This month, the SOS team are in New York helping reader Glenn Bucci with his recorded guitar and bass sounds.
Glenn Bucci, an SOS Forum regular, got in touch with us to see if we could help sort out a few mixing problems he was having in his basement studio, located around an hour from New York in the town of Huntington, just beyond the notorious Amityville — yes, it is a real place! Apparently Glenn was a regular follower of our Studio SOS exploits, as his wife unveiled a massive plate of chocolate biscuits almost as soon as we arrived!
Glenn's studio is based around a PC running Cubase SX running under Windows XP, though he had only 384MB of RAM, which was barely significant for his application, and we suffered occasional glitching during our visit that may have been due to this. However, as Martin Walker is not an allowable item of hand luggage, and because I have very little experience of Cubase SX, we didn't look into this any further, especially as Glenn had promised himself a new and faster computer in the near future. An RME card with ADAT I/O served as the audio interface, communicating with a Behringer DDX3216 digital mixer fitted with a pair of ADAT I/O cards. A pair of Mackie UAD1 cards in the PC provided extra processing power. Glenn's mics included a Rode NT1A, a Blue Blueberry, an Audio-Technica AT4033 and an Electrovoice EV237.
On the hardware side, Glenn's Manley Laboratories Langevin Dual Vocal Combo was his most recent purchase and favourite input device, though he had a Focusrite Voicemaster for when more inputs were needed, in addition to Focusrite Compounder and Mix Master units. He also owned some very nice guitars, a Mesa Boogie V-Twin recording preamp, a Fender Twin and a 12W Marshall amplifier. The monitoring comprised a pair of Tannoy 800s supported on adjustable tripod stands, with home-made wooden top plates and ribbed rubber mats to provide a secure and vibration-free fixing. These speakers have a fairly wide dispersion, which is a good thing generally, but in this case their potential was being compromised to some extent because of the room geometry.
The studio room turned out to be a slightly awkward 'L' shape, because a large cupboard for the boiler (or furnace as our US friends prefer to call it) projects into the room. Luckily, Glenn has not found noise from the boiler to be much of an issue, since most of the instruments go direct into his DAW. Also, during the cold months when it is in use, he can turn it off temporarily during a vocal or acoustic guitar take, and his well-insulated basement usually maintains a comfortable temperature.
The equipment and monitoring was all set up to face down the long axis of the room. The alcove to the right of the mixing position seemed destined to cause acoustic problems with sound bouncing back to the listening position, so we decided to drive down to the local Sam Ash (a chain of music shops across America) to buy a few panels of Auralex foam to treat this. Although professional acoustic engineers may cringe at these foam-based 'quick fixes', the reality is that they can make a significant and very cost-effective improvement in the home studio, even though they are unlikely to bring the acoustics anywhere close to professional standards. As the photographs show, we covered most of one alcove wall facing the front of the studio to try to minimise any reflections from that surface bouncing back to the listening position via the front wall.
The corner of the wall was also a potential problem, so we were careful to wrap the foam around the corner to soak up any high frequency reflections from that area (shown above). By aligning one of the notches in the foam with the corner edge, the foam folded neatly around the corner — although we allowed the main section of panel to stick firmly before trying to fold the shorter section around the corner. A further panel was fixed to the rather low ceiling in the traditional position, between the monitors and the listening position. The ceiling was low, because Glenn had specified a high degree of sound insulation in the joist space to minimise noise to and from the family room upstairs.
The outcome of our acoustic treatment was a slight but noticeable tightening of the stereo image and, as a bonus, the dead wall in the alcove would be useful when recording vocals, as the singer could stand with their back to the foam to help provide a more acoustically dry recording.
Glenn played us some of the tracks he'd been working on, and straightaway we recognised some of the problems he'd described — as well as a couple he hadn't! The first song featured nylon-string classical guitar, which Glenn had compressed, EQ'd and added quite a lot of reverb to. He wasn't initially too unhappy with the sound, but I felt that we could help him improve on it, especially as he had some nice mics, as well as the Manley preamp.
After bypassing all his processing, the guitar track sounded somewhat congested and 'pumped' on occasions, so we asked Glenn if he'd compressed it during the course of recording. He told us that he'd applied just gentle compression to prevent overloads during the recording. However, the effect was very obvious to the point of being detrimental, and it couldn't be undone by any further processing, so we suggested that he re-record the part from scratch using no compression, so that we could go through each stage of the process with him. As he was recording everything with 24-bit word lengths, we could afford to leave 8dB or so of headroom without losing any significant resolution — which in turn meant we didn't need compression as a 'safety net'.
