In this final part of our series, we look at how the major project studio-oriented MIDI + Audio sequencers (DAWs) currently handle surround.
Over the past few months, we've seen how it's possible to use analogue or digital mixers designed for stereo mixing to mix 5.1-compatible audio instead, and spoken to various producers about how they've achieved it. However, in most cases, it is of course easier to tackle surround mixing in some kind of software-based mixing environment, as this can be customised to suit the number of channels you're working with far more easily than hardware. This month, for the final part of this series, we look at how the major project studio-oriented MIDI + Audio sequencers currently handle surround. Nearly all of these packages have proper, purpose-built surround support — only Cubase VST does not, and even then it's possible to fudge a way of making it work, though with less flexibility than the competition.
It's certainly possible to create a limited surround mix using Cubase, but since Steinberg want to differentiate it from Nuendo, there are no built-in features to help you. It's important not to be misled by the various third-party 3D panning plug-ins for Cubase, such as SpinAudio's 3D Panner, Zeep's LocaliserDSP, and Pixelsonic's Surrounder. These all provide an insert effect that can receive stereo audio, pan it anywhere between left, right, centre, and surround outputs, but they then use various surround-encoding processes such as HRTF (Head Related Transfer Function) for use with headphones, or Dolby Pro Logic for speakers, to reduce the output to a stereo signal. While genuinely useful for these applications, they don't let you route sounds to more than two output channels, and are therefore unsuitable for Dolby 5.1 surround mixing.
Nevertheless, we have come up with a scheme that will let you position each channel in any static position in a surround environment (see screenshot). With the default stereo output pair of your soundcard allocated to the Master Channel for front left/right duties, you first need to click the Active buttons for two additional stereo busses in the Master Mixer, and name these 'Rear' and 'Cntr/LFE' in the Master Mixer.
Next, rename an unwanted Group channel as 'unused', pull its faders right down, and then route the output of each and every channel required for surround work to this 'unused' Group. Having dispensed with the normal stereo routing, we can now utilise six of the eight available aux sends as individual level controls to the six surround busses. To do this, click their On buttons, and select Master L, Master R, Rear L, Rear R, Cntr/LFE L, and Cntr/LFE R in their Send Routing pop-up windows.
You can now position any channel anywhere you wish within the surround environment, as well as sending low-frequency special effect material to the Sub channel. Since the aux sends are being used post-fader, the channel faders still provide overall level control for each sound, and channel level automation will still work well, although the channel pan controls are redundant. Of course, you'll have to have access to an external multi-channel recorder capable of receiving at least six inputs simultaneously on which to record your surround mixes.
This approach is perfectly usable (if somewhat fiddly) for setting up static surround mixes, and it's even possible to implement surround panning if you're prepared to record multiple automation passes of the aux send controls. Thankfully, although the full price of Nuendo is £799, Cubase users can upgrade for a much lower £579 if they want to throw in the towel and adopt a more elegant approach!
In contrast to Cubase VST, Steinberg's Nuendo incorporated a particularly flexible implementation of surround sound right from version 1.0 (see screenshot below. You can define any number of output channels from two to eight in its VST Master Setup window, and then define their azimuth (angle) and radius (distance) settings. Presets are provided for stereo, quadraphonic, Standard 3/2 (similar to Dolby 5.1, but without the Sub channel), Dolby 5.1, 6.1, 7.1, and LRCS (Left/Right/Centre/Surround), and a graphic window shows how the surround channels are positioned.
Once defined, the appropriate number of output meters appear in the mixer Master channel, along with a single master fader. Each channel in the mixer can then be routed either to any stereo output buss, any of the defined surround channels, or to surround pan. Selecting the latter replaces the normal panpot graphic with a surround pan plug-in whose pointer can either be moved with your mouse, or by any cheap PC analogue games joystick. For more precise control, double-clicking launches a much larger Panner window with further options, such as position or angle mode display.
