Taking on a new genre is a great way to test your mixing skills, as we found out when we took on this goth track by The Hiram Key...
One of the pleasures of working at SOS is that you never know what kind of music you'll be dealing with from day to day. There's no better way to broaden your horizons than by taking on a genre that's unfamiliar and new to you, and this month's Mix Rescue is a case in point. The Hiram Key are a four‑piece band whose music is firmly rooted in the old‑school goth of Bauhaus and the Cure. Although they haven't followed the trend towards electronic music, they've made full use of modern software technology in recording their album Amerikafka.
Having spent a long time recording the album, bandleader and singer Gary Ash found that he was having trouble mixing it to his satisfaction. With deadlines looming, he posted on the SOS forum to ask aspiring mix engineers to help out, and was also happy to send the multitracks of one song over for Mix Rescue to dissect.
The song, 'One And One', had been built up as a series of separate overdubs by band members, and featured a blend of real and sampled instrumentation. Prominent among the former were a conventional electric bass, a Fender VI 'bass' guitar and synth violin that shared lead roles with a distorted electric guitar, and a strummed acoustic guitar. Most of this action took place over a rhythm track, which had been recorded on a Roland V‑Drums kit and played back using FXpansion's BFD virtual drum instrument.
Listening to Gary's mix, it seemed to me that there was nothing fundamentally wrong with the balance of the instruments. There was not a great deal of separation between them, but in a genre where copious amounts of reverb and a generally murky feel seem to be par for the course, this wasn't necessarily a problem. The overall tonality was perhaps a little on the tinny side, however, while the drums lacked 'oomph', and the instrument that leapt out at me as needing most work was the DI'd acoustic guitar. Not only did it sound harsh and scratchy, it didn't keep good time with the drums.
My first step, after loading up the multitracks in Pro Tools LE, was to cull redundant tracks. BFD had supplied its usual array of arcane drum mics, and while I kept the two kick-drum mics and the hi‑hat mic, I decided I was unlikely to need spot mics on the cymbals, or the snare bottom track. I also dispensed with a rather brutal tambourine that was out of time with the drums, and with instrumental tracks that duplicated others with printed effects. (I didn't delete any tracks, but made them inactive and hid them so they didn't appear in the Pro Tools mixer.)
Through force of habit more than anything else, I started the mix with the drums. It soon became apparent that the techniques I might usually employ on a real kit recording weren't suitable for the BFD‑generated tracks. It was also clear that even though the tracks had their roots in a real performance by a live drummer, I was never going to be able to make them sound completely natural. For one thing, the tracks seemed to have been assembled by looping short sections of drumming, rather than recording them as a complete take. For another, there was no room mic, and very little spill between tracks, while the velocities of the drum hits were almost supernaturally even.
Using compression to try to 'big up' the drum sound was a complete non‑starter; there was no ambience for the compression to pull up, while the individual drum sounds were already rather flat, so squashing them further only pushed the drums right to the back of the mix. I decided to take an alternative approach that would be more in keeping with the band's '80s influences, and set out to find reverbs that could make the drums sound both a bit more lively, and suitably big and booming. After much experiment, I settled on a simple early reflections patch on the overheads to add the life, and a variation on the Pro Tools AIR Reverb 'Medium Hall' patch to add power and substance, fed from the snare and toms.
As usual, I fed all the drum tracks to a stereo aux, where I used a tiny amount of compression, with a 5.4ms attack time, to try to bring out the transients a little more, and EQ to add some top end and a bump at 100Hz.
The electric bass part that underpinned the song was decently played and recorded, but arguably a little colourless, so I set about making it sound fuller and weightier. I began with some compression and fairly typical EQ, boosting the 85Hz and 6kHz regions and cutting some mids at 1kHz. A bit of the AIR Distortion plug‑in added thickness and bite, and as another homage to '80s production, I also threw in an AIR Chorus plug‑in. The 'More Fatness' preset proved aptly named.
