In this article:
Photos too small? Click on photos, screenshots and diagrams in articles to open a Larger View gallery.
| Roland VS2480 MasterclassHands-on Power User TipsPublished in SOS January 2005 Technique : Recording/Mixing Get the most from Roland's flagship recording workstation with our power user tips.
The VS2480 is the flagship of Roland's successful VS multitracker range, offering much of the functionality of a complete recording studio from a single unit. Inevitably, however, when you shoe-horn a multitrack recorder, an automated digital mixer, flexible onboard multi-effects, and a CD burner into one box, the result is going to present something of a learning curve. So I'm going to follow on from the VS880/VS1680 masterclass in SOS March 2002 by sharing some of the techniques I've developed for getting the best out of the VS2480 in practice. Although my advice will be aimed primarily at VS2480 owners, the VS2400 and VS2000 have much in common functionally with their larger sibling, so much of what I'll be discussing will apply to the more recent machines as well. One of the things that really set the VS2480 apart from its VS-series predecessors was its redesigned mixer section. Not only was the processing resolution doubled to 56-bit, but the channel processing facilities were also dramatically improved, adding an extra EQ band, a separate filter section, and a dynamics processor with (joy of joys!) gain-reduction metering.
The improved equaliser is much more usable for mixing purposes than the equalisers in the previous VS multitrackers. However, the thing I like most about it is that it can be automated. As a result, rogue bass resonances can be tamed surgically only when it's needed, for example. Or, at the other end of the frequency spectrum, you can automate an EQ band to de-ess over-sibilant vocal tracks. There are several advantages to de-essing this way: you save on insert effects processing; the automated EQ approach tends to sound more transparent than any of the VS8F2's de-essers; and you won't get de-essing occurring accidentally where you don't want it to. Of course, automating the EQ is much more time-consuming, but I wouldn't do anything else for lead vocals now, to be honest. Despite the improvements to the channel EQ, though, I've still found that I get the best sounds by EQ'ing sources with dedicated outboard on the way into the machine. By doing this you're only really asking the VS2480's EQ to deal with mix-balance problems, rather than relying on it to effect dramatic tonal changes, so you can usually keep the channel EQ settings fairly gentle. If you're worried about processing something wrong on the way in, record an unprocessed version of the same audio onto another virtual track at the same time, so that you can return to it if necessary at the mix. Another thing that I've found works well for most chart-style material, and something that was mentioned in the interview with Mark 'Spike' Stent back in SOS January 1999, is to use your best outboard EQ across the VS2480's main outputs to add a broad global high-frequency boost while you mix. A shelf with a frequency of about 15kHz should be about right, but the exact frequency and the amount of boost should be varied to suit what you're working on. The most important thing about this approach is that it avoids a lot of high-frequency boost on individual channels, something that budget digital EQ is never really all that good at, and it should therefore help improve the overall sound. What I would say, however, is that you should record versions of the mix both with and without this buss EQ, just to keep your options open at the mastering stage. Filtering & Compression Tricks The EQ's multi-mode filter section has several nice little tricks up its sleeve. First off, you can easily create wah-wah effects, for which the BPF (band-pass filter) shape is probably the best choice. You can sweep the filter manually using the User Knob/Fader Assign mode (which I'll be looking at in more detail next month) — here a rotary control is the better bet, as you can easily get 'stepping' of the filter using a fader. Once you've decided on the sweeps you want to do, record them into the Automix system so that they play back the same every time. The filter section also provides a subtle phasing effect via the BEF (band-eliminate filter) shape, which can again be swept under the control of the automation.
The other way that I find the filter can come in very handy is where you want to add a bit of high-end sizzle to a sound, in particular to a lead vocal. Set the filter to its LPF (low-pass filter) shape, moving the cutoff frequency to 20kHz, and then adjust the resonance control to taste — you don't need to go much over a value off 1.00 to get a useful change. This is a really handy effect for bringing sounds forward in the mix without apparently changing their basic timbre, and as such you might use it in the same situations as you would an enhancer. The channel compressor has one weird little feature which can be used creatively: its Attack and Release parameters both go down to zero. If you set both of them to this value, the compressor reacts so fast that it tracks the individual waveform crests rather than their amplitude envelope. This results in the shapes of the waveforms being changed, rather than just their envelope, which means you get distortion rather than just gain-reduction. The threshold and ratio controls will adjust the harshness of the distortion sound, and there's actually quite a range of flavours to be had. If you're after something very subtle, then you might want to increase the Attack or Release parameter by a single increment to get a mellower range of timbres. It's usually good practice to create various different versions of your final mix, in order that you're less likely to have to do a remix. Probably the most common example of such an alternate mix is the 'vocal up', where the level of the lead vocal is a decibel or so higher than in the main mix. The 'vocal up' version can then be used if you later decide after hearing your track on many different systems that the lyrics are not intelligible enough, say.
