Topic: Best Approach For Multitrack/Audience Pair Mixdown

[DavidPuddy]

Hi All,

Recently, I recorded a show and was able to grab full ISO tracks from the soundboard along with my audience cardioid set. I have opened the files multiple times to process them but honestly, I am super intimidated by the scale of this project - there are like 15-20 tracks in the soundboard ISOs. Can anyone share their methodology?

I've read that the drum tracks need to be mixed down separately until they sound good as a unit, but I have no idea where to go from there. For the drums (and other tracks as well), do I need to apply reverb to the standalone tracks or to the entire mixdown as a whole?

My audience recording is pretty chatty so I'd like to lean heavily on the ISO tracks. If anyone has a link for reading I would be very grateful. Thank you in advance!

[Chanher]

Do you already have multi-track software? It is a bit daunting, and IMO it's best to take your time and get everything set up properly FIRST, then you can get started adjusting and mastering different tracks etc 

The first thing to do is get your audience recording synced up with the ISO tracks. In multi-track software, place it underneath the snare ISO so you can visually line up snare hits. You will sync the audience track at the beginning of the show, but due to digital drift (they were recorded on different recorders using different word clocks) the audience recording will drift out of sync with the ISO tracks. You will need to use your software's "stretch" feature to lock the audience track in place at the beginning where you already synced it up, then go to the end of the show, where you will clearly see (and hear) it out of sync, and STRETCH the audience recording so one of the last snare hits is lined up again with the snare ISO.  Give a quick listen to see if sounds good, and as long as you lined up the snare hit peaks properly, you should be ready to go.

Stretching a source from a different recorder is a bit advanced but it's not impossible, once you do it a few times it will become easier. I use Reaper.

I personally prefer to get everything setup like this before doing any mastering. At this point, I also like to save A COPY of this multi-track session I can always go back to a blank slate.

Once it's time to start mastering, I prefer to start listening and mixing all the track LEVELS right away. For me I want to hear everything as a whole before deciding if something needs EQ, reverb, compression etc etc.

[roffels]

I started working with multitracks earlier this year, and just chiming in that I'd love to hear from the experts. On one of my first mixes, i relayed my concern that I really was out of my league, and his response was "there's a really low-bar with live recordings, most folks are expecting cell phone quality"- which, fair.

I haven't really wrapped my head around how much room sound to mix in with multitracks, because with my typical matrix, my room mics are providing whatever is out of balance on the board feed. With the multitracks, I get to set that balance, and the room mics are mostly there to provide ambiance. I like the control with multitracks, but miss the simplicity of a basic board/aud matrix.

[opsopcopolis]

Great advice from Chanher about lining up the AUD with the snare channel. I do the same thing.

There's no real right answer for multitrack processing, but happy to lay out my process though if it'll help. I've spent a lot of time the last few years working live multitracks (I mix nugs tapes for Tauk, Squeaky Feet, Underground Springhouse, and have done random other shows for Consider the Source, Fruit Bats, Zoomst, etc..) and I've basically come to use essentially the same process and template creation for all of them.

• Import tracks and place them in order low to high, rhythmic to melodic (drums, bass, guitars, keys, vocals)
• Set up and route AUX tracks for grouped instruments (drum kit, multiple guitars, keys, etc..), overall effects (verbs/delays) and master (if you're not familiar with how aux work, there's plenty of articles online
• Go through individual tracks for basic eq and compression. I wouldn't really recommend spending a ton of time on processing solo'd instruments, but it's good for getting the correct amount of compression and proper use of a high pass filter (you should HPF pretty much everything recorded with a mic that doesn't have actual low end content imo)
• Go through the auxes and do any necessary processing. I don't do much here, but a little bit of shelf EQ and overall compression is nice, particularly on the drums. Will also add a master EQ and Limiter
• Build a basic mix from the bottom up. Start by turning up kick, add snare, rest of the kit, overheads, panning as you go until you have a nice sounding/balanced drum mix. Continue with the rest of the band. Panning guitars/keys and getting everything to a decent, balanced level dry
• Add verb. For most standard live stuff, I like to do one band verb and one vocal verb. Usually a plate or small room for the band and a bigger room/hall for vocal. I'll put each verb on their own AUX, 100% wet, and then use sends on each of the tracks for their own levels of verb. For drums often will only put verb on snare and OH, sometimes toms
• Correct any eq/compression issues within the context of the full band. Sometimes need to carve out some space for low end stuff, maybe add a sidechain or something
• Once I'm at a spot where I'm happy with the overall tone and balance of the mix, I'll run through the show and automate levels as if I'm mixing the band live, bringing vocals in and out (especially if there's a lot of bleed), turning up solos, etc... Sometimes this requires actually listening through the entire show, sometimes (for simpler stuff or bands I know really well) I can just jump around
• If it's a band I'm not super used to, I'll often bounce the full thing and just skim through over the course of a few days and see if anything sticks out about the mix. For most shows these days I'll just split and bounce immediately. Just don't really have the time to sit on mixes anymore

