DAWS...THE NEW GUITAR?
Music has forever changed, and we've all seen it coming. Computer-manipulated sound has become a ubiquitous component in music production, and the sounds on many of today's records are so far removed from anything that could ever be duplicated in the "really-real world" that it's not even worth debating. And we're not talking about some obscure group of German nihilists banging away at MIDI triggers; pick up almost any album produced in the last two years that's cracked the Top 200 and ask yourself: Which instruments are real? What's been tweaked? And could that sound ever be reproduced live?
Whether something was sampled, stolen or actually played by a human is a question that the record store employees of the world can answer. How to reproduce those sounds live, however, is a real-world problem that hundreds of musicians already face. Imagine that you've spent however many months in endless Pro Tools sessions crafting your White Album, and it's time to play out. An even more likely scenario for Mix-reading live sound engineers is that you're handed 70 minutes of material that has never seen the light of day outside of a hard drive and told, "We're going on the road in six weeks, make sure it sounds right!"
I can already hear what so many of you are going to say: "A good band doesn't have to worry about this." "A good band will never go out of style." "These are the worries of the Spice Girls and the Depeche Modes of the world." But this is not only the future, it's also the present. The crowds are moving with the music. Hip hop and electronic music are now the dominant forms of music, especially in many urban centers, and the trend shows no sign of slowing. With established acts like David Bowie, Melissa Etheridge and Tori Amos incorporating computer-generated elements into their music, gigs that require the combination of sequenced and live material are no longer going to be an exception, they're quickly becoming the rule.
So, how do you do it? Do you dump everything sans vocals and lead instruments to DAT, press Play and hide? Do you painstakingly create custom sample and synth banks and pray that the bird's nest of MIDI and SCSI doesn't hiccup? How to choose among ADAT, DA-88, Mini Disk, Cubase, Logic, Digital Performer and others? To shed some light on the issue, Mix picked the brains and racks of some of the artists and producers who do this day in and day out. Special thanks to producer/engineer Ray "SoL Survivor" Cham, engineer/producer/keyboardist Charlie Clouser of Nine Inch Nails and artist/producer Q of Uberzone.
SEQUENCING THE TEEN POP MACHINE
Ray Cham is just as comfortable manning the faders on an SSL, tweaking audio on a laptop with a pair of headphones or helping to bring teen diva Christina Aguilera to the arenas of the world. Cham, who co-produced Aguilera's hit single "Come On Over" and last year's Christmas album, did all of the live programming and sequencing for Aguilera's most recent tour and worked closely with musical director Alex "Adrenaline" Alessandroni and executive producer Ron Fair.
Cham began by transferring the original digital and analog masters into Cubase. He chose Cubase because of the program's integration of audio and MIDI and, the ability to work in both environments simultaneously. "While we were still editing and arranging the show, certain edits were happening on a daily basis," Cham explains. "We ran DA-88s as backup, but I was running Cubase for the live show. I was actually mixing and making mix changes live. At one time in the beginning, I'd be running 60 to 70 tracks per song during a show, just because we're still getting mixes together and making certain arrangement changes. Once we had everything locked in, we did end up dumping it down and condensing the tracks down to either 8-track format or 16-track format."
Cham's live Cubase rig consists of a rackmount G3 350, a Motor Mix control surface and a MOTU 2408 that interfaced with a DA-88 and sometimes an ADAT. "I would have a backup show running concurrently on tape," Cham continues. "But everything would be on input. So if anything ever happened with the computer for any reason, all I have to do is throw the whole thing out of input and the DA-88s will continue to run. Once the computer goes down, you're running off the prerecorded mix, but I have time to reboot the computer and prepare for the next song. I can reboot without missing a beat, because everything is backed up in DA-88: the click, the background vocals, everything."
For time compression and pitch shifting, Cham has recently switched from Digital Performer to the new ProSonic Time Factory, despite the program's admittedly lengthy processing times. Cham also maintains an "arsenal," as he calls it, of different sequencing software. "I don't believe there is one program out there that is doing all things for all people," he concludes. "I know cats that are using Digital Performer and cats that swear by Logic and cats who swear by Cubase. It's all about personal preference. In my situation, I've found that all of them have different algorithms that they use for their pitching and time compression, and at different times they all work great. It's a matter of not being locked in and saying, `I only have one program, and I have to get it done as best I can with Cubase or Digital Performer.' At the end of the day nobody really cares how you get it done - just that you get it done."
THREE TRACKS THAT GO CLICK IN THE NIGHT
A long-time fixture in the NIN camp, as well as a renowned programmer/remixer, Charlie Clouser was instrumental in helping to bring the NIN opus The Fragile to the stage last year. In addition to a couple of co-writing credits on The Fragile, Clouser's recent credits include production and remix work with The Deftones, Rob Zombie and David Torn's SplatterCell, as well as contributing a track to the current NIN EP Things Falling Apart.
