Pro Tools and Other Musings

Big Booty

From the very first conversations we had about the sound of the show, Arsenio Hall Musical Director Robin DiMaggio emphasized the desire of everyone involved for the show to have lots of good low end in the music.  “Big booty,” he said.  

I’ve been in broadcast audio for quite awhile.  Back before transmission was all digital, we tended to treat bass the way mastering engineers did vinyl--never let the low end get too big, but try to maintain the illusion that something big was down there.  Partially this was due to how easy it was to overload any part of the broadcast chain--a satellite transponder,  a microwave link, the preamp stage of the transmitter--and since every part of the chain had limitations and quirks, it was best just to keep the bass under control.  Bass overloads usually happened in an average-level rather than peak-level fashion, and the net effect was to crowd out everything else in the mix but the bass, especially since there were limiters protecting the signal at nearly every point in the chain.  The other reason to keep the low end under a tight rein was that the prevailing sense from producers was that nobody had a full range system on their home television anyways, which was largely true.  I do remember in 1978 hooking up the parent’s big stereo for the first broadcast of Battlestar Galactica, however, thinking it would be a Star Wars-type experience. Alas not.

Digital broadcast has changed a lot of things, and the old attitude toward bass isn’t valid any more.  Now, as long as you stay within whatever framework a particular network decrees about loudness (not the same as volume), you can do whatever you like with the mix.  And with the penetration of home theater in the general market, there are a surprising number of homes with subwoofers.

So, how to make the Arsenio Hall Show stand out when the Posse or any of the myriad guest artists are pumping?  Of course we go to the internet first and see how everyone else does it.  Dave Pensado, blessings be upon him, shows a method of thickening up synth bass tracks by duplicating the track with a 330Hz high pass on one track and a 300Hz low pass on the other.  The important part of this is that it allows the bass portion of the source to be treated on its own.  Dave uses the Waves Air plug and then a compressor set 6:1 with loose attack and release times to really focus the bass.

This approach works extremely well for synths, and with some tweaks can be used with other low-end sources.  For the purposes of television, however, I need the focus to be slightly different.  First, I want sub-bass information coming from instruments that may not actually have any.  Second, I want to optimize the signal to what the home viewer may be listening on.  The first item can be accomplished with any bass enhancement plug that uses some kind of synthesis or octave division.  The second, however, is a little more subjective.

Since I can’t control how a viewer sets up his subwoofer, I don’t worry too much about it.  I set mine with using the Blue Sky test files and a Radio Shack SPL meter, and I set the whole system to -20dbfs=78db at mix position and leave it there.  That’s loud enough to make me happy without making me deaf in the process.  So far so good, but what about the sub-less rest of the universe?  My mini speakers are an ancient pair of NS-10M Studios that have had the drivers replaced many times.  Truth be told, I’m not an NS-10 fan.  Don’t really love them, never have.  But they do have the virtue of telling one what something sounds like on a decent home speaker, albeit one with next to no bass response.

Screenshot 2014-02-14 13.01.02

Big Bottom set for High Booty

Now if I want to be able to make the subwoofer crank, I need to push subwoofer frequencies, right?  But what about the NS-10s?  Practically zero of that extra energy will make it to them.  So I have two aux sends available to every channel, called Low Booty and High Booty.  I twist Low Booty to make the subs thump, and I twist High Booty to wake up the NS-10s.

Screenshot 2014-02-14 13.01.02(3)

High Booty bandpass filter

Both sends end in aux inputs that feed the mix bus. Both inputs have the following chain: Aphex Big Bottom, Avid Pro Lim, and Avid 7 Band EQ.  If this sounds Avid heavy remember I set it all up in September of 2013 before the majority of third party 64-bit AAX plugs began to appear.  The difference between the two inputs is in the Big Bottom and the EQs.  Low Booty has Big Bottom tuned to 69Hz and the EQ HP at 48Hz and LP at 120Hz.  High Booty has Big Bottom tuned to 151Hz and the EQ HP at 67Hz and LP at 160Hz.  I set those frequencies by listening to the two different monitor systems with pink noise and then fine tuning them with music.  

Screenshot 2014-02-14 13.01.02(2)

Avid Pro Limiter High Booty Squish

The ProLim is set for Auto Release with the threshold set right down to -15.  I then selectively dial in various amounts of each send from the drum buss (both the clean and the parallel return), bass, percussion pads, and anything else that might add to the fun.  With the drums at first I experimented with just kick and floor tom going to the Booty busses, but putting the entire kit in there makes really wonderful things happen to the snare, which I did not expect.

IMG 1267

How it looks on the ICON.  The small "p" means the send is for the Posse,  Arsenio's house band.  There are a separate set of Booty sends for the Guest Artists.

The result is two inputs that add bass “feel” to the mix without overpowering it, while at the same time adding insane amounts of sub bass for anyone who likes that stuff.

