Hello, I've been experimenting with the headphone-only binaural monitoring, and no matter what I try I hear a distinct early reflection that is mysteriously added across the whole mix, which is very annoying and forces me to disable binaural capabilities and instead fall back to conventional stereo...
Is there a setting that can be added to adjust the binarual rendering processing? It seems like something is 'added' that feels distracting more so than 'spatializing'. I'd very much like to delivery binarual re-renders driectly from the renderer, but at the moment I don't trust it.
Is there something I'm missing?
Have you experimented with the Binaural Render Mode settings? Near, Mid, Far, Off can be assigned differently for each object or bed channel.
Hi Fin, thank you for this quickly reply! Changing everything from Mid to Near did make a noticeable improvement.
I do wish that I could adjust each bed to its own setting - for instance the dialog bed keep near, the ambience bed keep mid or far. Looks like if I change one bed, they all must follow suit. Objects I can adjust independently.
I'm also noticing a loudness discrepancy - Loudness measurements through the renderer seems higher than in my session, which is confusing me somewhat, as the limiter in the renderer seems to be kicking in when it ought nought to.
I'd very much like to request some more limiter settings (attack / release etc.) in a future version (hopefully soon).
Any other tips you can offer would be greatly helpful, thank you.
Yes currently each bed channel has to be the same setting as other beds, we may be looking into ways we can improve this in the future.
How are you measuring loudness in your session? If you are returning a re-render into your DAW to use a third party tool then make sure set the re-render to be a loudness re-render in re-render properties.
The limiter in the renderer output is the designed to mimic the limiter in Dolby encoding tools, that way you get to hear your mix as the encoder will "hear" it, and therefore the end consumer. It is a soft clip limiter (without it you would be hard clipping your signal), it is designed as a safety net for the occasional peak rather than a creative tool. If you are hitting it constantly you should maybe consider pulling your mix back a little.
At the moment I'm not doing live re-renders, but I'll likely be setting that up next.
For an off-line example, if I want to export a binaural re-render of my full mix, I should choose 'Loudness' in the layout properties? I though I was to select "BIN" for Binaural... It appears I can't do both on the same re-render and I want to make sure my binaural mix is still on target.
For deliverables I've been using the Dolby Encoder Client, if I'm using the default Lo/Ro mix down settings in the renderer for the Master, must I also specific -3dB attenuation in the downmix settings of the encoder? Or will that in effect be down-mixing everything twice as much?
Thanks again for your help!
You should use the BIN for the Binaural re-render.
The Loudness re-render type is for using external loudness measurement tools to measure the 5.1 re-render with the limiter applied to meet Loudness targets in delivery specifications.
If you want to perform loudness measurement in the Renderer use the Analyze Loudness function. Once the analysis is complete you can get a measurement of Atmos (the 5.1 re-render with limiter applied) or the Binaural render if binaural is set in preferences.
For 5.1, LoRo determines how Lsr is mixed to Ls, and Rsr to Rs. When LoRo is set for Stereo it determines how the Ls is mixed into L, Rs into R, C in L/R. This is downstream of the 5.1 downmix so LoRo for Stereo works on Ls/Rs regardless of the method was used to derive them in the 5.1 downmix (Direct Render, LoRo, or PLiix).
The 5.1 downmix mode is written into the master file and used by Dolby Media Producer in determining how the 5.1 core of the Dolby Digital Plus JOC encode is created. The stereo downmix mode is likewise written to the master file and included in the encoded bitstream to provide instructions for how a decoder should downmix the 5.1 to stereo. The -3dB attenuation values for Center and Surround are used by default. It is not additive but can be used to override the -3dB default values if you feel this is needed for a given piece of content. But in the vast majority of use cases it can be left at -3dB. You aren't doubling attentuation values with this left at the default -3dB for Center and Surrounds.
Hope that makes sense.