Spending thousands on fancy gear won't get you anywhere — at least, not unless you spend some time sorting out and getting to know your listening and recording environments...
Jez Mumford is a former music technology student who decided to set up a small studio in the family home. His very supportive mother allowed him to take over the conservatory as a control room, and also to convert his small adjacent bedroom into a dual‑purpose sleeping/live room. Some of his equipment arrived during the discussion we had prior to our visit, including a very nice pair of Focal CMS50 active monitors, so Jez hadn't had the opportunity to get his system properly set up and tested. Our challenge was to get his system running and to make sure that the sound was acceptable, within the constraints of a conservatory with a glass roof and glass front wall! Jez also wanted to find the simplest way of producing satisfactory vocal recordings in his live room/bedroom, which turned out to be a near-square space with two plasterboard walls, one solid wall and one wall occupied by glass patio‑style doors.
At the heart of the system is a commendably quiet rackmount PC, built specifically for musical applications and running Cakewalk's Sonar software. The computer has an M‑audio interface built in, but Jez's Allen and Heath ZED R16 mixing desk also serves as an audio interface via a Firewire connection, and can also double as a control surface. However, he'd had to sell his one and only decent microphone to finance his other purchases, so we took along an Audio-Technica AT2020 to test the system and to show Jez what kind of quality to expect from what is actually a very affordable budget microphone.
On our arrival, we found the system set up on a small computer desk, close to the end wall of the conservatory, and with the monitors pointing down the length of the rectangular space, which measured roughly 3 x 4.25m. The monitors were on Quik‑Lok metal stands, correctly positioned to either side of the pair of computer screens. These stands tend to be a little wobbly because of their design, but were adequate for such compact monitors as the CMS50s. Around half the ceiling was fitted with lightweight, corrugated blinds, which would provide a limited amount of diffusion, and the ceiling above the mixing position was also sloped in such a way that any reflections from the monitors would not bounce directly back to the listening position. As we worked, a pair of goldfish watched intently from a bowl next to the computer desk!
Our first impression was that the monitor speakers were a little too close to the end wall, so we pulled the desk and everything around it back into the room by around a foot. Our next step was to re‑check the sound of the existing setup by playing Hugh's test CD of well-known material, which also includes some useful left/right speaker checks, phase inversions and stepped audio frequencies, for identifying excessive peaks and dips in the bass response. As three of the four conservatory walls are lightweight (the ends are uPVC panels), we suspected that a lot of low end would either pass through or be absorbed by panel resonance — and this later proved to be correct, as the low end showed no significant dips or bumps in the room's response, and the Focal monitors tailed off very gracefully below about 75Hz. Of course, the sensibly‑sized monitors without excessive bass extension also helped!
Jez had hooked up his monitors to the main stereo mix-bus outputs of the ZED R16 console, but playing the test CD from the computer's DVD drive initially produced no sound. It turned out that Jez hadn't configured the PC to use the R16 as the default audio device for Windows and that was easily sorted... However, because Jez had connected his monitor speakers to the main stereo outputs of the mixer, the only level control we had was the main faders — rather than the purpose‑designed facilities of the Control Room section of the desk. Jez had done this because the main outs are on XLRs, which suited the balanced cables he'd connected to his monitors, whereas the monitor outputs are on balanced TRS jacks. This meant the only way he could control the monitor volume was to turn the master faders down, which would, of course, also affect his mixes if he chose to mix via the ZED R16. He didn't have any balanced jack‑to‑XLR cables, so we drove to the nearest music store, only to find that it closes all day on Wednesdays — and this was a Wednesday! So it was on to plan B, which was to cut up a spare moulded TRS jack cable Jez had lying around, and join its two halves onto the existing XLR cables using connector blocks, after first removing the male XLRs at the mixer end. That meant a trip to the local DIY store for some connector block and insulation tape. We also picked up a few bits and pieces to secure the foam panels we planned to put up in the conservatory to improve monitoring clarity at the listening position — but more about that later.
While Hugh set about the cables with a small screwdriver, a pair of wire cutters and a roll of PVC insulation tape, I glued some lightweight, plastic, electric-cable conduit across the top edges of six pieces of Auralex foam, so I'd have something to hang them by. I didn't want to make anything too permanent, as this was, after all, Jez's mum's conservatory, so I opted to use self‑adhesive plastic hooks fixed to the glass and uPVC frames, which would allow the foam panels to be hooked on or off as required. The lip on the plastic conduit sat nicely over the hooks, and the only problem I encountered was when sticking hooks to the patio-door glass leading to the live room. I tried cleaning the glass with detergent, glass cleaner — and finally even brandy! — but the hooks kept falling off after a few minutes. I came to the conclusion that as this door was originally on the outside wall of the house, the glass probably had a 'stay‑clean' coating, and that's why nothing would stick for long. The simple solution was to use a longer piece of plastic conduit, so that it stuck out at either side of the foam panel, and to stick the hooks to the uPVC door frame rather than to the glass. This worked fine.
