Future Proofing Your Sessions
I was having a conversation on Facebook with some friends about the current trend in subscription based software. Most of you are probably familiar with Adobe’s method of essentially “renting” the software from month to month and now Avid has given users the option of paying month to month as well. My initial perceptions of a subscription based model were not favorable. But the more I thought about it, the more I got used to the idea. It started to seem like a fairly harmless thing and in some ways (constant updates and bug fixes for example) quite advantageous. In this conversation, I explained that just as I was starting to see a few benefits in the subscription model of software, I thought of one major flaw that keeps me from ever jumping on board.
Lets say you have a big client that worked with you 5, 10, or even 15 years ago. They come back and need the files from an old project. You cant open that project; not because you don’t have the old computer or the files, but because the software company whose program you were using, now prevents you from even OPENING the software that opens those files. It could be because of licencing, or they dont want to support it anymore, or they shut down the authorization server. The point is, you have no way to open that file because the software no longer “exists”.
Currently, I can open old pro tools files going back 15+ years because we OWN the software perpetually and have an old computer running a version going back to Pro Tools 6.
But what if your DAW of choice went to a subscription model and years from now and then went out of business and you find yourself in a situation where you can’t open your old DAW files because you can’t even open your DAW software.
What To To
Well let’s start with the obvious first. Throw out any notion of retaining your session document, EDL, OMF, AAF, or any sort of file that is a pointer to where audio belongs on a timeline, what your mix settings are, plugins, etc.
We’re going under the assumption that THIS is there very file that will no longer work in the future.
So where does that leave us? We have the AUDIO….that’s all.
At the very least, I hope you are already bouncing a full quality WAV/AIF file of the mix at the native bit rate and sample rate of the sessions. So even if the client has requested to ONLY be delivered an mp3, you should still be bouncing a WAV or AIF. If you recorded the session at 24bit/96kHz, then you need to bounce a 24/96 WAV file of the FINAL mix. If you recorded at 44.1, then bounce a version at 44.1. That is the BARE MINIMUM of what should be done.
But lets also address some other things you should probably be doing already.
1. Instrumental bounce
2. A Cappella bounce
3. Vocal UP version
4. Vocal DOWN version
5. Any format the client may possibly request: WAV, mp3 at 320, mp3 at 192, etc….
6. If it’s a film, commercial, etc, stems are an absolute MUST. So you should be doing separate stems of voice, music, nat sound, effects at the very least.
Those are all pretty easy. You just have to bounce a couple extra versions….not a lot of additional time invested.
But let’s say you want to retain the “mix” the tracked out version in case something needs to change later on.
Going Above and Beyond
First things first, you should export all of the tracks UNMIXED. Just the raw files. But here’s the important part!
You need to make sure that you export/bounce all of the individual tracks from the same start point. Otherwise you’re in for a world of hurt when you have to manually line up all the elements on the timeline. No one wants that. So do yourself a favor and consolidate all of your files from the beginning of the timeline or grid line.
By exporting everything UNMIXED you ensure that you have a pure starting point if something needs to be remixed. And it DOES happen. Just like Hollywood like to to reboots, some artists like to do remixes of classic records, or surround versions. Pearl Jam’s re-release of Ten comes to mind. But something can only be properly remixed if we have the RAW source material.
But here’s another scenario, let’s say the client wants to retain most of the original mix, but make an alternate version. When Guitar Hero and Rock Band came out, many artists had to go back to their original master recordings and make new versions with stems of drums, guitar, bass, keys, and vocals. We can make this process MUCH easier by printing all the processors like EQ, compression, etc that belong to that track.
So what do we do in this instance? Well, we need to bounce or export that track WITH all the effects included. So if you have a chain of EQ->compression->limiter for example. The track needs to be bounced exactly as you hear it and at EXACTLY the same fader level as you have it in your mixer. This process is A LOT easier if you have a DAW with track freeze, otherwise you have to bounce to a new track in real time.
Another good idea is to print the click track (if you used one) and put the tempo in the file name of the click. For example: “Click 125bpm.wav”
For most of us, things will rarely be that simple. You have things like summing mixers, parallel compression, mix bus compression, external hardware inserts, virtual instruments, and many other issues to deal with. Let’s start with the easy ones first.
1. With hardware inserts and summing mixers, you can’t simply use track freeze or export for obvious reasons. We need to run audio in real time through the hardware, so that means busing the output of our track to a new track and recording the output. Unfortunately this makes for a much longer archival process.
2. For virtual instruments the same basic principle applies as the one above. Some instruments don’t always respond the same way on each play or have the ability to render the track, so you will need to route the audio from the virtual instrument to a new track and record the output.
3. Mix bus processing is where things get a little more tricky. Now, you could take the approach of just leaving it off and maybe writing down the settings of what you have on the 2 bus, or you can process each track THROUGH the mix bus chain when writing new audio files. This is much easier for things like EQ. But for compression things get hairy. First of all, you can’t just mute the other instruments and expect the compressor to react the same way. The 2 bus compressor is working based on the sum of the mix. You COULD do something like bounce the whole mix, send that bus to the side chain and have the compressor react to the sidechain while bouncing your tracks. But you run into the problem of the side chain only reacting to a MONO source rather than a stereo mix. I’ve always found that doing this never quite results in the same mix when I play the new files back. In my opinion, i would just mute the 2 bus compressor, write down the settings and hopefully when it comes time to pull the mix back up you can find something similar or better.
The point to drive home here is that if you are relying heavily on mix bus processing, to iron out your mix, you need to AT LEAST write down whats on it. Sometimes you can open a bigger can of worms than its worth by trying to export your tracks with the mix bus processing engaged.
The End of the Process
Ok, so lets say you’ve got all of your tracks WITH signal processing ready to go, all bounced from the beginning of the timeline, so they line up just fine when you pull them in. Now what? Well, I would suggest putting all of those files in a folder and then start a text document in a UNIVERSAL format like .rtf or .txt format and write down any notes that you think my be relevant if some random person had to pull up these files 30 years from now.
If you pull of these files should they all sum to the final mix, do you need additional processing, are there any problems, what’s the signal chain of these files, how many tracks should there be, WHAT are the tracks that should be present, etc.
There are many additional things to discuss and I may make an addendum to this at a later date. In the mean time. FUTURE PROOF YOURSELF!