One problem with doing a 'vocal up' mix on the VS2480 is that raising the vocal channel level has the potential to cause clipping if you've already maximised the peak level of your mix, so make sure that you have a little headroom to spare. You also have to bear in mind that you need to change the vocal level somewhere in the signal chain after any of the channel dynamics or insert effects — a change in level can interfere with any dynamics threshold settings, and will change the tone of effects such as amp modelling. You could use the channel fader to increase the vocal level, but if this is automated you'll have to offset the automation using the Automix editing procedures, so that leaves only the Att level control. However, if this is already set at 0dB you could easily clip the channel equaliser which follows it, especially if the vocal recording peaks near the digital maximum of 0dBFS. The solution to the conundrum is to start your mix by turning down all the channel and input Att controls by the same offset amount — a few decibels should do the job. This leaves you the EQ headroom to create whatever alternate mixes you like once you've finished the main mix. It also has the beneficial side-effect that it normally results in the channel faders being closer to their unity-gain positions, their area of greatest control resolution.
Managing A Complicated Automix I like the VS2480's Automix facilities a great deal, and I've recently used them for some pop-style tracks. Most styles of music, especially those you hear in the charts, are now quite demanding of an automation system, so I thought I'd pass on some of the the working methods which I've developed for getting the best from the VS2480's Automix when the going gets tough.
When approaching a complex automated mix, I've found that it's sensible to set up a basic balance for each individual song section first, before chaining these balances together to create a rough outline for the whole track. Finally, I concentrate on the fine-tuning: individual fader rides, detailed parameter automation, and special effects. My normal mixing 'rhythm' is to spend the morning doing the basic balances. After a lunch break, I set about chaining the basic balances together, before working on the fine-tuning for the rest of the day. I usually find it good practice then to come back to the mix the next day before printing it to my mastering recorder, as there always seem to be a few little tweaks that present themselves when you listen again with fresh ears. For the first step of this whole process, I work on the mix with the Automix mode switched off, using the Scene memories to store the basic balance for each section. For example, a pop song structure might have Scenes for verse one, verse two, chorus one, verse three, chorus two, the middle section, verse four, and the final Choruses. I find it best to work on these sections in order of importance (in other words, I usually work on the choruses before the verses), so that I can make sure I don't over-egg a part that isn't that important. Similarly, I try to balance later sections before earlier sections (in other words, I do chorus three before chorus two). If you mix from the beginning, there's a danger that the energy of the mix will peak too early, and you'll have nowhere left to go in the final choruses. I won't get into specific balancing techniques here, as there are plenty of SOS articles on the subject, although as a basic rule of thumb I try to add tracks to the balance in order of importance. That way, I end up processing the rhythm guitar to fit the vocal, rather than the other way around. Another general point I ought also to mention is that it's worth spending enough time on the most important section balances. These form the backbone of the mix, so if you don't get them right at the outset you'll make yourself a lot more work later on. For example, if you muck up the balance for your final choruses, and then base the earlier choruses on that balance, you'll have to change all of the chorus Scenes to sort out the problem, rather than just one.
Transferring Scene Data To Automix Data
Once you've created Scenes for all the main sections of your mix, you need to transfer these to the Automix engine. The thing to bear in mind here is that Scenes store information which an Automix doesn't. There is no Automix function for controlling the channel dynamics, signal routing, or stereo linking, so any such changes you make while creating the Scenes have to be kept in place for the Automix to sound right. Let's say, for example, that you have already created the mix Scenes for your song's choruses, but you then decide you want to set up a flanger on a spare auxiliary send for the verse. The change in the input assignment of the flanger's effects-return channel won't be carried out by the Automix, so you have to make a note of it and make sure that it's properly set up before you switch to Automix mode. You could do this on a scrap of paper, as there aren't usually many such changes, but I just use another spare Scene memory (I usually choose Scene number zero) to store the final routing which I require at Automix, updating it as and when any routing changes are made for any of the other Scenes. That way, before I go into Automix mode I can just switch to Scene zero to avoid the sound of any of the song sections changing when they are chained together by the automation.
