Information accurate at the time of publishing - last edited 15 May 2020
My last blog was focused on loudness, and of course arguably NUGEN’s “flagship” product is VisLM, our timecode-locked loudness meter. But in close second place is our upmixing plug-in, Halo Upmix. So today’s blog is about upmixing!
More and more commonly, upmixing is becoming a key part of any audio post workflow. Whether we’re mixing various surround formats for a new project, or simply repurposing older archive material to meet modern standards. For me, it has been heartening to see so many Dolby Atmos-equipped studios installing Halo Upmix on their machines as standard.
Here are some of the biggest misconceptions I frequently hear about upmixing. As ever, please get in touch if you have any questions or comments.
Is upmixing "lazy"?
Commonly, I hear upmixing described as “lazy”. Ruy Garcia (Midsommar, Wendy) alluded to this in his recent NUGEN interview, and explained why he has now changed his mind. Previously, he thought there was “no creative decision-making” involved, and that upmixing was simply a set-and-forget operation, where you picked a preset and that was that.
Of course, in some situations you might want a set-and-forget operation, and there isn’t necessarily anything wrong with that! If you’ve got a stereo or 5.1 mix you’re already in love with, and it needs quickly repurposing for 7.1 or Atmos, then a preset might be exactly what you need. But there are creative decisions to be made, too.
A plug-in like Halo Upmix gives you precise control over how far and wide your audio gets dispersed, as well as control over placement of the vertical “energy” if you’re upmixing to a 3D surround format. Plus, it doesn’t always have to be as simple as placing an upmixer at the end of your signal chain after the entire project has been mixed. Plenty of engineers will upmix individual elements within a larger mix - a great example of this is in feature films which use classic rock or pop songs as part of their soundtrack, the whole film might be mixed in 5.1 or 7.1, but the mix engineer might only have access to a stereo master of the song. In those cases, the music alone could be upmixed on its own bus, in order to sit nicely within the full mix, without having to resort to simply panning it to the front Left and Right channels.
Does upmixing create artefacts and phase issues?
No-one wants nasty artefacts or phase issues in their audio. Especially if many viewers or listeners are likely to hear your project downmixed at home, rather than your full surround mix.
In the past, this certainly was a big problem! And that’s a big part of why we included delay compensation within Halo Downmix, in order to allow engineers to create better downmixes, without the phase cancellation caused by nasty artificial chorus and reverb involved in some legacy upmixing practices.
Halo Upmix does not use any artificial chorus or reverb - in fact, it doesn’t really add any extra audio. It simply takes what is already there, and uses its proprietary algorithm to reconstruct a realistic surround image. So if (or, more likely, when) the audio is downmixed at a later stage, there are no nasty surprises caused by phase smearing due to doubling up of audio. This is especially precise with the plug-in’s ‘Exact’ function enabled.
In 2018, Pro Tools Expert ran a blind upmixer shootout. Sadly, Halo Upmix didn’t win, but in his writeup Alan Sallabank noted that ours is the best upmixer of those tested in terms of downmix compatibility, and the only one of the three which upmixes to 7.1.2 and Ambisonics.