[The Michael Dual Audio 720p Down
Rancul Ratha
It used to be that making music involved getting musicians together in a room, recording multiple tracks, and layering them together in a balanced and cohesive way. Now, as the world has become more virtual, the process is evolving. In this interview, Michael Lewis, owner of The Song Mill Studios, discusses how creating music has changed, the importance of bass in the soundstage, and why his dual SB-2000 Pro subwoofers have become an indispensable part of his monitoring system.
Later, as I got into my own music, I became a prog rock fan early on, then I got into jazz and studied jazz in high school and went on to study jazz in college. The music professor that I studied under became the director of jazz studies at UC Berkeley in California so he was a big inspiration too.
We want a nice balanced sound and as I said, an appropriate, industry standard level, because a big part of the job of the mastering engineer is to get the final product to where it would sounds good on a playlist with similar artists. Say you're mastering a country tune, and it's modern country. Well, that modern country song has to fit on the playlist with all the other current country artists that are out there. You don't want it to sound out of place or too quiet in comparison with the other songs on that modern country playlist, for example. Every genre has a particular sound that you need to aim for so you need to know what the current sound is for any given genre.
There's many projects that we work on where we're using players from all over the place, and I've done sessions for people as far away as Russia. And once we were all on lockdown last year, the whole idea of collaboration grew tremendously, but I don't know if people in the real world realize how much we had already been doing that.
The 110 channel mix and master we finished recently is for an album that's coming out for an artist by the name of Kika Morelan. We also finished a double Christmas album for her last year that features our Grammy Award winning session players for these projects. My son Jordan also filmed and edited an epic music video that went along with this song, you can check it out here.
In a normal year, our band One Street Over will probably go out and play live about 75 shows on average. 2019 was a typically busy year for us as we were playing a lot of cruise dates on the Colombia and snake rivers that travel between Oregon and Washington from the coast all the way up here to Idaho. We actually have a couple videos posted from our cruise performances, such as this one.
When using source footage that has multichannel audio, Premeire - and AME to an extent - refuses to allow proxies. It will claim to need every audio channel to match, which AME will not do even on the right settings. Proxies make everything simpler on the video end, not the audio, so why should it require every audio channel to match anyway?
My workaround has been to load the file into the free version of Davinci Resolve, set it to 720p, and export using their DNxHD settings that will match the audio source exactly. I then have to attach this file as a proxy to the original in Premiere. It works, but it slows me down a lot.
The current Proxy workflow requires matching audio channels. Some of the exporters in PPro/AME have limited audio channel support, which makes this challenging for Proxies. The QuickTime exporter supports the greatest amount of audio channel options, while also providing some good edit-friendly codecs such as DNxHR/HD, Apple ProRes and GoPro CineForm. These would be good options to use in PPro to create Proxies.
gotta say I spent weeks trying to make this work with some sort of sony footage with 8 tracks. Don't remember the exact format but it wasn't quicktime. Could never match a proxy setting to the source in terms of the audio. Just wasn't possible and I tried everything. Since the source material just had 2 channels of actual audio, I just transcoded to prores 422 hq with just 2 channels of audio and then made my proxies from that. Not ideal, but got me over the hump.
I'm in the same boat. Anything requiring multitrack audio means I have to go use an external, competing video editing software to render out a proxy file. Sometimes, that doesn't work either. I'm left with 2K footage that takes anywhere from 10-60 seconds to load on my timeline without a proxy. I'm also editing files that run in excess of four, five, and even six hours long in duration.
Is there any way this could be categorized as 'higher importance' because the workflow I currently have to do for this kind of thing is far more complicated than it ought to be. I'm sure myself, lanzamanza, mgrenadier, and a lot of other people who work like this would very much appreciate some kind of way to implement a better solution than the one that currently exists. It would save a lot of time, headaches, and frustration going forward.
Their system for getting things to the attention of the management is that UserVoice system. It goes directly to the engineering team, and yes, every post is read (I've talked with engineers that do so) ... and also every post is delivered in collated form to the upper managers who decide budgets.
