This month, one of our engineers sets out to fake a big, lush, recorded string sound with only a handful of musicians.
The brainchild of former Broken Family Band guitarist Jason Williams, I Strip For Couples is an ambitious project that features a large collective of musicians from around the Cambridge area. The group was originally put together for a one-off live show towards the end of 2013 and features two live drummers, bass, guitars, piano, string section, brass and rap-style vocals, combined with various samples and sound effects. Fuelled by the success and warm reception to the live show, Jason is attempting to create an album based on the same material, whilst also preparing for a second live performance this coming May. I've been tasked with helping him with the recording side of things, and for this month's Session Notes I'll be explaining the process we recently went through to get the string elements of the project captured, and to achieve certain stylistic aims along the way.
Like many sound engineers or producers these days, I didn't have the opportunity to learn my craft in one of the big studios where large-scale string recordings might have taken place. At Half-Ton Studios here in Cambridge, I deal mainly with traditional guitar/electronic band setups, and my experience of recording strings is limited to overdubbing cello or violin parts, or perhaps the occasional mini 'mock' string section with two or three players. Jason has a much grander scale in mind for this project, however, and whilst he's keen for the project to have a very contemporary style, the string sounds that excite him are from a different era, namely the '70s, with examples being War Of The Worlds and ELO. Thinking about the scale and budgets that were probably involved with these kind of sessions, I realised that capturing similar strings for Jason's project would be no mean feat, especially when the budget was almost non-existent!
Jason is quite an accomplished user of software string packages such as LA Scoring Strings, and had achieved fairly convincing results when putting the original concept together. We discussed various options for how to get the strings sounding how he wanted, and although I appreciate how good sampled strings are getting these days, we were keen to try and get as close as possible with the real musicians we had at our disposal. We'd always have the option of supplementing the live strings with some samples to fall back on.
The string section we had available consisted of three violins, a viola and two cellos, and our thinking was that the parts would be arranged such that recording two to three different layers would create the grander orchestral sound that was desired. Jason had managed to get a conductor on board for the live side of the project, and with his help, we attempted to turn this idea into reality.
My recording partner Rob Toulson is involved with a local university, and had managed to secure us the use of a quiet, large room, with a seemingly nice-sounding acoustic for strings. As you can see from the pictures, the room was big enough to provide the grander orchestral feel we sought. It was pretty interesting from an acoustic perspective, in that, apart from one protruding wall for a video screen, it was almost perfectly circular. Although the room was quite large (approximately 15 metres in diameter) it had a bright, smooth-sounding reverb characteristic that was very promising indeed. I've never worked in a room this shape before, and walking around clapping and probing for acoustic characteristics showed that if you stood dead centre, you got the most pronounced flutter echo I've ever heard. Probably a spot to avoid, then! Thankfully, by moving just a few feet away, this almost completely disappeared.
We were keen to build in a degree of flexibility when it came to mic placement, as we anticipated that we'd need enough separation to be able to do a few edits and to tune the odd note. Rob's mobile recording system consisted of a MacBook Pro running Logic Pro 9, a TC Electronic Studio Konnekt 48 interface and an RME Octamic II eight-channel preamp, which could feed into the Studio Konnekt 48's digital inputs. In short, we had had 12 inputs to work with, so we reckoned a close mic for each player and two stereo room options would give us enough control without going overboard. For the artist monitoring side of things, we had an ART HeadAmp 6Pro six-way headphone splitter and several pairs of Sennheiser HD201 headphones for the players. And for our own monitoring, we used a pair of Adam A7 powered speakers alongside Sennheiser HD650 open-backed headphones.
We chose to position the players in a loose semi-circle shape with the two cello players in the centre, to keep the lower frequencies focused towards the middle of the sound stage captured by the room mics. These were flanked by a pair of violins on one side, and a violin and viola on the other. We wanted to utilise the natural width that the room gave us, whilst also maximising the amount of separation on the close mics — but we resisted the urge to put more than about two metres between the players, and arranged them in a curve, just enough so that the musicians could get a sense that they were working together, and that the room mics would pick up something sounding like a collective unit.
There wasn't a huge amount of time to experiment with mic positions, so we started off by placing the room mics just behind the conductor, who was positioned centrally, facing the semicircle of players. We had two stereo-mic options. The first was my Royer SF12 stereo ribbon mic, which is set up in a Blumlein configuration, and would hopefully bring some characteristic ribbon smoothness to proceedings. I've owned the SF12 for a while, and was particularly keen to hear it in this application as I'd heard great things about using it on larger string sessions. The other option was a spaced pair of Earthworks HTM TC25 omnidirectional capacitor mics. I've always been a fan of spaced omnis where you have a wide enough source to get a proper sense of stereo, and we felt that these used along with the ribbons would give us two good but different-flavoured options. As you can see from the photos, we had the room mics set up fairly high in relation to the musicians; we did try them lower down, but the height seemed to bring out a nicer quality from the room.
For the close mics, we used a selection of good-quality capacitor mics, all set to cardioid or hypercardioid polar patterns. We had a pair of Earthworks EW30s, a pair of AKG C414s, an AKG C451 and a Blue Baby Bottle. Things can sound a little harsh when close-miking strings, so you need to try to capture the body and character of the instrument without getting too much of the bowing sound. The mics were all positioned about two to three feet away from the instruments' soundholes, slightly off-axis from the point where the strings were being bowed. There simply wasn't time to refine individual placements: we knew we'd have a short sound check in which to identify and solve any problems, so at this stage it was very much a case of trusting our instincts about the right places to position the close mics for each instrument.