It's obviously a very personal process, and also very dependent on the DAW you're using, but at the end of the day, the more time you spend with individual instrument tracks the more used to it you'll get. I don't think the normal process most tapers use for AUD/SBDs/Matricies really translates very well, so it's certainly a good thing to get some practice with if you're interested in it and can totally understand it being overwhelming the first time you open something up. There are sites around that you can grab sample multis from to mess with, or I'm not opposed to sending over some random old multis if you want to mess around with them.

[goodcooker]

I've done a few of these and my process is very similar to what others have mentioned so far.

I put my audience source at the top as track 1/2 then add everything else in this order

kick
snare
overheads
toms (and any other drum mics here like if there's some percussion mic somewhere)
bass
guitar
keyboards
vocals (each on it's own track)

Things I had to learn the hard way to look out for -

Keyboards were usually submixed before hitting the desk but sometimes each had it's own channel - ie piano, organ, synth were separate.

Mic on the amp and DI separate for guitars and basses. Put these next to each other and after you find the balance leave it. I usually did a low cut on the bass cabinet to leave some room for the kick drum but it's a balancing act. Most of the presence and attack from the bass came from the DI.

Even with gates on backup vocal mics I would still draw a volume envelope or automate to cut it out when not in use.

I learned to be lot more heavy handed with EQ when doing multitrack editing. It's kinda backwards from doing it for audience (and/or sbd) recordings where you need a light touch. With multis you are intentionally carving out space where other instruments need a place to sit. All instruments and voices have a range and you only need to open up the frequencies in that range. I hardly ever left any frequencies above 8kHz for anything except drums and vocals and often much lower than that.

Everything but the bass and kick drum (and synth if there's a lot of low frequency content) gets a low cut shelf and sometimes a high cut shelf as well. It increases intelligibility of vocals, makes guitars and keys cut through and get's rid of some of the low bleed that happens on stage with not much space between the mics assuming the FOH didn't do it already or maybe not enough.

I used to have one of these (not this one exactly but close) on the wall I'd refer to when dealing with some problematic tracks
https://www.amazon.com/Best-Music-Stuff-Ultimate-Frequency/dp/B07YQ9SW3K

I had one show in particular that had what felt like a million tracks. I ended up importing them all and cutting it into chunks and working on the 4-5 song per chunk pieces independently. I saved my template and added the new batch when I was done with the old.

[opsopcopolis]

I usually did a low cut on the bass cabinet to leave some room for the kick drum but it's a balancing act.

Sidechained/keyed compressor is a great solution for this. Send the kick out an empty bus, and use that bus as the key for a compressor on the bass. Set attack and release times to fit what you've done to the kick and you should be able to dial it in so you get the attack of the kick without having to compromise the overall tone of the bass. Can complement it with a bit of a cut in the bass at the fundamental freq of the kick too

[Gutbucket]

All good info above.

Taking a step back, consider early on how you want to approach it.  The factors influencing that decision are most likely to be how good your AUD mic source sounds, the expectations of the band or whomever your primary audience happens to be, and how much work you want to put into it.

The most basic decision will be the same as when making a simple 2ch AUD / 2ch SBD matrix - which source is going to be the primary one and which will be used in more of a secondary support role, AUD or SBD?  Personal philosophy and preference are drivers there, but often a bigger factor is a practical one - how good is the AUD and how good is the SBD?  If you have the full multitrack SBD, it's presumably good and useful, but the same can't always be said for a stereo or mono SBD patch. Sometimes it's very good, other times useless crap.