I'm working in Premiere 14.3.0. I have 4k clips shot on Canon C300 with 4 channels of audio: Ch1 is boom, Ch2 is lav, and Ch3-4 are junk. My original clips appear in the Project panel as 4 Channel mapped to 4 Mono. I right-click on the clips to Create Proxies and choose the ProRes Medium setting. The first strange thing is that, as the proxies are created , the clips switch to 4 Channel mapped to Adaptive. This doesn't work for me, so I map them back to Mono, and cut them into a sequence.
When I play the sequence with proxies toggled on, the audio channels are properly isolated. When I play the sequence with proxies toggled off, they aren't. Specifically, I get what sounds like a blend of all channels when I solo A1. The rest of the channels are blank.
Here's what I wound up doing: I brought the clips into Resolve and transcoded them to ProRes 422 Proxy and "Same as Source" audio. I saved the clips to a separate Proxies folder. I used Mac OS's built-in Finder capability to rename all the proxies with "_Proxy" at the end. Then I batch-attached the original clips inside Premiere to the proxies. The audio now matches.
Thanks! I did get in touch with Jarle, who was very helpful. As it turns out, the ProRes proxy presets now automatically create proxies with a matching number of audio channels. (At least, they worked that way for my clips with 4 audio channels - not sure what would happen with 8.) So, farewell custom presets! I have continued to see the following bug: when trying to create proxies for clips brought in as multichannel mono, Premiere likes to change the first batch in the clip to adaptive. The fix is to detach the proxy, re-modify the audio channels of the clip, re-attach proxy. Sometimes you need to close the project and relaunch as well.
Premiere likes to change the first batch in the clip to adaptive. The fix is to detach the proxy, re-modify the audio channels of the clip, re-attach proxy. Sometimes you need to close the project and relaunch as well.
This fixed my issue perfectly! Coudnt understand why Adobe was interpreting footage with incorrect audio when every other software I used would show me my individual channels with their respective mics isolated. Thanks so much mi...@citizenfilm.org
Wow finally. I think this is the answer to my problems. Never realized my proxies actually had the right number of audio channels. One question though, in your instructions what do you mean by re-modify the audio channels of the clip? Sorry if I'm being daft.
Hi @benjamins19660537 , what the re-modifing is referring to is that when you batch modify the audio channels some of the clips don't actually change the way you told them to, so when you switch on proxies some clips suddenly display the wrong audio channels in the timeline, so the fix is to go back to the project panel and detach the proxies from those particular clips that aren't showing the correct audio channels, then modify the audio channels again to how you wanted and then reattach the proxy. For some reason it won't work while the proxy is attached. I'm not sure if this has been fixed with the latest updates or not, but this at least was an issue and sometimes you needed detach the proxy and restart Premiere, then modify the audio channel and reattach the proxy. Hope that makes sense!
Basically it extracts the audio of the original footage and attatch it to the Proxy, this is done very fast because it doesn't re-encodes anything, literally it takes half a second to execute the line of code and get the new file.
Whenver I'm away from home and get a new video, I connect throw remote desktop to my Desktop computer, download the files there and execute the "Extract the two audio tracks of the original Video" code for all the files. I then upload those MKV (Extracted audio) files to Frame.
On my laptop I download from Frame, the MKV files and then Frame has an option to download Proxies of the files that you have there, so instead of downloading the original Footage I download the Proxy that Frame.io provides. The proxies that Frame provides are 1/10 less heavy than the orignal files so this is perfect whenever I am in a place with bad internet. Also it helps a lot to not having to create the Proxies on your own.
To the proxy I downloaded, with the code "Remove audio from Proxy" I delete the audio track of the proxy, I do this because I didnt find a way of deleting the original track of the Proxy and injecting the new audio tracks in the same step. So if anyone knows how to do it please share it with me!
Now that I have the Proxy without audio and the audio tracks in a MKV file, I use the "Add the two audio tracks to the Proxy" line of code to inject the audio tracks into the Proxy. And just like that I have a Proxy file that has the original audio tracks of the footage. When I attach this proxy in premiere it works flawlessly.