We load up the magic buses to take a psychedelic rock mix on a fresh trip.
We're often told that technical quality is less important than musical quality. As long as you have a good song and a good performance, it's said, everything else is secondary. That may be true, but it's also true that a good performance of a good song can be spoiled by technical problems, and so it was with this month's Mix Rescue candidate. The Vietnam Flashbacks describe themselves as a "psychedelic jam band", and unlike some acts of that ilk, they have genuinely strong material to underpin their sonic experimentation. What they didn't have was mixes that did justice to this material.
The Vietnam Flashbacks are a psychedelic band from Leicester who record live in their rehearsal room-cum-studio. They are influenced by artists like Love, Hendrix and the 13th Floor Elevators as well as '70s music like Hawkwind, mixed in with indie and modernish bands like the Black Angels. The full line up is: Ben Moore (keyboards), Dave Kidd (bass and vocals), Lance Bennett (drums and vocals), Pete Illson (guitar and vocals) and Riki Maru (vocals). "Psychedelia is our trip," they say, "and if you like psychedelic rock then we are what you're looking for!"
The process that we now call mixing used to be known as 'balancing', and I still think this is a better description of the job that needs to be done. There are two main senses in which a good mix needs to be 'balanced'. First, the relative levels of each sound source need to make sense, with nothing being either too loud or too quiet; and, second, the overall tonality of the mix has to be effective. There must be enough action at the bottom end for the track to sound full, yet not so much that it comes across as boomy or muddy. Likewise, the mid-range needs to be present without being boxy or tinny, and similar considerations apply to the high-frequency region.
The two kinds of balance are interconnected. If you find yourself in a situation where nothing seems to sit at the right level, it's often because the overall tonal character of the mix has gone awry; and a bass-heavy or top-heavy frequency balance tends to make particular instruments poke out unpredictably on different playback systems. The Vietnam Flashbacks' own mix of 'Cherry Cola' was a case in point. The overall frequency balance was heavily 'scooped', with too much bass and a strong emphasis in the upper treble region. Consequently, the instrumental balance was also wonky, with the bass guitar dominating the mix, the hi-hat more prominent than the snare drum, and so on.
Two things are key to avoiding this sort of problem. The first is having a reliable monitoring system, and checking your mixes on headphones and other speakers where possible. The second is referencing. To keep your mix on the straight and narrow, identify a couple of commercial tracks in a similar style to yours, and return to them every so often during the mix process — but be sure to match the levels, because our perception of tonality is heavily dependent on loudness. In this case the band hadn't mentioned specific references; as the track seemed to me to recall the finer moments of the Rain Parade, I used one of their songs as my main comparison.
Frequency balance issues quite often originate in tracking. If you're listening on an unreliable system, or you're working in a single room and don't have an isolated playback system, it's all too easy to position mics and adjust amps in such a way that your sources are captured with too much low end, or an overly aggressive mid-range, excessive sibilance and so on. I find this is particularly common when people position mics while listening very loud on headphones, and if you're forced to work like this, it's definitely worth making test recordings and playing them back quietly before committing to a sound.
The Vietnam Flashbacks' multitracks were a mixed bag from this point of view. Drums are fundamental to rock music of every stripe, and the basic drum tracks on 'Cherry Cola' were both well played and decently recorded. I had the luxury of two kick-drum mics and no fewer than three snare mics to work with; often this just means more flavours of not-quite-right, but in this case, they provided useful alternative characters. The only fly in the ointment on the drum front was that the overheads sounded a little too distant, with the snare noticeably off-centre.
Moving on, the bass guitar sound was a little on the woolly side, but not disastrously so. This was lucky, because there was enough drum spill onto its mic to make wholesale resculpting a tad dicey. The multitrack also contained four keyboard parts, essentially playing variations on the same thing using different sounds; these required some work, as I'll explain shortly, but the most problematic raw tracks were the electric guitars and the vocals.
Riki Maru's lead vocal was the feature that really sold me on 'Cherry Cola' when I first heard it: a characterful and very English performance that was perfect for the song. What the multitracks revealed, however, was that this vocal hadn't been terribly well captured. It had evidently been recorded while monitoring on loudspeakers rather than on headphones, making spill an issue, but a bigger problem was an inappropriate mic choice that sounded both boomy and sibilant.
The four guitar tracks presented slightly different problems. One was a conventional, slightly dirty electric part; another, in proper psychedelic spirit, had been recorded backwards, and sounded quite muddy, perhaps an indication of the kind of monitoring-while-getting-sounds issues described earlier. The last two guitar parts posed more of a headache. Both of these ran through almost the entire song, and clearly were important to the arrangement, but both had been played through heavy wah and fuzz. The fuzz obscured note definition and added lots of fizz, making them into a fairly indiscriminate and harsh-sounding wall of sound, while the wah ensured that they never presented a consistent spectral signature that could be targeted with EQ.