We didn't do anything unusual foam‑wise, so I won't dwell on it for too long — you've seen us sticking up foam plenty of times before! We placed one piece each side of the listening position at the so‑called mirror points (which happened to be on the glass door panels) to intercept early reflections from the monitors, one piece behind each monitor, and two more on the rear wall. While hardly an exhaustive treatment, we expected it to produce a reduction in direct reflections at the listening position, bringing about a significant improvement in the stereo image and a slight drying up of any room reverberation, which our subsequent listening tests confirmed. While all that glass and the wood floor did little to dry up the sound elsewhere in the room, Jez also had a soft sofa on order to place at the rear of the room, which is always beneficial to the acoustics, as well as to the creature comforts of the room.
With the modified cables in place, we now had control over the monitors via the Control Room section of the ZED R16 mixing desk, so we did another monitoring check with Hugh's CD. Thankfully, everything came out of the correct speakers without any unexpected buzzes or hums (despite the unconventional wiring arrangements). Importantly, the in‑phase mono and out‑of‑phase checks were substantially better than before the foam had been installed, too. However, there was a slight balance offset and too much gain in the speakers, which prevented the console monitor knob being set to a sensible position. A quick adjustment to the sensitivity switches at the back of the CMS50s resolved the latter issue, and turning the speakers' front-panel level trims up to full cured the former. More critical listening tests led Hugh to try adjusting the rear-panel EQ controls on the Focal monitors, and we ended up with both the bass and treble filters set to +2dB, which we all agreed sounded good. In theory, this would impart a gentle 'smile' or mid‑dip curve to the overall response, but in conjunction with the inherently bright room character, it sounded subjectively accurate.
Next, we booted up Sonar for a test run. Although there were two monitor screens, Sonar's window was initially maximised to fill just one, which made for a very cramped display. Hugh resized the Sonar window and dragged it across to occupy both screens, the idea being to have the mixer window on one, and the arrange window on the other. However, this then exposed a problem: when dragging windows from one screen to another, the window size changed and the position of the window on the second screen didn't line up with that on the first screen. It was pretty obvious that the screen resolutions of the two monitors were set differently and misaligned, but a brief trip to Windows' monitor settings menu soon remedied that, after which the window position lined up perfectly — though there was a slight colour discrepancy between the two screens that we didn't waste time fixing: Jez could play with the monitor colour settings after we'd set off up the motorway!
After a few tentative prods at the ZED R16's transport buttons produced no results in Sonar, it was back into Sonar's configuration options to find how Sonar could be set up to recognise the ZED‑R16 as a hardware controller. This turned out to be in Options / Control Surfaces, where selecting the ZED R16 brought about the desired effect. Now the transport controls worked, and the first eight mixer faders operated their corresponding on‑screen faders, but to get the rest of the mixer's faders doing anything, it was necessary to assign them manually to the desired mixer controls. We assigned the remaining eight mixer faders to Sonar channels 9-16 and the four Group faders to four output buses we set up in Sonar. (Allen & Heath also provide a Sonar Template with 16 separate outputs that feed back into the mixer channels, for those wanting to mix outside the box.)
As assigning faders every time you start a new song would be a waste of time, I suggested to Jez that he save the setup as a template song, which he could lock to prevent it being changed accidentally. It can always be unlocked and tweaked if something needs to be added or changed. This song could also have the arrange and mixer windows open on the two adjacent screens, with a reverb patched into send number one. A copy of this 'template' song could then be opened when starting a new project. Of course, it's possible to create multiple templates for different types of project, such as remixing, tracking, mixing in the DAW, mixing in the ZED R16, mastering, and so on.
Jez called up a latency-checking window and showed us that this peaked repeatedly to a very high value for no apparent reason, which seemed to worry him. We explained that, as computers are multitasking devices, he should shut off everything that wasn't needed while working on music, and disable any energy‑saving and auto‑sleep options. Once we'd deactivated the wireless Internet connection, the latency bar-graph remained stable at below 10ms, so the wireless Internet had clearly been interrupting the proceedings unnecessarily.