You can export an audio file in 'Multi-channel Split' and 'Multi-channel Interleaved' formats, as well as mono and stereo modes, but Nuendo also includes Matrix Encoder and Decoder plug-ins for converting to and from LRCS format, and for Pro Logic-compatible encoding. Provisions are built in to let you apply effects plug-ins to various combinations of channels when mixing in any surround mode.
The optional Nuendo Surround Plug-in Pack will let you apply global effects such as compression, loudness maximising, seven-band parametric EQ, and reverb across up to eight channels simultaneously, to provide precise control over the final 3D mix, while its LFE Splitter lets you create a subwoofer channel from an existing mix. The LFE Combiner, on the other hand, integrates a separate subwoofer channel back into the other surround channels. TC Works have also released the impressive Surroundverb for Nuendo which uses technology based on their System 6000 hardware surround processor. Martin Walker
Logic handles surround in a very straightforward and intuitive way, the only requirement being a soundcard or audio interface with at least six output channels. From Logic's Audio menu, the Surround page provides the means to select any of the currently standard surround formats from stereo and quad (up to 7.1), with or without centre speaker, though the most common format is 5.1. This window shows which outputs carry which surround channels and also shows the file name extensions that will be given to any bounced or mixer surround files.
The next task is to set up a surround mixer in the Environment, which at its simplest means a number of channels with surround panners plus six outputs (I used two stereo and two mono) to carry the six components of the surround mix (the Centre and LFE outputs were fed through the mono outputs). The LFE or Sub channel would normally be low-pass filtered at 120Hz for film work, and in Logic, this is achieved by inserting a suitable low-pass filter plug-in into the LFE Output insert point (an 18dB-per-octave or third-order filter is good for this). Your monitor system is fed from your soundcard's outputs, so if these are also feeding a mastering recorder, some kind of signal-splitting and level control system may be required.
Your individual mixer channels are first set to surround mode as shown in the screenshot above, then, by double-clicking on the surround pan control, a larger pan window opens (visible above the Mid and LFE faders) where a pull-down menu allows the user to pick the required surround format for the channel (I used 5.1 for all channels, but with the no-centre option selected for channel 7 and 8 — I'll tell you why in just a moment). By moving the virtual surround pan control (much like a joystick), the signal is distributed between the five surround channels, while an LFE control at the bottom of the surround panner window sends a proportion of the channel signal to the LFE/Sub output (6). If you need more control over what signals appears (or not) in the centre channel, you can set up the individual channels for no centre speaker, then configure a post-fade Aux to send any desired amount of the channel signal directly to the Centre output (5) via a Buss. This would allow the contribution of the centre speaker to be adjusted manually for each channel, independently of the surround pan control setting. In my screenshot, the last couple of channels have been set up in this way, the centre signal being routed to Output 5 via Buss 1. Note that as Buss outputs are stereo, the pan control is used to steer the signal to Output 5 only.
While controlling and mixing surround within Logic is easy, there's always the question of how to deal with external sources, such as MIDI controlled hardware instruments. The best solution is to record these sources as audio back into Logic. Once recorded into Logic, they may be treated as any other audio tracks and positioned within the surround environment accordingly. Similarly, virtual instruments behave in exactly the same way as audio tracks as far as surround is concerned.
Once a surround mix has been set up (and automated if required), it can be bounced to a new set of mixed files, which in the case of surround means six discrete audio files labelled as in the first Logic screenshot bottom left. These may then be reloaded into Logic for playback or imported into a suitable surround editing/authoring package ready for transfer to surround media such as DVD-A or DTS. It avoids complications if the sample rate of the original recording matches that of the intended delivery medium and that the recording bit-depth is equal to or greater than that of the final format. Paul White
A degree of confusion surrounds the issue (pun not intended) of suitable I/O hardware for PC users keen to try surround work, but suitable products do exist. Although Creative Labs' SB Live! has built-in DSP algorithms with surround capability, and does provide separate front and rear outputs, it doesn't let you access these separately from within a music application, so you can't use it for surround mixing — only for listening to specially encoded surround effects in games, for instance.
