Premium Membership - All Voxengo Plugins at a Fixed One-Time Fee - Click Here For More Info
Forums     Plugins     r8brain PRO Normalization in R8brain PRO simply won't work!

The updates look good.  I have a question with regard to the difference between setting "prevent clipping" to on, and setting "normalise peak dBFS" to 0: is that the former uses limiting, and the latter an overall gain change?

Prevent clipping won't normalize to 0 dBFS if the peak level is below 0 dBFS. 'Normalize peak dBFS' will normalize the output irregardless of the actual peak level.

Normalizing to peak values will NOT set the various levels so that they are the same apparent loudness - this is not how our ears work.

All it will do is raise everything to the same PEAK value, which may be a very brief transient.

Please, why the normalizing?

It is a very bad thing.

See, and read what Bob Katz has to say about it.

You are merely adding a layer of DSP to the file and introducing problems.

If the files are too quiet, raise the gain levels in a DAW or an editor before running SRC.

Multichannel Audio Specialists

DVD-A, DVD-V Authoring

Mixing & Mastering to most formats

Good suggestion, Neil.  It is also pretty reasonable to run conversion on files which peak at -6 dBFS - this way no additional step of normalization will be necessary.

Hmm, why not implement a simple 'gain' stage that measures RMS or some average sound energy levels according to a fletcher&munson curve or something? (like the C or A+C modes in Soniformer2) Then this could automatically try to make all the files sound about the same level.  I mean, what's the difference in doing this inside R8brain instead of rising the output fader manually in the DAW before bounce?  None I think, unless dithering was applied and the files bit depth is less than 32bit but who the heck bounces less than 32bit float if they want to samplerate convert? not me at least.. :P

It would of course take one file as a reference in the batch list and then match the others to that file, either by lowering or raising the volumes of each file.  The user should be given some options to specify the maximum output peak and minimum output level in dBFS, making a target 'bracket' like [min -8dBFS, max -0.3dBFS]

This would be very convenient and another selling point for r8brain.



Yes, this sounds reasonable.  At the same time I doubt this may be useful to many users as most engineers already have their workflow working.  They may not be interested in replacing their favorite tools with a new one.

Moreover, automatic RMS matching may not work good in all situations - average RMS is a too vague representation of tune's dynamics.

Yes, I think you're right.  I usually have my tunes at quite similar levels after the final bounce.  Maybe this still would be useful for converting big sample libraries from one sample rate to another and at the same time average the volume?

In either case this does not sound like a necessary function in r8brain PRO.  Thanks for the idea anyway.
This topic was last updated 180 days ago, and thus it was archived.  Replying is disabled for this topic.