For drum recording, Jez planned to use his Roland TD3 kit as a MIDI controller to drive the Steven Slate Drums plug‑in instrument. This is a good way to work, but even a small amount of latency can ruin a drummer's feel. I explained my own workaround to Jez, which is to feed a DAW mix (which could be a simple headphone output on the desk) into the aux input on the drum kit 'brain', so that the kit's own headphone amp can be used for monitoring while playing. This would require a special cable or adaptor, as the brain accepts a 3.5mm stereo jack at its aux input. The aux input level control can then be used to balance the DAW mix against the Roland kit's own sounds for monitoring purposes, and as these drum sounds are not going to the computer and back, there's no latency. The drum instrument being triggered via MIDI in the DAW has to be muted or turned down during recording, of course, to avoid doubling the Roland kit sound with the slightly delayed sampled kit. The outcome is that the drummer gets latency-free monitoring using the electronic kit's own sounds, while the engineer records MIDI that can then be used to trigger any drum instrument or drum sample set that is available to the DAW.
That left us with the live room to look at. Jez had already started working on a hole through the wall for the wiring, but as this wasn't finished (and was too small for an XLR plug) we simply laid an XLR cable through the patio doors and used my AT2020 mic for the tests. The wooden floor of the room would be good for acoustic guitar recording, while the mattress Jez would be using to sleep on could be leant against one of the walls to absorb some reflected sound. However, Jez was more worried about getting clean vocal recordings, as all the walls were bare and reflective, other than a thin blanket draped over a spare sheet of plasterboard leaning against one wall.
Rather than trying to treat the whole room, we decided to use our customary cheapskate methodology, which combines a Reflexion Filter (or similar device) behind the mic with a heavy, folded duvet behind the singer, to intercept reflections from that quarter. This provides a reasonably dead environment from the microphone's viewpoint, without the need to treat the rest of the room. To test the idea, we used a boom mic-stand set up in the shape of a letter 'T' to support the duvet (heavy polyester is best, as feathers sink to the bottom!). Meanwhile, Hugh rejigged the mounting hardware of the Reflexion Filter (kindly donated by Sonic Distribution) so that it could be placed on an old vertical mic stand (with a small but heavy base), without the risk of it overbalancing. We placed the duvet across one corner of the room so that the singer would be facing the glass patio doors and have good eye contact with the engineer.
A better device for supporting the duvet would be a budget lighting T‑bar stand, as these go higher than mic stands and have a wider, more sturdy tripod base — though a curtain rail fixed across the corner of the room could also be fitted to make the arrangement more permanent. In fact, if Jez wants to get really creative, he could fit corner shelving behind the duvet for clothing storage, which, when full of clothing, would add to the acoustic absorbency of the area — as well as providing him with an invisible wardrobe!
With everything set up, we recorded a simple speech test, which confirmed that the results were, indeed, adequately dry. Jez was also surprised at the sound quality from the Audio Technica AT2020 microphone (a favourite of ours, given its price!), and told us he'd probably buy one for himself.
Before leaving, we suggested to Jez that he do some experimental recordings and mixing sessions as soon as possible, both to get familiar with his new system and to see what subjective differences were apparent between mixing the same project entirely in the DAW and feeding the tracks and buses out to the ZED R16 mixer, and mixing it there. Many people still come down on the side of analogue mixing, both for convenience and sound quality, but the down side of working in that way is that the mixer settings are not stored along with the other DAW data — which means that it becomes harder to bring back an old project exactly as it was.
The up side is the improved ergonomics of mixing, with discrete physical controls, such sonic benefits as analogue mixing may confer, and, of course, the ease with which hardware signal processors and effects can be patched into the mixer. One way to make project recall less painful is to take a digital photograph of the mixer panel, along with the front panels of any hardware processors connected to it, then save the photos in the same folder as the song, so they can be used as a reference at a later date.
Jez seemed very pleased to have his system operational, and was clearly keen to put it to use, so after thanking his mum for plying us with food and drinks, we braved the gathering storm clouds and headed for home.
Jez Mumford: "I knew the guys would have a hard job on their hands, as working in a glass conservatory is far from convenient, but wow! The studio is so much better! The stereo imaging has dramatically improved, phase problems are much more noticeable, and the speakers are set perfectly. I'm going to feel a lot more confident in my mixing now that my control room has been treated. I never thought I could get such clean and dry-sounding vocals in my live room with such little treatment. The AT2020 will definitely be on my shopping list of mics!”