The Fender VI, meanwhile, played a melodic figure that was sometimes alternated on and sometimes doubled by the violin and distorted guitar. All of these were mono recordings, and as I was going for a big and powerful mix, rather than a natural one, I made them stereo by creating duplicates, panning these to the other side and inserting a short (30ms or so) delay on them. I also set up global reverb and delay buses. The former used a rather darker take on the AIR 'Medium Hall' preset, while the latter was a short cross‑panning stereo delay with plenty of feedback, filtered to give it some tape‑style warmth. All the lead instruments benefited from plenty of both.
The Fender VI required no other processing apart from a swingeing high‑pass filter to stop it fighting with the real bass guitar. Things were less peachy with the violin, which sounded thin and nervous. In the context of the overall sound, I didn't think it necessary to keep it sounding like a violin, so experimented with a variety of fairly extreme processing. I ended up with separate instances of the AIR Fuzz Wah plug‑in on the original and duplicate tracks, with different frequency settings. The resulting sound was much wider and richer than the original, and was able to compete with the Fender VI on an equal basis.
The distorted guitar doubled the violin part, but also did its own thing during the chorus. With the violin sounding big enough on its own, I decided to mute the doubled sections and save the guitar for the chorus. As with the VI, a high‑pass filter was the only processing that seemed necessary.
Gary's lead vocal had been recorded fairly cleanly but, again, sounded a little lifeless and thin, and — perhaps thanks to Gary's having recorded it with some compression — had a noticeably boxy ambience to it. I tried various ideas to restore a bit of low‑mid richness, but all of them turned out to be muddy dead ends, and in the context of the mix, there wasn't much room for bottom end on the vocal anyway. My existing global reverb also sounded a bit underwhelming on the vocal, so I ended up creating another reverb bus specifically for the vocals. The AIR Reverb has a preset called 'Between The Buildings', which sounds almost like a cross between a delay and a reverb, with a very pronounced stereo effect; I toned down the pre‑delay and the results proved perfect for adding the requisite doomy grandeur. This was joined by a slap delay created using the AIR Dynamic Delay plug‑in, and as always — for me at least — a little compression was augmented by a lot of volume automation.
At this point, I could no longer ignore the elephant in the mix that was the acoustic guitar. It had been recorded as a DI'd signal from an Ovation Balladeer and, as with most piezo pickups, it sounded incredibly mid‑rangey and harsh, with brutal transients on note attacks. I tried heavy‑handed EQ, and it still sounded vile. I tried compressing it with a fast attack to dull the transients, and it sounded distorted. I tried distortion and amp simulation to make it sound like an electric guitar, and it sounded like a distorted acoustic guitar. I tried editing it, but it remained audibly out of time with the drums. I tried muting it, but its propulsive rhythm and stereo width were pretty fundamental to the arrangement, and it left a huge hole.
Something more radical was required, and in the end I decided the only option was to turn the acoustic guitar into something else entirely. Beginning with the least unpleasant version of my distorted, amp‑simulated processing chain, I chopped up the audio part so that only the first eighth‑note chord in each bar was sounding. I tidied up the timing of each of these chords, then added an eighth‑note delay with plenty of feedback, the idea being to create a tight rhythmic part with enough movement to drive the song forward, but without the distracting transients.
This seemed to work quite well, so I set about taking it a step further. I duplicated the track and muted alternate regions on each, so that they were in effect playing alternate bars. I then automated the pan controls on the two tracks so that they started hard right and left, and swept across the stereo field as the delay progressed. In order to lift the chorus and the last verse, I also allowed the original guitar back, in its distorted guise but at a low enough level to fill out the mix without being too prominent.
All in all, this took a fair bit of work to get right, and by the end of it I felt I had something that was, if not a perfect replacement for the original guitar, at least doing some of the same work within the arrangement.
At this point, I thought I'd better check that I was on the right track, so I fired off a rough mix to the band. This gave me a couple of days to reflect on it myself, and although the band seemed pleased with the rough mix, when I came back to it I still felt there was something missing. The band's instinct to record an acoustic guitar was undeniably the right one, and the mix reference Gary had given me featured exactly the sort of driving acoustic part they'd been going for, so I decided to throw up a mic and attempt to record a new version.
