago, and thus it was archived. Replying is disabled for this topic.