Version 6 of DP comes with improved hosting of Audio Unit plug-ins — but what does that mean for the MAS format? We consider the issues, and bring you a guide to DP's new bundled convolution reverb, ProVerb.
In DP6, compatible AU plug-ins have working side-chain inputs (as shown here on WaveArts' TrackPlug) and fully‑featured automation. The reasons to use MAS versions instead are disappearing fast.
One of the more unsung features of the 2008 update of DP to version 6 has been its improved Audio Unit plug-in hosting. Of course, you've been able to use Audio Units in all the OS X versions of DP, and many users have done so for years with no difficulty whatsoever. But in DP4 and DP5 there are limitations: there's no way to access side‑chain inputs on plug‑ins such as compressors or gates that have them, it's impossible to automate a smooth value change (a so-called 'ramp') for continuous‑type parameters, and AU instruments don't benefit from the same super-tight timing as their MAS-format counterparts (although it's still pretty good!). These shortcomings have been addressed in DP6, and consequently the question arises: why bother with MAS-format plug-ins at all any more?
Some DP Plug-in History
The MAS (MOTU Audio System) plug-in format dates back to the days before OS X, when the only other format we Digital Performer users had to concern ourselves with was VST. Back then, developing MAS plug-ins made a lot of sense to many developers, as it meant that DP users could buy their products without also needing a third-party VST host (which at the time was either VST Wrapper or Pluggo). Native Instruments, Applied Acoustics, IK Multimedia, Antares, PSP, Arturia and GMedia all developed MAS plug-ins in those days.
Audio Unit support dates back to DP4.1, released in Autumn 2003, and from that moment on, third-party MAS support dwindled, presumably since the MAS market was comparatively small and DP users could, in most cases, be catered for very well with AU. And so we reach the present day, where about the only companies now continuing to produce MAS plug-ins are MOTU themselves, UVI, Audioease, iZotope and WaveArts. Even Waves, one of the most long-standing MAS developers, have dropped the format with the recent release of their version 6 software.
Where does all this leave us? Presumably it's only a matter of time until the remaining MAS developers also commit to the Audio Unit standard, since DP6 has got so good at handling it. And that being the case, the days when you'd specifically seek out a MAS plug-in over an equivalent Audio Unit seem to be coming to an end. Having a MAS‑format plug‑in installed for DP as well as an Audio Unit for all other audio software is a complication most of us could do without, and can even be confusing, since DP 'sees' both. Even MOTU's own 'extra cost' instruments, such as MX4, MachFive 2 and Ethno, work really well as Audio Units, and are no longer automatically disabled in DP6.
Now, if you're thinking at this point 'what about project compatibility?', you have a point. As a general rule, if you try to open a project that had used a specific set of MAS plug-ins when those plug-ins are no longer installed, you get a 'missing plug-in' dialogue box, and a devil of a job trying to reconstruct the settings and sound of your project. But there is, apparently, a mechanism that plug-in developers can use to automatically replace missing MAS plug-ins with AUs — and it's already active and working effectively in Waves v6. If other developers chose to implement this, they could save themselves the effort of producing further MAS plug‑ins, but not inconvenience DP users in the process.
MAS Exodus?
So will we see MAS disappear? I don't think so. It's almost bound to continue as the format for DP's bundled plug-ins, to prevent their use in other applications, and quite probably for MOTU's other instruments too, for obvious reasons. And we might find third-party support hanging on for a while, after all, to cater for users of DP4 and 5. Looking to the future, though, what should you do about the AU/MAS situation? That sort of depends on whether you've upgraded to DP6 yet. If not, MAS plug-ins will continue to give you the best performance. If you have, though, now could be the time to install and enable AU equivalents of all your favourite MAS processors, and start using them in preference to MAS in new projects. It's a transitional time (we Mac users are used to those by now...) and nothing's certain, but this is definitely worth thinking about. And, if anything, it reminds us all of the importance of archiving projects carefully, perhaps bouncing or freezing all plug-in tracks before finally putting the project to bed, so that if you ever have to open it again it'll work independently of whatever has happened to your plug-ins in the meantime. I'll be keeping an eye on further MAS developments in the coming months, so stay tuned.