My efforts were hampered slightly by a combination of lack of time, my own cack‑handedness on the guitar, and the discovery that the drum part in the chorus departed quite significantly from the click. Nevertheless, I got the timing closer than on the original parts, and when I took the two least sloppy takes and hard‑panned them to create a double‑track (something I usually hate with acoustic guitar, but it seemed appropriate here), it was clear that this had indeed been the missing ingredient. I bounced out another mix, took it home and ran it through Slate Digital's magical 'make it louder' FGX processor, called it 'Final Master' and sent it to the band, who were either happy with it or too timid to complain!
Just as 99 percent of the problems Paul White and Hugh Robjohns uncover in Studio SOS have to do with acoustic treatment, or lack thereof, in Mix Rescue it nearly always turns out that when a track is hard to mix, the problem lies in the source recordings. That was definitely the case here: I did my best with the source files, but it wasn't until I was able to replace the acoustic guitar that the mix felt complete to me. The other parts, meanwhile, were mostly well played and recorded, but as is often the case with songs built up from multiple overdubs, they didn't seem to gel as naturally as one might expect. I suspect that if the band had been recording together, with a producer on hand to crack the whip, they'd have come up with basic performances that cohered more easily to start with. The remote recording possibilities opened up by the Internet are wonderful, but at the end of the day, there's no substitute for the human interaction you get from a bunch of people playing together in the same room.
"We are all old bastards who got sick of the current trend in goth bands to go totally electronic and basically play rave music with goth lyrics,” says Gary Ash. "There were some amazing goth bands who threatened to make goth cross over into the commercial charts, such as the Sisters Of Mercy and The Cure, but bands such as Bauhaus, Siouxsie and the Banshees, Killing Joke, Alien Sex Fiend and lesser known bands such as Specimen, Nosferatu, the Marionettes and Killing Miranda were overlooked as goth disappeared into the electronic direction.
"I personally disappeared into forming a Cure tribute band, which has toured around the world for the last 20 years, allowing us an avenue into being able to buy decent pro gear and play it on decent stages to good audiences. So, in a story not dissimilar to the Kaiser Chiefs, who formed around the remnants of a Stones tribute band, we decided to start writing the sort of music we would like to be able to buy and listen to.
"The band is made up of Gary Ash (ex Nosferatu) on vocals, guitars and Fender VI; Belle Star (ex Nosferatu, Lahanya, Sister Midnight) on drums; Roi (Mechanical Cabaret), keys and noises, and Sean Flude (Noah), bass guitar. This, our debut album, is named Amerikafka, and is a collection of songs inspired by Kafka's last novel Amerika (or The Stoker).
"The song that has been mixed by Sam is light relief on an otherwise dark album. After hearing the amazing mixes by Sam, it will probably end up as a single. I think it's kept the Bowie and Cure influences and at the same time has a commercial feel. I hope that will allow more people to enjoy it.”
Gary Ash: "On the first mix I did, I was not happy with the bass, as it was ringing out in the high LF range too much, and didn't really centre the mix properly. The vocals were recorded OK, but didn't seem to come up to the front, and it didn't sound as though the singer was in charge of the song. I had huge problems getting the acoustic guitar to sit in the mix (I had referenced Echo & the Bunnymen and really wanted that scratchy piezo sound playing fast 16ths). The biggest problem that let the mix down, in my opinion, was the drums. I couldn't get BFD to sound powerful enough.
"Now... the vocals are exactly where they need to be, and some of the percussive sounds I put in have had a 'whoosh' put on them that provides a turnaround into the chorus, which makes it sound like a whole song and not two different pieces glued together. I love that the backing vocals are almost transparent but really do their job of supporting the vocal. The drums now power out and drive the bass, which is well in with the kick. All in all, I am overwhelmed with what Sam has done, and this has gone from being an album filler to a possible single.
"The song is about telling a girl who keeps coming around your house and moaning at you for recording 24/7 that your neglect of her is in her head. The arrangement on the intro represents a moment of inspiration and then the door bell goes... this time it's time to tell her the score, but in a light, honest, jovial way. I think anyone would agree that Sam has caught this perfectly.”
We've placed a number of downloadable audio files on the Sound On Sound web site, so that you can hear for yourself the changes Sam made to this mix.