Recording Acoustic Guitar

I’ve been researching and experimenting a lot with recording acoustic guitars recently, here are some notes and observations.

1) Every room, player, string, guitar, mic, mic preamp, mic position, playing technique, song, and style of music is different. It may seem obvious, but it makes the question, “How do I record my acoustic for the best sound?” pretty meaningless. There is no one way that will always give you the best results.

2) Play the guitar in different parts of the room, or different rooms. You may be surprised how different it sounds. Where it sounds best, that’s the playing position where you should record it.

3) Rehearse a lot, warm up before recording, get used to recording a lot so you don’t get “red light fever”. Remember, “An amateur practises until they can get it right, a professional practises until they can’t get it wrong”. The recording will only be as good as you can play it. Technique is very important for the best recorded sound, as well as emotion and feel.

4) New strings/old strings. Pros usually put new strings on within 24 hours before the recording session. However some people (myself included) quite like the sound of older strings. One person’s zingy and bright is another’s harsh and brash. One person’s old and muffled is another’s warm and woody. So choose wisely for the track, which would sound best in the mix?

5) Is the guitar well setup, comfortable, and easy to play? How does it respond to dynamics? As mentioned above, all guitars are different. Try to play to the strengths of the guitar.

6) Mics are a huge subject in themselves. They all record sound, but they all sound different. ;) Do you want a large diaphragm condenser, a small diaphragm condenser, a dynamic, or a ribbon? Should you use the omni, cardioid, or figure 8 polar pattern? A single mic, a stereo pair (and if so, in X/Y, A/B, ORTF, DIN, NOS, M/S, or something else entirely?), or even more mics, perhaps also a distant mic for the room sound? Do you use the high pass filter and pad built into the mic, or not? With mics, a great one will make all the difference, but even some of the best mics might not work on all sources. Experiment with what you have.

7) Mic pres are another huge subject. Most of the best classic acoustic guitar tones have been recorded through very high quality, all discrete, transformer balanced, class A preamps. These aren’t cheap, and the ones built in to your audio interface are almost certainly not as good, and won’t sound as good either. That’s not to say you can’t achieve great results with them, just that a really good pre will make a very positive difference. What is the output impedance of the mic? What is the input impedance of the pre? Are they a good match? And don’t send phantom power to your ribbon!

8) If you take anything away from this entire post, let it be this: Mic position is the single most important variable in getting the best possible sound. Most people don’t realise that moving the mic just an inch or two can change the recorded sound drastically. The only way to find where that best position is for you (see point one above), is to experiment, take notes, and listen. In pro studios you’ll often see the engineer running around at the beginning of the session, mic in hand, a pair of headphones on, wildly trying different positions in front of the acoustic player, trying to find “the spot”. This is really the only way to do it. If you can get someone else to play your acoustic in the recording room/position, have a go at this. You might be surprised how different it can sound. Try to use headphones that cut out the surrounding sound, and turn up the volume so you can mainly hear the miced sound, rather than the sound in the room. Note how the sound changes as you move the mic, where it sounds good, and which position might be best for the track. Write it down. Record some tracks. Far better to get the best sound at the source, than have to worry about EQ and other “surgical” processing later. There are many, many articles on the various “go to” mic positions for acoustic, so I’ll leave you to look those up.

9) It is considered by most professional guitarists and sound engineers that an acoustic piezo pickup is a horrible, quacky sound, to be avoided at all costs. For live sound, it’s convenient, of course, but when you’ve compared a DI’d acoustic to a well recorded miced acoustic, you are never likely to want to use the pickup/preamp/DI for recording ever again. However, for effect, why not? There are no rules in audio!