Using Pro Tools and ICON for Live Mixing with Snapshots

For many years now live event mixers, FOH, Monitors, and Broadcast/Record, have relied on snapshot and scene recall automation to make shows go smoother and have performances as consistent as possible.  I've used lots of snapshot/scene systems, but the better ones all rely on some kind of combination of storing the present state of the console and then recalling that state on the fly.

Pro Tools, on the other hand, was not designed as a live mixing platform.  It is a powerful recording and mixing system meant for the recording studio and not the stage, and Digidesign recognized this when they rolled out the Venue series.  If you do happen to own an ICON, however, you already have a powerful live mixing system that you can set up to emulate a Yamaha or Venue.  Here's how:

1.  Gather all of the tracks that will be involved in a recall into a group--let's name it Guest for now.  Open Modify Groups (Control-Command-G in Mac) and uncheck Follow Globals.  In Attributes uncheck all the Main boxes (Volume, Mute, Pan, LFE). In the Mix Attributes sub window on the bottom check Record Enable, Input Monitoring, and Automation Mode.

Screen Shot 2014-02-14 at 3.34.05 PM

What you have done here is make a group that when enabled the faders and mute buttons are all independent, but changing automation modes, input switching, or record enable on one track will change all the tracks.

2.  If you haven't already, open Preferences/Mixing and enable Allow Latch Prime in Stop and Plug-In Controls Default to Auto-Enabled.  Make sure automation is enabled.  My personal preference here is to enable all the boxes in the Automation window except Mute since I frequently use mutes to audition different balances, and I don't want those moves stored.

3.  Make sure Guest is enabled (highlighted) in the GROUPS window, then toggle one member track into Latch mode.  All members of Guest should change to match.

4.  Set up your initial balances, inserts, send levels, etc. as usual.  Any touched fader will drop into active Write mode and the Latch light will flash.  It's not a bad idea here to occasionally hit Write To All, which will switch all the Latch lights from flashing back to steady.  As you adjust things, a quick glance at the flashing lights will tell you which tracks have been modified and which haven't.

4  In the GROUPS window click to the left of Guest   All members of Guest should now show that they are selected.

Screen Shot 2014-02-14 at 3.38.12 PM

Click on the black dot to the left of the group to Select all members of the group.

5.  Press Snap on the ICON soft key pad to open the Snapshot buttons.

6.  When you have a starting balance, hold Shift/Option on the keyboard and press the CAPTRE soft key.  What you have just done is to write all selected channels into the snapshot buffer--note that only the channels that are selected go into the buffer--any tracks that aren't members of Guest will not go into the buffer, and hence, will not be subject to a future recall.  Pick one of the soft keys below, and press and hold the key until the window says Stored.  (If you're in Select mode and not Focus you can now double-press the soft key you just stored to open a naming dialogue for that snapshot.)

7.  As you modify your balances and settings, keep overwriting that soft key by loading the buffer with the Shift/Option/CAPTRE combination and press-hold on the target soft key.

8.  When you need to store a different snap, choose a new soft key.  There are four available on the first page and another 44 available on succeeding pages.

9.  When you're ready to recall a snap, press the soft key of the scene you want (which pops it into the buffer) and PUNCH CAPTRE to execute the recall.  You will note that all tracks that are members of Guest will go into active Write mode instantly, and tracks that aren't members of Guest aren't affected at all.  This behavior will not be affected by whatever channels you happen to have selected when you recall the snapshot.  The concept to get here is that the channels that are selected when you store are the channels that will be affected when you recall.

You now have a system that can reliably store and recall 48 snapshots on the fly.  I know it seems like a lot of keystrokes, but really, once you have things set up the important moves are the left-click to select all members of the group and the key combinations to store the selected tracks into the buffer and subsequently the soft key.  Note that you can have more than one SCENE group--I use this to have separate recalls for the house band and guest artists on Arsenio.

I would avoid using VCA Masters when doing this.  The Coalesce options can make the behavior of slave channels somewhat unpredictable on recalls.  The exception is VCA groups where you never change the internal balances (FX returns, for example).  You can emulate VCA spill by making Custom Fader layouts for any logical groupings (drums, strings, horns, etc.) with only the members of the group in the layout.  Then assign the members of the group to a bus and re-enter the bus on an Aux Input on the main layout (also a great way to pop a limiter across a bunch of like instruments).

Keep in mind that this is an automation-only recall, and as such has no effect on bus assigns and plug-in selection/ordering.  If you need a different snaps to use different bus paths, all needed paths must exist throughout the whole session.  If you need different plug-ins at different times they all need to be enabled throughout the entire session and brought in and out with mutes or fader moves.

One last tip--get into the habit of hitting Write to All until you actually roll the transport for the first time.  From that point on, use Write to End to save all your previous moves.  Once you're finished, you now have all your original live moves intact and you can use them as the basis for quick touch-ups to the mix.