We set up the mic (an Audio-Technica AT4033, which is a cardioid back-electret capacitor design with a medium-diameter diaphragm) over a piece of plywood placed on the floor — Glenn had got this idea after reading about the technique in a previous issue of Sound On Sound. The idea of the board is to reflect some of the sound back up from the floor in order to give a more lively sound when recording acoustic guitars and similar instruments. The mic's suspension shockmount uses an elasticated cord, providing isolation between the inner and outer frames, which is threaded through holes in the inner frame and hung on hooks on the outer frame, above and below. After 12 years of use, the elastic had stretched well past the point of no return and consequently not only provided a complete lack of shock isolation, but also fell off the suspension hooks with the slightest provocation, putting the microphone at risk from the effects of gravity!
Replacement suspensions can be purchased from Audio-Technica, but it's fairly easy to replace the elastic yourself — suitable elasticated cord can be purchased from any haberdashery shop, and the lacing is fairly self evident. The original design uses a metal crimp to secure the ends of the cord, but a simple knot will suffice. The critical thing is to get the length and tension of each support loop right, and that takes a little trial and error, but the results are very worthwhile. If you find yourself repairing a similar shockmount, pay attention to the condition of the rubber bands which hold the microphone in place as well, as these tend to dry out and go brittle with time, and then they split easily.
Placing the AT4033 level with the guitar and aiming it at the point where the neck joined the body gave a fairly natural sound over the headphones. I made some further adjustments while listening to Glenn play and suggested using a higher mic position to lose some of the low end from the sound. Glenn was a bit sceptical of this, so we didn't try it for the first test recording, but as soon as the recording was complete he felt the need to add some high-end EQ within Cubase to brighten the sound. To avoid being over-reliant on EQ, we suggested recording again, but with the mic raised slightly higher above the neck of the guitar, so that the sound was closer to what he wanted at source. This would avoid having to use so much EQ, essentially to correct an inappropriate mic position.
The resulting recording was nice and bright, with less muddiness at the low end, and it sat in the mix much better, so we went on to try to establish some suitable compression and reverb settings. This time we used the Urei 1176 compressor simulation plug-in from Glenn's Mackie UAD1 card to apply fairly gentle compression. We used a faster release time than I suspect Glenn had used before, so that the compressor would have time to recover between notes. This worked fine with just three of four decibels of gain reduction showing on the peaks, so we moved on to fixing the reverb.
Originally Glenn had used the UAD1 Dreamverb plug-in to set up a church-like acoustic, but both Hugh and I felt that this was swamping the guitar sound, and also robbing the mix of much of its valuable space. Instead, we set up a shorter, brighter room sound that complemented the attack of the guitar and created a nice impression of space without washing out the sound. The exact settings we arrived at can be seen in the screenshot, though I've no doubt we could have improved on this further had we had more prior experience with this particular reverb plug-in. Glenn's drum part, which was sequenced using a Steinberg LM4 virtual instrument, was also struggling against overwhelming reverb, so we used the same small bright room and backed off the level to get a more natural sound.
We then went on to tackle Glenn's bass guitar sound, so he opened up a cover version of Eric Clapton's 'Stepping Out' he'd been working on (originally from John Mayall's Bluesbreakers' Beano album), where he'd simply DI'd his bass via the Manley preamp, while adding some compression. There was nothing intrinsically wrong with the bass sound, though it lacked mid-range definition, which meant it wasn't being heard as well as it might have been when the rest of the mix was up and running. We suggested re-recording the part using his Mesa Boogie V-Twin tube preamp, which we then fed into the Manley preamp so that we could use its compression facilities if we needed to. We used the V-Twin's clean channel with Gain set at around 25 percent, Bass set halfway, and Mid at 90 percent. Treble was set low at 10 percent, while Presence was at 30 percent. These settings gave a nice crisp sound with plenty of mid-range definition that would really cut through in the mix.
We actually used very little compression while recording, and probably needn't have used any at all, but it still sounded DI'd to me, so we agreed to try some 1176 compression to give the sound a bit more attitude. We also added a little EQ boost at 80Hz and 250Hz (using the UAD1 Cambridge EQ plug-in) coupled with some gentle HF roll-off, giving us a sound that sat rather better in the mix.