However, Creative's new Audigy soundcard does provide individual access to its Front L/R, Rear L/R, and Centre/LFE outputs when using its ASIO drivers, so you can use these to create a Dolby 5.1 audio mix. The card also has a built-in decoder to listen to existing Dolby 5.1 mixes. Terratec's DMX 6Fire soundcard (reviewed elsewhere in this issue) also helps the budding surround-sound musician, since its Control Panel utility provides a handy global fader controlling its Front L/R, Rear L/R, and Centre/LFE outputs. Creamware's Luna II is yet another soundcard that provides special surround features — this time a 16‑channel surround mixer — but you have to mix all your songs within its environment, rather than using your main sequencer application. Martin Walker
When MOTU set about adding surround mixing to Digital Performer they weren't messing around. The implementation in DP3 is superb — easy to set up and highly flexible. What's more, DP's Mix Mode facilities make converting stereo mixes to surround (and vice versa) straightforward.
The key to DP3's surround flexibility is the Audio Bundles window, which at first glance doesn't appear to have an awful lot to do with surround. But it's here, under the Output tab, that you can select and configure output 'bundles' ranging from mono and stereo, through LCRS, Quad and 5.1, right up to 10.2 format (!). Once these bundles are set up (and this can be done on a project-specific or global basis) they're available as output assignments in normal mono and stereo tracks.
The type of bundle you select for an individual track determines what panner is assigned to it in the Mixing Board. Select a mono bundle and you get no panner at all; stereo, and you get a conventional pan pot. If you select any bundle greater than stereo, though, a surround panner appears, with a representation of speaker placement, and an indicator running around the outside showing spread between the speakers. This Mixing Board panner is just the tip of the iceberg, though. Click the little button next to it and you open one of four 'specialist' panners that allow much more accurate imaging control. You can switch between these at any time, and every track can have its own type of panner.
Arc Panner is like a magnified version of the Mixing Board panner and additionally provides an image focus control. n-Panner is effectively two pan pots combined — one each for left/right and front/rear — and again it has a focus control. TriPan is similar except that it offers a 'three knob' mode which allows signals to be very easily panned in straight lines around the surround field. It also offers full control over divergence — the extent to which signals panned to one speaker 'bleed' into the others. Amongst other things, this lets you exclude the Centre channel from general imaging duties, freeing it up for use by other tracks. All these panners offer flexible stereo placement modes along with configurable low- and high-pass filters which help to manage the frequency content of the LFE channel.
The remaining panner, Auralizer, is a little different. Basically it combines surround panning with a fine-sounding reverb, and adds doppler effects and psychoacoustic processing for good measure. Used well, it's a jaw-dropper, and can create wonderfully believable cathedral-size simulations as well as unnerving 'fly buzzing around your head' effects!
DP3 has some plug-ins dedicated solely to surround mixing, such as Calibration, which invites you to feed it the output of an omni-directional mic placed at your listening position before proceeding to play a range of tones that are used to automatically balance the output levels of your monitor speakers. Then there's Bass Manager, which is dedicated to keeping the LFE channel under control and can help to prepare mixes for home theatre systems which lack dedicated subwoofers. There are also surround versions of many of the standard MAS plug-ins, including the superb Delay (shown bottom right), which offers separate delay and feedback parameters for each channel, and eVerb, which can handle any number of input and output channels that you want to throw at it.
Finally, DP3 makes provision for surround tracks, so you could, for example, bounce down your 5.1 mix to a single six-channel track on which you've placed the surround edition of the MasterWorks Limiter plug-in. Additionally, multi-channel tracks really facilitate editing, archiving and distribution of final surround mixes. Robin Bigwood
Purpose-built surround mixing came to Pro Tools Mix systems with the release of software version 5.1 — a happy concidence in the numbering scheme! Prior to this, many Pro Tools mix engineers involved in film and DVD production used the standard kludges for using a stereo mixer as detailed earlier in this series, and owners of systems that pre-date Pro Tools Mix still have to employ these (more on this in a moment). Two significant things set Pro Tools Mix and the new HD systems apart and make them fully-fledged surround mixing environments: the ability to set up multi-channel busses, and surround panners.