Good luck,


A Qualified Panegyric

As the process of innovation gets ever faster, I’ve noticed that combinations of workflow and gear now resemble snapshots more than portraits.  When I was at Post Logic years ago we had rooms that didn’t substantially change for some time-- a big analog desk with fader automation, and an Adams-Smith AV system controlling various 24 tracks, DATS, video machines, and the odd 2 track.  Certainly we tried to innovate as far as the technology would allow--it was a big deal when we put all the sound effects CDs in a central jukebox with a Mac in each room to search and audition, for example--but a picture of the room taken in the late 80s would have been substantially the same five years later.

What a difference a few decades makes.

These days, a workflow/gear combination may go roaring past me with only a fleeting glimpse before being swallowed forever by progress.  So I want to take a moment and talk about the present workflow “snapshot”.  Some things that have been around for a while have been joined by some newer things, and some older things have seen some dramatic improvements.  Since this is largely a love letter to the D-Control, I’m putting it here, but it involves a variety of other gear not necessarily limited to the Avid marque.

Here are some highlights from the present snapshot:

Blueface 32 fader D-Control

Pro Tools 11HD

Pro Tools HDX|3 talking to Focusrite Rednet 5’s and HD I/Os

Aphex Aural Exciter and Big Bottom

Avid Pro-Lim, Reverb One, Revibe

Sources from the stage on Yamaha Rio 3224Ds via the Dante network

And here’s my workflow:

In the morning, set up and sound check a guest artist of some type.  Could be hip-hop, could be rock, could be blues, could be pop, could be latin, could be anything.  Anywhere between 10 and 60 inputs.  After that, rehearse with the house band (48-58 inputs depending) setting the show order and possibly recording some new cues.  Then camera blocking, and if time is available I may get to play back the last camera pass of the guest artist to get it as dialed in as I can.  Then and at some point we do the show.  Depending on time available before the network feed, maybe some quick touch-ups to the musical performances, but mostly it airs as it went down.

Here are the things that make coming to work fun.

First, as I said before, the D-Control.  Being able to have a dozen different Custom Fader setups a single button away is really what makes this whole thing possible.  I set all 32 faders for Custom, with drums on one setup and the rest of the band on another (the drum submaster lands on the band setup).  It keeps things calm on the active layer while making a quick look at the kit for a touchup simple.  Multiply that by two bands, and you get an idea of why it’s important to me.  And when last minute things get thrown my way, after a quick layer edit I’m ready to go.  I keep another layer with just masters on it for switching stems in and out of record and selecting which busses feed the external meters.  The surface is also very good at letting you know what the automation is doing: the status lights on the channels really yell out what state they are in, and since my workflow involves hitting Write To All during rehearsals and Write To End when recording, having those buttons dedicated on the surface is really useful.

I also find D-Control amazingly responsive.  I know it’s just a big mouse, but it reacts instantly and doesn’t fight me.  Some other digital desks I know seem to love to fight.

Up top I mentioned this observation was “qualified”.  Of course, the biggest irritation for me is Avid abandoning the platform just as I'm starting to fall in love with it.  I understand corporations have to chart their own courses--I just wish they had done a redesign/relaunch of ICON instead of abandoning it for the S6 project.

And of course I share the usual gripes about the surface that have been with us since it came out.  The meters are strange, no way around it.  It’s always a shock to go from the meters on the desk to the ones on the Mix screen, especially now that we have such a great selection of meter ballistics in 11.  My personal grump is the control spacing--the console could have been probably a third smaller without sacrificing utility (which would also have made the top row of encoders a little more reachable by standard-sized arms).  I’m not as unhappy with the scribble strips as some others seem to be--they’re a reminder of what the state of the art was ten years ago, but still an enormous source of critical information while working.

Now that I’ve mentioned Pro Tools 11 I can list a few things about it that make my present workflow possible.  First of all, in the configuration I have, it’s much more stable than any other Pro Tools version I’ve ever used.  There was release a few years back that would drop record during concerts, which was not good for my blood pressure.  11 feels completely different in this regard from its predecessors: the spikes are gone from the system status bars, and while this may just be the effect of re-writing the graphics software it certainly doesn’t feel that way.  Taken together this means a reduction in much of the pucker factor when mixing in the box.

Second is the Offline Bounce capability.  Since most of my work involves song-sized chunks I guess I could use the regular Bounce to Disk function and not give up too much, but I always seem to be in situations where minutes count.  The fastest I’ve seen Offline render a selection that involves quite a few plugs, re-entry paths and a bit of automation is around 4-5 times real time, and mostly it’s slower.  But it works, except for one strange bug (11.0.2).  If your session start time is anything other than 00:00:00;00 (I work in television and therefore drop frame) the original time stamp on any bounce you make will be offset by exactly the difference between your session start and 00:00:00;00.  Annoying, but the workaround is fairly simple: keep your session start time all zeros.  Although for me this means the Session Start time is now 15 hours away from the first recorded material, it also means that any bounces I send over to editorial will fall in to their timeline frame-accurate to the camera master time (which saves them a lot of grief spotting).  I hope Avid fixes this soon.