If your AUD is very good and you decide to use it as the primary source, that makes things easier.  In that case you won't necessarily need to use every channel of the multitrack SBD. You can instead just use the channels that are needed to improve whatever happens to be lacking in the AUD.   You essentially gain the ability to dial in your own stereo SBD patch which perfectly suits your AUD and might only need a few channels of the multitrack to shine.  I tend to prefer this approach.  Partly because I feel it tends to better translate the magic, liveness, sense of being there, stereo depth and image, audience, room.. all that goodness that typically only comes from a great stereo pair in the the sweet spot.. but also because it's simpler and a lot less work!  Of course it requires a good AUD, or stage-lip pair, or mics on stage, or whatever as the primary source, so this approach won't always be an option.

The other way is treating the multitrack SBD as primary and the AUD as a secondary "ambience and audience" type of source.  In that case you'll have more work cut out for you in creating a balanced SBD source that stands fully on its own, as described by the others who've posed above, to which you'll add your secondary AUD source for "ambient live flavoring".  If you know this is going to be the approach you will take before you record, it helps to setup your AUD mics differently than you otherwise would if they are to be the primary source in the mix.  You don't need to record good clear PA in your AUD source because you will have everything that was feeding the PA in the multitrack.  The AUD will actually be more useful when that stuff is minimized as much as possible and it instead focused on providing just audience reaction, room sound, and general ambience.  You then gain the ability to use more AUD in the multichannel soundboard mix when that's appropriate, because it doesn't primarily consist of PA sound which is essentially duplicating what you are creating in the multitrack mix.  Sort of the opposite of the typical audience taper approach where good clear PA articulation in the AUD is vital.  An arrangement such as a pair of mics facing out at the audience, or a spaced omni AUD pair can fill this role better than a directional AUD pair pointed at the PA.

[opsopcopolis]

Probably obvious, but I'll always opt for option 2 in those scenarios. For me, the aud is a secondary source, sometimes useful, sometimes not. I tend to find taper made mixes that feature the aud and a smattering of multitrack channels to be lacking in full arrangement and to be too lively in an attempt to make the aud the focus.

When I first went out on the road, I taped all my shows from FOH to use in the mixdown, but didn't end up using a single one.

[Gutbucket]

Sounds like the most appropriate approach for that scenario. Kudos to you and others here building mixes from the multitrack as primary source. That's a lot of work.  Setting up templates as you mention are a help in that regard.

I like the first method when I have a really good on stage / stage lip / up-front sweet spot recording to start out with that doesn't really need much added to it, yet the addition of just a touch of definition and clarity to whichever elements could use a bit of help really makes it shine.   Vox, keys, bass DI, a little more upfront snare crack, whatever.  Really nice to have that stuff separated out, as it allows me to pull up just a bit of whatever I want, without having it already mixed in with a bunch of other stuff I don't need that gets in the way.

I do think for that to work really well the sonic distance perspective between the AUD and SBD shouldn't be too different from each other, so the taper recording part of it needs to be pretty upfront, present, and not overly distant sounding. 

Deciding which way to go with it is essentially the first step of the mixing effort, since the approach from that point forward will be different even though the same basic elements are being used.  Same ingredients from the cupboard, different dish to the table.

[goodcooker]

He very clearly states in the OP that the audience recording is chatty and that he needs to lean heavily on the multis 

[Gutbucket]

He certainly does. Which suggests the second approach may be more appropriate in this particular case.

OP also states this-
"honestly, I am super intimidated by the scale of this project - there are like 15-20 tracks in the soundboard ISOs".

In a situation where the first approach is appropriate, he or anyone else coming across this thread can simplify things dramatically by exploring that route, which additionally preserves the depth and spatial imaging cues of a really good taper recording.

Horses for courses. Good to have options, and good to have this forum where folks can ask about and discuss them.

[Gutbucket]

You guys mentioned a few specific scenarios about applying gating, compression, EQ. 

Along that line of thought, I want to make it clear to other folks reading that in addition to multichannel SBD ISOs being a bunch of separated individual sources rather than a mixed stereo or mono output, those individual ISO channels will typically not have the same gating, compression and EQ applied to them that one would expect to find in a typical taper SBD patch, although it is possible some channels may have some of that applied prior to reaching the board.