As on the other MIDI + Audio sequencers mentioned in this article, the variety of playback formats now available (from mono to 7.1) demands that the fundamental architecture of the mixer be flexible. In order to achieve this, Digidesign created the I/O Setup page. While most hardware mixers are hard-wired to provide mono and stereo input channels, and stereo mix busses (such as control room, group, and aux mix busses), I/O Setup allows you to configure your routing options from scratch. This can include a variety of inputs, inserts, output busses, and internal busses, which are anything up to eight channels 'wide'. It's easier to understand how this works by looking at something more familiar, so the screenshot shows the template for a normal stereo output structure in a Pro Tools system with eight audio interface channels. A graphical grid system maps the stereo mix busses on the left (which Pro Tools calls 'Paths') to the available physical outputs along the top. The paths declared in I/O Setup appear as destinations in the mixer, available to any track's output or aux send sections. In addition to these four stereo paths, we could also set up mono paths, linked to single outputs. With a routing template like this (probably the most common in use), if you wanted to route a mono track to, say, output 1, you could either set its output directly to 'A1', or set the output to 'A1-2' and set the stereo panner fully to the left.
The are just an extension of these ideas. The screenshot top right shows a well-specified output set up for use in the 5.1 format. There are two main paths: a six-channel path (5.1) mapped to outputs 1-6, and a stereo path using up the two remaining outputs. There are also a number of sub-paths, mapped to various relevant output groupings, such as Left/Right only. If you wanted to place a track in the Centre speaker, you could just set its output to Centre, but what if you need to place it in between the Left and Centre, or have it spread out across the whole front wall? This is where surround panners come in, and the track would be given one automatically if you set its output to Full 5.1. On a 5.1 panner, the green dot represents the position of the sound in the mix, while the blue square indicates its 'width' or 'spread'. Additional controls serve to feed the Sub, and attenuate the Centre channel.
As mentioned earlier, people with Pro Tools 24 and host-based systems built on Pro Tools LE as don't have access to these nice new surround facilities, but there are plenty of workarounds, such as that described by Martin Walker earlier in this article for Cubase. Space precludes me going into too much detail here, but my Pro Tools Notes column in SOS November 2001 tackles the subject in full. In summary, as in Cubase as described earlier, a combination of a track's main output and aux sends must be used for surround placement. For example, a vocal track might have its main output going to the centre, with a send feeding the Left and Right speakers to widen it. This works fine, although it's not as convenient as a surround panner, and it's difficult to achieve moving pan effects. Pro Tools 24 systems have a considerable advantage over host-based systems running Pro Tools LE, in that the former can run two third-party plug-ins which solve the problem. Smart Pan Pro, from Kind Of Loud, and the Dolby Surround plug-ins use clever behind-the-scenes trickery to provide surround panners. They both take a signal and distribute it to more than one of the existing stereo busses, achieving the same effect as the built-in panners in Mix or HD systems. Simon Price
And that brings us to the end of SOS's series on surround sound, although we'll be returning to the topic at some point in the future for some practical advice articles.
At the time of writing, although it seems clear that some form of surround is here to stay, the most important format for audio production remains stereo, and most recording gear is still designed with that in mind. Whether this balance will shift over the coming couple of years, or whether surround sound will merely continue to exist alongside stereo production as a specialist format is far from certain, even now. Arguments continue to rage over the best format for presenting surround, from the optimum number of speakers to the best playback medium, which makes choosing 'future-proof' surround gear a risky business at present. Nevertheless, one of the aims of this series was to show that if you are interested in working in surround, it's possible to attempt it without throwing away your entire studio and re-equipping from scratch — and hopefully, we've achieved that.