DP6 Upgrade Guide: ProVerb
They say it's a wise man who knows his proverbs, so here's the essential guide to DP6's new reverb plug-in!
Just like commercially available rivals such as Altiverb, ProVerb is a convolution reverb — it loads up 'impulse response' audio files (which often sound like a gunshot, were you ever to listen to them separately) that encapsulate what happens to a dry sound when it's played in a particular acoustic space or, indeed, through an effects unit. It can then apply those same changes in real time to any other sound you play through it.
You insert ProVerb in a Mixing Board plug‑in slot just like any other plug-in, but then the first thing to do is to load the impulse response you want. If you click on the little triangle under the central waveform display, you'll open a pop-up menu giving access to impulse responses stored in the Factory, User, Shared or Project libraries — but you might find that the last three are greyed out. By going down through the hierarchy, you eventually end up with an impulse response you can choose, and this loads and shows up in the waveform display. MOTU provides a good range of Cathedrals, Halls, Rooms and Plates, and also some Delay-based presets showing that ProVerb can do more than just 'normal' acoustic recreations.
ProVerb, DP6's new convolution reverb plug-in, gives potentially unlimited access to high-quality acoustic treatments.With an impulse response loaded, for many applications the key settings to make are then the wet/dry mix (the central control at the top), and perhaps an adjustment of reverb length, using the very large 'length' knob. The latter is not calibrated but adjusts reverb length by a factor of +/- 4. How could something that sounds so nice be this easy to use?
Control Zone
There's a heck of a lot more to ProVerb, though. Even the lowly mix control has hidden depths: from its self-effacing pop-up menu you can choose different wet/dry crossfade characteristics, or turn it into a volume control for the wet level while the dry level remains constant (or vice versa). Pre-delay is a familiar feature of conventional reverbs, and you can use it in the same way here. However, the 99-millisecond limitation points to the fact that this is more of a tonal tool, helping to integrate the sound of the reverb and the dry signal without phase issues. That also explains why it offers negative pre-delay too, shifting the onset of the reverb backwards with respect to the dry signal. The Width/Pan parameter is another dark horse. On a stereo-to-stereo ProVerb it's labelled 'Width' and controls channel routing, so that when it's at zero you get normal stereo operation, when it's at maximum the left channel feeds into the right input (and vice versa, to reverse the stereo field), and when it's in the middle the channels are mixed, effectively producing a mono-in/stereo-out reverb from a stereo source, which is often very desirable. However, on a mono‑to-stereo ProVerb, the same knob is labelled 'Pan' and does something odd: now it's like a panner that places your mono input in the stereo field at the position of the knob, and pans the reverb the other way.
While the Damping (a simple low-pass filter) and four-band EQ offer no great surprises, the Dynamic Mix section (on the right, at the top) is most unusual. Its controls hide a nifty side-chain compressor arrangement that allows strong dry signals to momentarily compress the reverb. It works best with varying, discontinuous signals such as vocals, and you set it up a lot like a normal compressor. The Thresh parameter determines at what input level the effect will kick in, while Comp and Sens work together to control how much compression occurs.
Acting On Impulse
Where ProVerb can get really exciting — and really weird — is in its ability to load other impulse responses. You can find decent ones on the net (try http://noisevault.com/nv/, for example) or use ones you've made yourself. In both cases, you just drag an audio file from the Finder on to ProVerb's waveform display, and it loads it as a 'Project' impulse response (which can then be copied to your User or Shared libraries, using the pop-up menu). ProVerb isn't particularly fussed whether you use a mono or a stereo file — you get good results from both. And if experimental processing is your bag, your impulses don't have to be anything like 'normal' impulses; you can use anything. For example, I made one by doing an acoustic piano glissando from one end of the keyboard to the other. Imported into ProVerb, its length reduced, and then applied to a drum track, the effect was seriously odd, but beautiful, and hard to achieve in any other way.