The point is that a multichannel SBD ISO set is a lot more flexible than a typical taper SBD patch, but also a much rawer form of output that is likely to need more of that stuff applied to it to work well in the mix.

[Chanher]

It’s funny you guys are discussing this, because this past weekend I just did a multi-track live recording and for the first time tried pointing an ambience mic out towards the audience. I really like the results and, at least based on this one recording, prefer this method to mixing an actual audience recording (as we know them) with multi’s. It was a single at853rx hyper off to the side of the stage, pointed out towards the crowd. I’ve seen pro video guys use 2 mics (and 2 shotguns as well) in this manner, and I would have done it, but I only had one channel left on my recorder. I also ran a separate stereo recorder in the audience and also tried inserting it into the multitrack software to see what it sounded like, and I definitely preferred the single ambience mic. My least favorite variation was when I muted the single ambience mic AND the stereo audience recording, so it was just the close-mic’d instruments in the mix. It was a little dry and lifeless, but honestly not bad but it just didn’t sound like the actual show I had attended.

To stay on topic, I applied EQ and then compression to each track (guitar, bass, 1 drum mic, 1 kick mic, and keyboard). In the past I’ve used VERY minute amounts of reverb on some tracks in multi’s if they’re too dry, but I didn’t this time because the ambience mic kinda took care of that. I mixed the ambience mic pretty low in the mix, just enough to hear the room but not dominate the recording at all. The result is a very clear, borderline professional recording but with plenty of audience clapping and cheering. I’m very pleased, it was my first recording with the Zoom F6.

[opsopcopolis]

I'm also a big fan of the crowd facing ambiance mic. Does a better job at catching crowd interaction as well as the room tone without muddying the overall mix in the way a true AUD can

[Gutbucket]

I've included one or two rear facing mics in most of my rigs for a long time now, and almost always use some of that even without soundboard.   One rear facing channel works quite well.  I still use only one in my stealth setups.  Two adds ambient openness and is the obvious next thing to do when there is an extra channel available, but isn't really a 2X improvement over the first.  Best to place them where the audience is most engaged, so on stage facing the audience is the obvious place.  To simplify things I've integrated them into my main mic array that I place on stage or out in the audience, hopefully not too far back where the ambience can still be useful but the audience tends to be less engaged.  The primary thing is minimizing pickup of the PA and direct onstage sound into the rear facing mic or pair.

Most of the time I'm not using a whole lot of level from the rear facing channels, just enough to add a better sense of depth, room dimension and to make the audience reaction a lot more natural sounding and properly placed around the perceived listening position.  It doesn't really make the audience reaction louder, just more natural and properly placed in the stereo image, helping it to be less of a secondary annoyance while providing more of a sense of actually being there.  Sure, if the audience is distractedly talking I use even less of it so as not to emphasize that, maybe even none, but the right amount of it can actually make the distracted talking that's going to be on the recording anyway sound more like a natural part of being there rather than being just plan annoying, sort of getting it out of the way of the music to some extent.


When the soundboard elements are the foundation of the mix I lean more heavily on the rear facing microphones and less heavily on the forward facing mics, using that to add the necessary ambience once the soundboard parts are working well enough together.  The fundamental part of the mix is the balance of soundboard channels, to which I add dimensional ambience and sense of space from the microphone array, biased so as not to be focused so much on the front arriving sound.  This approach starts with direct sound placement and clarity > to which ambience is added.

When the foundation of the recording is my microphone array, I balance that first, listening for the most natural sounding overall portrayal of the room and overall the environmental envelopment, rather than for instrumental and vocal clarity, and then go on to reinforce whatever elements are weak with the soundboard elements.   This approach starts with natural environment > to which more direct sound clarity is added as needed.

When I don't have any soundboard and am relying entirely my mic array, the starting point is the same as before, but the primary focus is switched.  I first concentrate on achieving good front clarity, proximity, and image distribution using the forward facing mics (sort of like starting from the soundboard elements), and then support that with balanced ambience from the rear facing mics.  I'm still focused on achieving the most natural sounding portrayal, but start with a focus on good front clarity and image since I don't have SBD available to further support that, and the ambient part follows.

The end goal is the same, but the approach differs depending on what ingredients I have available and which are the most fresh.