Clouser began by importing each of the multitrack files into Logic and dividing the tracks into groups. One group consists of possible tracks that guitar player Robin Finck would handle; another group might be assigned to the bassist Danny Lohner, and so forth. "And [then] there's what's not playable live," Clouser explains, "i.e. 16th note `typewriter' bass lines and 808 hi-hat parts that have crazy obviously machine-like bits to them." For the numerous keyboard parts, Clouser cut the multitracks into individual notes and exported them to an E4 sampler. The same process was used for a number of percussion elements that the drummer, Jerome Dillon, was intended to trigger live. The samples were dumped into an Alesis DM Pro drum module, which interfaced with triggers on the acoustic bass and snare, as well as an array of D-Drum pads.
For reliability reasons, the band uses DA-88s as a playback medium. "Back in `the day,' it used to be a Tascam 234 4-track cassette deck," Clouser remembers. "It was a rackmount Portastudio with a mono click, a mono bass and a stereo pair of synths and ambiences. And we've expanded on that just a tiny bit; it's still basically three tracks: a stereo pair and a mono bass - those things that are going to have ridiculous low frequencies wind up in mono on their own track - and then a track of click. And this time out we were more daring and put timecode on because we had had some video elements we wanted to trigger at specific points in a song."
The process of assembling the sequenced tracks and assigning parts to be played live began with the band's initial round of rehearsals last summer in the Bahamas. "I had been making a lot of the live tapes for The Fragile stuff on a portable rig as we were in rehearsals. In the morning, I would usually confer with the guys and say, `Here's three parts I've found for you to play, is that acceptable?' In many cases, there would be a bit of digital trickery involved because we find a lot live that you have to up the tempo by a few bpm. So then I would spend the morning doing my time compression and so forth and give the guys a really terrible version of the tape to play to. By the end of the day, after hacking through it a few times, we would have figured out the ultimate tempo."
The band then moved to full production rehearsals with the P.A. and a full crew, which included tape and trigger tech John Van Eaton. "We get the actual system down there," Clouser continues, "and we get an actual venue that's big enough to make it sound like it's actually going to sound." The band would then rehearse to either the DA-88 or Clouser's portable Logic rig, depending on how far along the band was with a certain song. Clouser's rig interfaced over separate ADAT bridges with a Yamaha 02R, which handled all the compression and EQ for the individual elements that would eventually end up on tape. The eight buses off the 02R would then be routed to the DA-88 for the final mix. "We can literally play along to the computer through the 02R. We can run though the song, and the FOH guy [John Lemon] can say, `there is some tambourine loop that comes in there in the middle that is so freaking loud.' And I'll go, `great' and walk over to the O2R and find that channel and pull it down and store that mix. And we would do that to very endless detail while we were in production rehearsals."
According to Clouser, another integral part of the process was creating a click track for the drummer to follow. "The drummer and I have worked out what his favorite sound is for the click and what sort of pattern he'd like it to play, because it's not just, `one, two, three, four.' But it has eighth notes and 16th notes in there. And there are little extra sounds; lets say there is a long break where there is some weird ambiance playing, so that he'll know when to come back in, there is, unfortunately, my voice counting him back in. And there are many songs where he's counting the band in, because we all start playing together and none of us are listening to a click except for the drummer. He follows the click we play to him. From our end, it's a very natural musical experience."
BEATS, BREAKS, DECKS AND SCSI
Recently picked up by Astralwerks, Uberzone is the brainchild of longtime artist/producer/remixer Q. Working out of his L.A.-based studio, Q has had a number of well-received, independently released UK and domestic dance singles. Q also remixed tracks by The Crystal Method, Sarah McLachlan and Herbie Hancock, in addition to picking up a co-production credit on the Afrika Bambaataa track, "2 Kool 4 Skool." Currently, Q is working on his debut effort for Astralwerks, slated for release this year.
"I approach [playing live] from the philosophy, first of all that I want to do something that is obviously recognizable and sounds like the original material but has a total fresh angle to it," Q states. "And then that emancipates me from having to figure out how to try and justify playing to backing tracks."
Q begins by making subgroup mixdowns of all the separate elements (drums, bass, synths, etc.) while he is still working in the studio. He then samples each of those elements and imports them into Acid for further tweaking. "You can take those subgroups," he explains, "put them in Acid and type in the bpm and take the sample, literally, and stretch it to whatever bpm you want to work at. Whatever bpm you're working at in Acid seems to really translate well to Cubase. So I'll take those samples that I've made in Acid that are perfect, perfect loops, and just export it to a new track and SCSI-dump it to the sampler. And if you write it in as a full bar loop, using Cubase, just drawing in a bar all the way across, it loops flawlessly."
Q uses the E-4 line of samplers and sequences live with Cubase running off of a laptop. "The laptop just becomes a sequencer," he continues. "I've got a simple MIDI interface on it. [The] E-4 acts as like ground zero for all the staple parts that are playing back lines and hooks that represent pieces of gear that I wasn't able to take with me. Like the Jupiter 6, all of its parts would be playing out of the E-4 now. A lot of the times you take a sound and sample it off a Jupiter 6, it kind of looses some of its impact. So I'll take the phrase and sample it. The E-4 is capable of like 128 Megs, so you really don't have any memory issues.