Third is the ability to automate while recording.  This may not sound terribly important, but it’s a huge timesaver.  Now we all know that the automation in Pro Tools is some of the simplest and at the same time deepest ever written for mixing.  Everyone who was big into SSL or Flying Faders automation, raise your hands: automation used to be really scary, and only those who have seen 60 faders suddenly fly to the wrong position after an hour of careful balancing because you missed one button push know what I mean.  Not anymore--I never even bother to turn automation off these days.  The PT automation suite is really the only one that an operator can start using right away and still be confident the automation isn’t out to ambush him or her.

Those advantages really come together when doing music for television.  When a show has a daily delivery, the feed to the network very often starts an hour or less after production finishes.  In our case, it’s always less,  sometimes 30 minutes or less.  Let’s consider the case where something needs to be tweaked in that time.  If you weren’t able to automate while recording you would either have to rebuild the entire mix very quickly or just punch in spots that really need fixing.  If you wanted to make a minor change for one instrument or vocal across the whole song you would very quickly run out of time, losing whatever good moves you made during the live performance.  With the ability to automate while recording your moves are all safe and you can pick and choose which elements require touching up.  When you have the remix finished, you can quickly select that area, do an offline bounce, and 11 will deliver it to editorial in a third the time it takes to play it out (all the while miraculously saving all the reverb tails, compression ballistics, etc).  My favorite application of this is when we do hip-hop and the artist uses language that Standards and Practices is unhappy with.  When this happens I just select the entire performance (with sufficient handles for the editors), mute the channels that may contain questionable language, label the bounce “Song no vocal” and send it on.  Editorial can then quickly drop it on the timeline in sync, and wherever the lawyers object just swap from the line mix to the remix for a few frames.  It means the show isn’t forced into silence, tone, or some other substitute sound and the song doesn’t get interrupted.

Lastly, let’s talk about sonics.  I don’t consider myself a “golden ears” type, but even I can hear the difference between HDX and TDM.  The amount of detail in the audio now is startling.  Overhead mics now actually reproduce cymbal sounds and not the ripping paper sound we’ve all put up with for so long.  For better or worse, things actually sound more like their sources now than they used to.  I don’t know how they did it, but the sonics are dramatically better. 

Once again I’ve taken way too many column inches to say something simple.  I’m grateful for the combination of tools and toys available now that make mixing easier and more fun.  Which is what it’s always supposed to be, isn’t it?


A Little Love for the BeyerDynamic M 160

I guess I've always harbored a deep and unreasonable suspicion of Beyer ribbons because a place I worked at long ago had an M500 (maybe two?) that always seemed to be broken.  After multiple repairs it never seemed to last more than a month before something fell off or the ribbon gave up.  Looking back I'm sure this has more to do with us not treating the thing with some respect, but it was a long time ago.  Maybe we stuck it in front of the kick too much.

Flash forward 25 years or so, and while mixing music for the new Arsenio Hall Show I'm getting requests from three separate acts in the last few weeks for an M 160 on something--guitar, kick, whatever.  Then I get the advance info for the Wayne Shorter Quartet with Esperanza Spalding, and Wayne lists two mics he likes to see these days.  The first is a Soundelux U99, the second, you guessed it, was an M 160.  (The U99 is a little huge for TV and besides I don't know anyone who owns one, much less rents one.)  This is now a trend, I think, and place a call to Ron Cheney over at RSPE to get an M 160.  It arrives the morning of the day Wayne and Esperanza are scheduled, and we literally took it brand new out of the box and put it on Wayne's stand.

Kind of revelatory. Soprano sax is a tricky instrument to reproduce--it can get honky or screechy pretty easily--and when Wayne, who is a master, living legend, sax god of the highest order, started playing it was creamy and warm and still had enough character to stand up in a dense mix with almost no EQ.  They're not cheap by any stretch--street price around $600+, but it's probably the only ribbon I know of that sounds great and has enough pattern control to make it useable on a noisy stage.

So I hereby admit my prejudice was unfounded and give an unqualified endorsement of the M 160.  At some point I'm going to bring the sucker home and try playing some trombone into it.  Not that I'm any good, but I am curious how well that character--slightly forward for a ribbon--will translate for brass.

The performance was amazing--very outside and something of a challenge for mainstream television.  I loved it.  And I got to meet Esperanza!  Let me know what you think of Wayne's sound, if you feel like it.  Thanks again to Ron and RSPE for the hustle.