We only had the room and the musicians for one day, and although there was the possibility of another session later in the year, we had to get at least five or six songs done in this session for the project to remain vaguely on track. Yet by the time everyone was present and ready to go, and all our gear was set up, it was approaching midday. The musicians are all excellent players, but as is always the case with volunteers, there was a limit to how demanding we could be. Although there had been no rehearsals, they had at least been provided with scores beforehand, and had the conductor to help guide them on the day. In sessions like this it's vitally important to establish a workflow very quickly; even trying to manage toilet and food breaks with a group of unpaid musicians can be a challenge!
From a technical point of view, everything was running very smoothly. We did a quick line check of all the players, and did the best we could in terms of soloing individual mics in our headphones to check for any problems whilst they rehearsed the first track. I sensed a certain tenseness in the room for the first 15 minutes or so, but with the excellent help of Matt the conductor and Jason's relaxed direction, things started to come together.
In hindsight, the first track perhaps wasn't the greatest choice to commence with, as it has some tricky staccato rhythmic elements that required me to create a new click track in a faster time signature, just to help tighten things up. It took a few hours to get this first piece down, but from there on in, things moved much more quickly — and by the end of the session we had managed to get the strings for seven tracks recorded.
It was very interesting to see how the dynamic of the session developed over the first few hours, with the recording team, musicians, conductor and Jason communicating to get things done. Whilst Matt the conductor naturally wanted to continue to fine-tune certain aspects of the performances, Rob and I had spoken in advance about what we thought was possible in terms of editing in the post-production stage, and given the time constraints, we often found ourselves trying to move things along a bit more quickly.
Although the room provided an excellent acoustic for string recording, it made it a little difficult to tell with confidence if things were sounding as good as we thought until we got back into a more familiar and reliable listening environment. Rob had very kindly agreed to do some preparation of the multitrack files before we embarked on further mixing, and a few days later he sent me separate stereo files of the close mics and the two room options. Listening to the files it was very obvious that we'd been lucky enough to record in a good room. The two stereo room setups each sounded fantastic, with a lovely sense of stereo, natural reverb and general 'bigness'. The Royer SF12 was my favourite — it just oozed smoothness — whilst the spaced pair of omnidirectional condensers offered a slightly brighter option. The close mics provided us with less smooth-sounding but more editable alternatives, and of course we'd be able to manipulate the panning and relative levels in order to tweak the performance dynamic if required. Even listening to the close-miked files alone, though, I still got a slight sense of the room we were recording in, and they seemed to take the adding of artificial reverb well.
Did we achieve what we set out to do with this session? Considering we only had six string players for the session, I was pleased with the size and grandness of the recordings, and Jason was very excited listening back to the first passes of the recordings. The real proof of the pudding will be sitting them in the mix of course, but we set out to capture some big, authentic, orchestral-sounding strings and I think we got very close — perhaps closer than I thought we would, in fact. At the very least, we have a strong foundation on which to build with further string samples if need be, although with finding enough space and headr
oom in a mix for big drums, bass, keyboards, vocals and so on, I think they'll be plenty big enough as they are!
For someone who has spent almost his whole recording career working out how to get things sounding good in more 'real world' rooms, getting into a nice-sounding, big room to do a proper stereo recording has been a real treat. As well developing as appreciation for the roomitself, it was great to see and be a part of the teamwork ethic of a project such as this. Unless you have the budget to just pick, choose and pay who you need, there's no other way I can see of getting something like this done. Right... I'm now off to lock myself in a small, dark control room, and will probably not emerge until I've mixed the album!
Jason Williams: "As I can't read music, had no experience of writing for a string section and had little or no budget, the original idea for this project was to create the songs using software, and then to go on stage, playing a guitar or piano part live over backing tracks. Despite the obvious quality of Audiobro's LA Scoring Strings (LASS) software, the programmed parts didn't even convince my half-deaf grandmother, so I embarked on a quest to put as big a string section together as possible. The LASS tracks didn't go to waste, though: they worked very well as demos and provided the basis for a score, produced in Apple Logic Pro with the help of Matt Sharrock — who has proved invaluable, both as a conductor and as an intermediary between me and the instrumentalists.
"The odd thing about the process of writing is that it's almost entirely solitary, lonely even, and technical — but then, when the band's playing live and recording, I often find myself playing nothing at all! When writing, I found I had to avoid grabbing handfuls of chords for each instrument. Even though LASS works in a divisi way (ie. a six-note chord would be played by six different sampled instruments, to avoid a synthetic sound), I realised that the chords would have to be built across the different instruments, and I also restricted the chords to two notes per instrument when writing (not including double stops). Partly that's because this is how it tends to be done in an orchestra, and partly it's due to the small number of musicians I'd managed to recruit.
"I'd advertised and spoken to, emailed and called hundreds of people, and ended up with seven players, comprising two cellists, one viola player and four violinists (though illness meant that when recording we were down to only three violins). If anyone's reading this who is interested, please get in touch: I'm still looking for good players, including a bass player. Thankfully, this turned out to be a very cool group of musicians, and things worked better than I could have hoped for, with our first gig being voted 'gig of the year' by a local radio station. Unfortunately, it wasn't possible to rehearse with the rock band and strings more than twice, due to financial constraints, so backing tracks of the strings were used in the rock band rehearsals, blasting over the PA at deafening levels!
"Listening back to the recorded tracks, before mixing and with no reverb added, got us excited, because they were starting to sound pretty big straight away. The main thing I noticed was that in place of the sheen of perfection of the LASS parts, our recordings offered more of the individual sound of each instrument: warts and all, you can hear what each of the instruments is made of!”
I Strip For Couples are playing at the Junction 2 in Cambridge on Saturday 17 May.
Stereo audio files of the dry close mics, close mics with reverb and the two stereo options are available for you to download and listen to. These will be used in the mix of I Strip For Couples' track 'Faultjam'