"Then I'm bringing a Jupiter 8000 and a Virus out as well to play parts that I'm able to tweak live. One of the things that I do, that makes the live performance more fun, is break it all out of the E-4 in stereo pairs of drums, synth parts, samples and that comes up in the mixer. I'll do it all in loops, and I'll do it so that I have sections of the song where I can move the locator points around in the sequence. I can extend a certain part of the song, and I can kind of do a live mix on-the-fly, using mutes, and I'll use the effect sends and send those out to time delays. So I can kind of play the mix. One of the other things I do a lot is use the trap-cap to trigger parts. So I'll have sounds in a second E-4 at the same bpm as the rest of the track, and I can trigger the loops off the trap-cap using the Lever function to start and stop breaks. Between that, being able to play the mix and playing keyboards parts gives me a lot of options and really involves you in what's going on onstage."
AN ABBREVIATED HISTORY
If there are any historical accounts of the use of prerecorded sound to augment live performances, then they likely start in the dawn of live radio drama. Broadcast sound effects artists would typically supplement live effects with playback from vinyl records, adding realism or, in the case of the legendary anarchic BBC radio show "The Goons," surreal absurdity.
Outside the controlled environment of a broadcast studio, playback from vinyl was relatively impractical, so it was not until the advent of tape machines that prerecorded sound cues became common in live theater. During my early years as an ASM and later as a sound designer, Ferrograph 2-track tape decks were the first choice for playback, later replaced by the ubiquitous Revox. Two-track cartridge players soon became popular for providing spot sound effects, underlining the fact that theatrical sound playback devices must provide two distinctly different types of cues: long multitrack sequences and "instant" cues, which are typically short. And, on some projects, the sound design may call for a multitrack player that needs to be triggered instantly.
Though the laserdisc format has almost followed the 8-track and DCC into oblivion, laserdisc players were once a common choice for theme park playback systems. Laserdiscs can play back more than two tracks of discrete audio, and the format's robustness recommended it over multitrack tape for productions that typically included 10-plus shows a day, every day. However, the turnaround time to burn a new laserdisc was at best 10 days - and it was expensive. This usually had the effect of locking in the show early in pre-production (sometimes a good thing). Spot sound effects were typically stored on 360 Systems and Alcorn McBride devices, to name but two.
The appearance of hard disk-based recording systems offered sound designers a new set of tools. I first used such a system - New England Digital's Post-Pro - 11 years ago for the Seigfried and Roy show in Las Vegas. The requirement was to play back all the prerecorded music on multiple channels, and, in fact, the system was capable of instant 8-track playback and stored the entire 90 minutes of show program in a towering rack of Patriot drives. This was considered quite a technical feat back then and came at an appropriately jaw-dropping cost - around $200,000. Backup was on MiniDisc. Ouch.
Other notable systems of the time included Doremi Labs' Macintosh-based DAWN system, an 8-track system with instant (buffered) playback. With Seagate hard drives and a lower price than the NED system (the New England Digital company later dissolved), the DAWN system became the new standard, and I specified it for the EFX show in Las Vegas.
By using two complete systems, it was possible to program A/B rolls from cue to cue with eight discrete tracks of playback; all the short sound effects were loaded into a Roland S-770. This show has now been running for over five years with minimal failures - backup is on a bunch of spare hard drives. Another Las Vegas spectacular, the Rio Hotel's Masquerade Parade, uses Wild Tracks from Level Control Systems. This has the ability to play back 16 tracks of audio with cue-by-cue mapping of track requirements (asynchronous) and an online backup hard drive.
Two-track MiniDisc players are a popular choice for relatively simple shows, in part because of their relatively low cost and ease of use and labeling. It is also possible to perform basic editing on MiniDiscs with minimal equipment. However, the MiniDisc format has only two tracks, only offers compressed audio (sometimes an issue) and will only play one sound at a time. On Broadway shows, prerecorded sequences typically include a click track for the musicians or dancers or both, so a multitrack player becomes necessary, because the click will eat up one or possibly two tracks. Akai's DR8 and DR16 units are among my frequent choices for multitrack playback, and the DR16 offers the potential for automatically triggering mix automation cues in order to re-route sources and cue outboard effects via SMTPE/MTC. Another unit I currently favor is the Akai 6000, which can be used for short sound cues and will also play back straight from the hard drive for lengthier cues.
Rereading this partial list of equipment that I have specified and used over the years, I am struck by how many pieces of gear have come and gone. Hardware obsolescence is almost inevitable, but there is no reason why the cues cannot be migrated to new hardware. Archive source material in .WAV or .AIFF data formats, which are easily transferred among most modern playback units. And, above all, make sure that you can safely and efficiently back up your work after you have finished programming.