Transferring the Scene data into Automix data is simplest if you want the settings for all the mixer channels and busses to change simultaneously at exact section boundaries. Switch on Automix mode by pressing the Automix button — it'll light up. Arm for automation all the channels and busses that you have adjusted when creating your Scenes. You can do this quickly by holding the Automix button and pressing Ch Edit buttons until they turn red. (You will probably have to repeat this process for the four different banks of fader assignments.) Next hold Shift and press Automix to get directly to the main Automix page in the Utility menu. Quickly double-check that all the correct channels and busses have been armed, and then choose the parameters you want to record. You'll probably want Level, Pan/Bal, Mute, and Aux Send enabled for most mixes. I'd only bother enabling EQ or InsFXLvl if you've altered them between scenes, otherwise you'll have to edit or overwrite the automation data to change their settings once Automix is switched on, rather than just being able to grab a control. EQ automation has often caught me out while mixing on the VS2480 — forgetting that I've set up automation for a channel's EQ, I'll tweak the controls while the track is playing, only to find that the curve reverts to its stored Automix settings when I stop the transport! If you're planning on automating effects algorithm changes, I'd leave that until after you've done everything else, so leave the Effect buttons at the bottom left-hand corner of the screen set to Manual status for the moment. For the next stage of the process the order is vital — if you do things out of order you'll find that your Scene settings won't transfer into the Automix properly:
Things get a bit more complicated if, say, an instrument has a pickup to the Chorus that overlaps the ending to the previous verse. The thing to do in this case is to go through the Scene-transfer process for groups of tracks which have their song boundaries in the same places, which takes a little longer. Once you've transferred the Scene settings for each set of channels, change their Automix status to Read by holding the Automix button and pressing the Ch Edit button until it turns amber. If you're planning to automate any effects-algorithm changes, it's also best to do these on a separate pass — that way you can stagger the algorithm changes with regards to the song-section boundaries in order to avoid problems such as delay or reverb tails being cut off unnaturally.
Fine-tuning Your Automated Mix
One of the missed opportunities of the automation system in the VS2480 is that you can't use its controls to 'trim' existing automation data in the way you can on more advanced automation systems. This is a shame, because it means that it's rarely practical to do anything but the first automation pass using the faders. The 'punch-in' facility is also a bit of a kludge given that the VS2480 doesn't have touch-sensitive faders: firstly, you need to hold Automix and press the Ch Edit button at the same time, which is a bit of a juggle when you're also wanting to move a fader; and, secondly, the fader motor is sluggish in letting go of the fader, so it tends to fight you for the first second or so. I've even tried to use the Utility menu's Marker Stop to accurately re-record sections of the Automix data, but to no avail — the Marker Stop function seems to fool the system somehow so the fader moves aren't recorded. For me, the upshot of these Automix limitations is that I end up approaching the automation process much as I did back on my old VS880, which means that I create most of my small-scale automation adjustments by a series of snapshots, adjusting parameters from the rotary encoders or the data wheel for the sake of accuracy. A side-effect of this is that things therefore stay simple enough in the Automix Micro Edit screen for it to be useful. The automation editing commands on the VS2480 are also a great help, especially for doing the kind of level trimming that isn't possible via the hardware controls — let's say I've not de-essed enough via an automated channel EQ, it's a simple job to just increase the relevant EQ cut settings for the whole track.
A nifty thing that the automation editing also lets you create is effects which closely resemble dynamics processing. For example, I've managed to give drum sounds extra attack by creating an SPL Transient Designer-style gain spike for the first few milliseconds of each drum hit — this might seem an incredibly tedious task, but the automation Copy command makes it fairly manageable if you have a regular drum pattern. A related technique also lets you re-balance mixed drum tracks, which I've found incredibly handy on occasion, and this is especially easy to do if you're working with a loop. Set a stepped pattern of fader automation data to adjust the loudness of the different hits, and then copy it to each instance of the loop. In fact, you could even automate the EQ as well if you have a serious 'remixing' job to do. When you get down to automating on a small scale, it's incredible what kinds of creative effects you can achieve: tremolo, auto-panning, and de-essing are just some of the simpler possibilities. That's enough to be getting on with for this month, but look out for the next instalment of VS2480 tips, where I'll be taking a close look at the some of the weird and wonderful things you can do with the onboard VS8F2 effects boards, as well as passing on some editing and routing tricks. Published in SOS January 2005 | Sunday 5th July 2009 July 2009
Click image for Contents
Photos too small? Click on photos, screenshots and diagrams in articles to open a Larger View gallery. |