Many of the impulse responses MOTU supply with ProVerb come in three versions, called Left, Right and Center. There's no explanation of what these are for in the DP6 manual, but they make a difference to the sound: try making ProVerb 100 percent wet and then listen carefully to the stereo effect as you switch between the different versions. Basically, a strong stereo image of the dry signal is produced either at the left, right or centre of the stereo field. This may be down to how the impulse responses were recorded, and also probably points to the fact that even a stereo-to-stereo ProVerb actually sums its inputs to mono, but what does it mean from a practical point of view? Basically, I'd suggest that for normal reverb treatments you select the Center variant. But if you're using multiple ProVerbs for orchestral scoring, for example, you might want to run your 1st violins (which traditionally are on the left side of the stage) through a Left variant, and your trumpets (which sit on the right) through a Right. This could give you better imaging and perceived width in the final mix.
There's more about what's possible with convolution, and some third-party libraries you might want to use with ProVerb, in Martin Walker's excellent article in the April 2005 issue of Sound On Sound, online at /sos/apr05/articles/impulse.htm.
Strum Together: Virtual Guitar Instruments & Digital Performer
There's no doubting the rhythmic drive that strummed acoustic or electric guitar parts can bring to a track, or their role in creating a range of textures in an arrangement. What do you do, though, if (like me) you don't play guitar, or you're working on the road and want to lay down some guitar tracks but don't have a way of making a decent recording? The answer is a virtual guitar instrument, and there are a couple available that work very well in Digital Performer.
• Applied Acoustics Strum Acoustic ($229, Audio Unit)
As the name implies, Applied Acoustics' (www.applied-acoustics.com) latest virtual instrument is dedicated to simulating acoustic guitar sounds. Like AA's other synths, this one isn't sample-based but instead uses physical modelling techniques, so what's lost in terms of sheer realism is arguably compensated for in the flexibility and 'life' of the sound.
Unlike your typical sampled guitar, where different pitches are mapped across the keyboard in a normal piano-like fashion, Strum Acoustic uses what's arguably a much more useful approach. You 'fret' individual pitches or chord shapes with your left hand, and then trigger various strums and individual string picks by playing dedicated keys, all within one octave, with your right hand. Essentially, this process mimics the way guitarists actually play, and the plug-in interprets the pitch information you feed it to generate appropriate guitar‑like chord voicings, which can be controlled to a fair extent. The result is a degree of realism that could otherwise only be achieved with perhaps hours of MIDI programming of a conventional guitar multisample.
It has to be said that as a solo instrument Strum Acoustic has a distinctly synthetic twang, but the playability is undoubted, and the sounds work well in a mix. I encountered no difficulties, and pre‑rendering was reliable. Used in real-time, though, Strum Acoustic's modelling process does eat a chunk of processing power, so you might want to check out the demo version first, especially if you're running DP on an older Mac.
• MusicLab RealStrat 1.1 and RealGuitar 2.2L (bundle $399, Audio Unit)
RealGuitar has been around for a while, and it was one of the first plug-ins to use the 'fret-and-trigger' model that the likes of Strum Acoustic have since copied. RealStrat is more recent, operates in pretty much the same way, but has a few more tricks up its sleeve (see www.musiclab.com). The difference between them is that RealGuitar offers a range of acoustic guitar sounds, and RealStrat (yes, you've guessed it) sounds like a Fender Stratocaster. Both are sample-based, but we're not talking multi-gigabyte libraries: at a typical 44.1KHz sample rate RealStrat's sample set is around 40MB.
Both the Real guitars offer five modes: Solo, Harmony, Chords, Bass & Chord, and Bass & Pick. Solo mode is really clever. You play what seems to be a conventionally mapped multisampled sound, but the plug-in 'voices' what you do in real time, sometimes re-fretting a string (and therefore muting nearby pitches) and at other times allocating notes to another string. If you play typical arpeggiated parts it somehow, magically, ends up sounding like a real guitar. Harmony mode allows a single note to trigger a range of simple intervals and power chords, while the three remaining modes offers variations on the the fret-and-trigger scheme to cover anything from simple strumming to more complex picked accompaniment patterns. In all modes the sense of realism is remarkable. MusicLab's programming is smart, too; it sounds like you never get the same two samples back to back, so there's none of the boring uniformity that we often associate with sampled guitar. Of course, you're still going to have miles more flexibility playing a real guitar, but this is pretty phenomenal.
I really rate both MusicLab guitars, and note that you don't have to buy both as a bundle. They're available separately, and in some cheaper versions.