I started my own mix with the drums, partly out of habit and partly because it seemed particularly important in this case to establish a solid foundation that would put the other elements in context.
Home-studio drum recordings can often be improved by paying attention to polarity, and sometimes by using microscopic time adjustments to make the elements of the kit work as well as possible with the overheads. When I'm recording drum kits myself, I usually pack a tape measure to ensure that both overheads are equidistant from the snare drum. That evidently hadn't been done here, as close inspection of the waveform showed it arriving slightly earlier in one mic than in the other. Compensating for this difference using Eventide's Precision Time Align plug-in brought about a noticeable improvement, especially when the kit was auditioned in mono. I also adjusted for the off-centre snare by panning the two overhead mics at different settings.
Other drum processing was fairly conventional. I find Sound Radix's ingenious Drum Leveler plug-in much more natural and controllable than a conventional compressor when it comes to evening out the dynamics in a drum kit, and it offers controllable leakage suppression into the bargain, so I brought this into play on both the kick drum and snare drum. A small drum room impulse response from EastWest's Spaces II added weight and depth to the snare, and although three tom close mics had been recorded, there was only a single tom hit in the entire song, so I cut that out and muted the rest. Finally, the raw kick sound was a bit lacking in power and substance, so I resorted to a couple of my favourite tricks to beef it up. First I applied some saturation using SoundToys' Decapitator: by enabling its Thump parameter and carefully adjusting the setting of the Low Cut dial, it's possible to add exactly the right amount of heft to the bottom end, in this case at 50Hz. The resulting signal was fed through FabFilter's Pro‑MB multi-band processor, with a mid-range band set to compress and a treble band set as an upward expander to enhance the attack.
Getting the tonality of a mix right can be especially challenging if you proceed by attacking all the individual elements in turn, because it's not obvious at the start of a mix what any given element should contribute to the overall frequency balance. The tonality of the mix as a whole is the product of all its elements and, as such, is most easily shaped using EQ on the master bus. So, having got the drum sound close to where I wanted it, I put up a very rough balance of all the other tracks and started experimenting with master bus EQ.
This isn't the time for surgical notches or dipping out specific frequencies: rather, it's about using shelving or tilt filters to put the tonality in the same ballpark as your references. Usually, this means adjusting the overall frequency balance to favour the treble and upper mid-range, and that was exactly what I did here, adding a very broad shelf to lift everything above 1.5kHz by several dB. These settings evolved as I worked on the mix — my final effort had two EQ plug-ins and a multi-band compressor on the master bus — but the important thing is to be working within the target zone from the earliest possible moment.
Next up was the bass guitar. Sonically, this wasn't a million miles from where it needed to be, but it took a surprising amount of mucking about to bridge what should have been quite a small gap between the recorded sound and the sound I wanted in the mix. There was noticeable saturation in the raw bass sound, which should in theory have added welcome brightness to it, but in practice it had a muffled and slightly vague tone that needed firming up. My eventual plug-in chain involved EQ, full-band and multi-band compression, but the most important element was the SansAmp PSA-1 distortion plug-in that comes with Pro Tools. This might be old and free, but it's still incredibly useful on bass instruments, thanks to the way its effect can be focused in different frequency ranges. In this case, I used a fairly high setting of the Punch control to add some mid-range emphasis, and turned the Buzz parameter right down to eliminate lower mid-range bloat.
Meanwhile, applying an EQ boost on the master bus had revealed noticeable upper-mid harshness in the vocal track, on top of the other issues. The boominess at the low end was easily tackled using EQ, a boost at 2kHz gave it some much-needed bite, and a fairly aggressive de-esser setting got the worst of the sibilance under control — but the heavy lifting was done by Oeksound's Spiff plug-in. This can be rather a mysterious processor, and it takes some practice to get useful results from it, but one of the things at which it excels is smoothing out a vocal that contains spitty or aggressive consonants. There wasn't much I could do about the spill, but since it proved necessary to automate the vocal level in any case, I dropped it right down in the gaps between phrases.
The four keyboard tracks presented a different sort of problem. They had presumably been recorded by direct injection, and there was nothing wrong with them from a technical point of view; the challenge lay in fitting them into the mix. Most of the classic 'rock band' instruments are either bandwidth-limited by nature (electric and bass guitars), or percussive, leaving gaps for other instruments to fit into (drums), so they interlock fairly naturally to form a complete picture. Bringing four keyboard parts into this picture complicates matters — especially when each of them seems to fill the entire frequency spectrum on its own!