As a final tweak, Glenn inserted a Steinberg tube-simulation plug-in and adjusted the drive until the sound was just starting to roughen up, which gave more of an amped feel. One tip here is that, if you do use a tube simulation plug-in and you feel it is making the sound too edgy or gritty, you can put a sharp high-cut filter after, with a slope of, say, 18dB/octave at 5kHz. Because every bass sound is different, you should try adjusting the filter frequency so that the sound smooths out, without actually becoming dull. Though these various tricks and adjustments gave us more of an 'amp' sound, a dedicated processor, such as a Line 6 Bass Pod XT, would make this very much easier and offer a far greater choice of tones.
Glenn's monitoring setup actually worked pretty well, so a minimal amount of basic treatment was all that was needed. He has been careful to choose good-quality equipment, both hardware and software, so the main mix problems he was encountering were related to overprocessing, specifically with compression and reverb. Again, our visit underlined the importance of getting as close to the desired sound as possible at source, both by choosing a suitable mic and by adjusting its position, as this minimises the amount of EQ needed and invariably leads to a more natural and open sound. It's also important to make the final EQ adjustment with all tracks playing, as the subjective result can be quite different to EQ'ing in isolation.
The UAD1 1176 compressor proved very effective on the bass and guitar tracks that we worked on, and setting up a brighter, more intimate reverb helped retain clarity and create a sense of space. However, compressing for no good reason (or over-compressing) can seriously compromise some sounds, and is best avoided by recording without compression where possible and then trying various compression settings afterwards at the mixing stage. The downside of over-compression was also evident on a sampled piano part Glenn had recorded, where the compression made it sound somewhat hard and electronic. As a rule, I wouldn't compress acoustic pianos (or good piano samples) at all.
Glenn is a big fan of '80s productions where reverb was used more extensively than it is now, but to get away with that type of production you need to use a very high-quality reverb unit or plug-in and, equally importantly, choose a setting that doesn't muddy your mix. Dreamverb was clearly up to the task, but it meant doing some patch editing. Rolling some of the low end out of the reverb is one way to avoid mix congestion, but getting the effect exactly right takes some experimentation, and it's worth saving any good patches you come up with. Adding 60-80ms of pre-delay can also help when you're treating vocals, because this provides some separation between the original sound and the reverb. Glenn had originally chosen church-type reverb patches, which are characterised by a lot of rolling low end and very little high end. While the UAD1 Dreamverb produces extremely convincing church reverbs, these weren't really suitable for the type of music being worked on. As a rule, smaller, brighter rooms, plates or ambience settings create the required sense of space and complexity without obscuring the sound or compromising its dynamics.
Finally, we felt Glenn would benefit from dual flat TFT screens on his computer, as these would help manage the windows in Cubase better and also help reduce the amount of hum on his Fender Stratocaster guitar due to magnetic radiation. When Glenn played his Strat, there was a further interference problem that we couldn't track down in the time available (we had an appointment with a curry downtown!), but during the drive back to Manhattan we concluded that it could have been radiation from the rather large TV set in the room above. The only solution is to turn off such potential sources of interference while recording and/or fit the guitar with good noise-cancelling replacement pickups (such as those made by the Australian company Kinman) that won't compromise the tone.
"First I want to say it was great to meet the staff of SOS — I had a lot of fun! Having now listened to the effects of putting up the acoustic foam, I can confirm that my stereo image has indeed improved, and I know this will help me get more accurate mixes in the future. As for the shockmount, I've now repaired it, and it works and looks better.
"I was pleasantly surprised by Paul's mic placement on the guitar, as I normally just put it a couple of feet in front of the instrument. It seems it really is worth spending the time to get the correct mic placement to capture the sound you want, rather than using an EQ to correct the sound later. Paul and Hugh also taught me that, while getting a great sound on your bass is important, you also have to find the correct sound to suit the type of music you are doing. My smooth jazz bass sound wasn't right for a blues number, but Paul's punchier, more mid-range sound was just what the doctor ordered!
"With 24-bit recording, Hugh emphasised that leaving enough headroom to avoid clipping would not degrade the sound, as the final track would be converted to 16-bit resolution for CD use anyway. Leaving 8dB of headroom is usually better than pushing your signal close to the red and then using compression or limiting to avoid clipping. I also learned that adjusting (reducing) the low-end EQ on a reverb can help take the boxiness out of the sound. When I recorded my acoustic, I was trying to get the same reverb Larry Carlton did on his recording of the Lord's Prayer (in the late '80s/early '90s, mind you). However, I actually preferred the more natural sound that Paul came up with, which sounds like the guitar is in front of you, not at a church recital. Thanks again to the SOS team — I really enjoyed meeting up, and I learned a lot."