I used a combination of tactics to try to shoehorn them in. One was simply to drop some of them out for different sections of the song, generally thinning out the verses and creating a denser texture to lift the choruses. Another was to limit their respective frequency responses using a combination of EQ, cabinet simulation and other fidelity-destroying plug-ins.
That left the four guitar tracks, all of which also played pretty much throughout the entire song. One of them was a pretty straightforward rhythm guitar part, transitioning into a short solo right at the end. I used automation to drop it down in level in the verses, and to bring it from the left into the centre for the solo, but it didn't need much additional processing. That left the backwards noodling and those challenging twin fuzz‑wah parts.
Although the band's own mix had its share of technical issues, as described at the start of this article, it also had good qualities. Above all, it was undeniably psychedelic, with a trippy, head-spinning vibe that was spot‑on for their chosen musical style. Thus far in my own attempt, I'd concentrated mainly on the meat-and-potatoes stuff that would deliver a technically sound mix. With that now in place, it was time to add the audio hallucinogens back in, and since I couldn't find conventional ways to make the additional guitars sit in the mix, I decided to treat them less as parts in their own right, and more as sources of spacey effects.
This is generally something I prefer to leave to near the end of a mix, even if those trippy effects are going to be quite prominent, because otherwise it's too easy to get caught up in generating ever more creative noises and neglect the fundamentals. Often it's also the most fun part of mixing, so saving it until last is a way of rewarding yourself for all the hard work that went into the basic building blocks of the mix. And given that this track was explicitly billed as being psychedelic, I felt I had carte blanche to lay on the crazy stuff pretty thickly.
After I'd removed a lot of low-mid mud with EQ, the tone of the backwards guitar part was basically OK, so I didn't feel the need to try to turn it into something it wasn't. However, it was obviously crying out for some psychedelic delay effects, and on top of that, I also used Soundtoys' PanMan auto-panner to make it lurch unpredictably across the stereo field, in classic Eddie Kramer style.
The fuzz-wah parts were harder to deal with. Adding effects such as delay didn't really make any difference, since the core sound was little more than a continuous wash of noise. So, as well as trying to tame the fizz using EQ and cabinet simulation, I took a more off-the-wall approach to making things spacey, using LMDSP's fascinating Superchord plug-in. This gives you a set of tuned resonators modelled on piano strings, which respond sympathetically to the input signal and which can be manipulated in all sorts of odd ways. By loading up promising presets and tinkering with the controls, I was able to generate something that had more movement and life than the raw tracks, and a warmer, less harsh tone.
Emboldened by this success, I decided to take my psychedelic experiments further. A second instance of Superchord helped the keyboard parts become less static and more otherworldly, and I created another off-the-wall modulated delay effect using Eventide's Fission plug-in, fed both from the keyboards and the guitars. This got a bit out of hand if I left it running all the time, so I used automation to bring it in at strategic points.
Finally, I also wanted to create something a bit unusual to differentiate the song's middle section. This was much more sparse than the rest of the song, and left the snare drum sounding a little stark and proud of the mix. I discovered that I could use Soundtoys' Crystallizer plug-in to create a delay with a rapid pitch dive, so I fed this from the snare bus and added extra reverb.
Sometimes artists get very attached to a rough mix, and it can be a struggle to persuade them of the merits of a remix, even when you feel you've effected a technical improvement. Thankfully, that wasn't the case here, and the Vietnam Flashbacks were very receptive to the changes I'd made — which was lucky, because the end result of all this work was a mix that sounded quite different from the band's original. As I mentioned at the start of this article, frequency balance and the balance of instruments go hand in hand, and both were altered fairly radically in my interpretation. At the end of the day, though, it's still the song and the performance that count, and the goal of mixing is to put them across most effectively. Reshaping the balance of the mix allowed it to translate better to different speaker systems and brought it more in line with commercial tracks, but above all, it helped the most important elements of the song to shine.
"The original mixes, whilst sounding good to us, just do not compare to the depth and clarity of these mixes. The stand‑outs are how amazing the bass and drums sound and how polished it is. Everything just sounds better. It took us all by surprise, and the rougher original live recordings now sound like songs ready to be released.
"So far we have had them played on the radio and received these Twitter comments."
Listen to the Original Mix and Sam's Remix above. In addition, you can download a ZIP file of hi‑res WAV audio stems and mixes in the righthand Media sidebar or use the link below.
Download | 148 MB