With my own experiments, I’ve drawn up info sheets such as the following:

Date: 08/12/2016

Track: The Jester’s Revenge

Instrument & Setup: Faith Naked Venus Acoustic Guitar, Standard Tuning, Broken In Strings (Martin MSP3100 12 80/20), Dunlop Prime Tone .73 plectrum

Mic/DI: CM3 mic (hyper-cardioid) & Jack Out (Bass: Flat, Treble: Flat, Volume: Max) to DI to TG2

Mic Position & Setup: Single CM3 pointed directly on axis at the 14th fret, 30cm away

Mic Pre Settings: Mic Channel 1, 55dB Gain, 300 Ohm. DI Channel 2, 35dB Gain, 1200 Ohm

Bax EQ Settings: HPF 54Hz, LPF 70kHz

Recording: 24/96

I hope this has been in some way useful to some people! There is no one answer, it’s all about the process, lots of practice, lots of experiments, and above all, lots of listening to find out what works for you, in your space, with your gear, for your tracks.

For strummed chords recently, I’ve really been digging my Advanced Audio CM48T tube large diaphragm condenser in cardioid mode, no pad, 125Hz HPF switched on, into the Chandler TG2 pre at 300 Ohm. Position at 14th fret (where neck meets body), about 18 inches back, angled at 45 degrees toward the sound hole. Sounds great! That mic/pre combo is very “coloured”, if I want things a little more flat/transparent/natural I’ll use the Line Audio CM3 small diaphragm condenser in a similar position.

Perceived Loudness & Album Flow

I was recently in discussion with a client about his album of spacey, dynamic, almost modern classical, electronic, ambient type material, and his concern about track to track level. I wrote a whole long rant about loudness and albums, which I thought some people might enjoy reading.

For a start, RMS is pretty bad as a measure of perceived loudness. It’s a mathematical measurement of power in a system, has nothing to do with the human ear, and doesn’t take into account frequency response or dynamics like the human ear does. Bass frequencies have a much higher RMS than treble frequencies, so a bass heavy track may read far higher on an RMS meter, while subjectively sounding much quieter. The new LUFS loudness standards are a step in the right direction, (taking frequency response more into account), but still don’t correlate particularly well to perceived loudness in the room, on many occasions, and were designed with Film, TV and Radio broadcast in mind, more than for serious music listening.

I usually get a “ballpark” figure going while I’m mastering, from track to track, using Integrated BS.1770 LUFS on just the loudest sections of a particular track, so that those sections are all within about 1.5dB Integrated LUFS of each other. (The actual figure I use will depend on the artist’s wishes and genre etc. What might work well for an ambient album would probably be considered way too quiet for my Hip Hop or EDM clients, for example). When every track has been mastered I go back to the whole album from start to finish and fine tune the levels by ear, because ultimately that will be the experience of the end listener. It has to sound good and flowing and coherent coming out of the speakers in the room at various different volume levels, for best translation across a wide variety of systems. I’ll also do a final QC listen on headphones.

That brings me on to the next point, which is track order. If the order is set, then that’s the order I work with, trying to make sure all the tracks flow well together, one into the next. This includes timbre, loudness, and “gapping” (the length of silence between each track). I will not focus extensively on A/Bing tracks next to each other that are not next to each other on the album. I might do so more for an album where the artist has requested it be more like a “collection of singles” than an album, or are concerned about wanting it as loud as possible overall, or that all tracks will sound similar in a randomised playlist etc., but this is not usually a request when I work on proper “album as album” projects.

I think you have to decide if you want an album where every track will sound at a similar level next to all the others, or if you are happy with a bit of dynamic ebb and flow between tracks for a better “album feel” and contrast. For me personally, Ambient albums usually benefit more from the latter approach. I can advise, but ultimately the creative decisions lie with the artist and their vision, and I will try to fulfil that in the best way I can. Overall volume changes are a pretty quick and easy tweak, as it’s mostly done in the final stage at the limiter, just before render, so it’s usually not too much work to make any changes required.

DJ Mixes

As well as mastering, I am just as passionate about playing and listening to music as I am about working on it. I started DJing as a student radio DJ at The University Of Essex in 1995, and last year had a weekly hour long show on the UK’s Future Music FM station, called “3×3” (three mixes of three tracks from three different artists or genres). These days I tend to do a mix when I feel like it and upload it to Mixcloud, and will hopefully have some live gigs coming up in the near future. I recently received a Condesa Lucia rotary DJ mixer, which is a total pleasure to mix on and sounds just fabulous (all discrete, class A, transformer balanced etc.) If you’d like to hear the latest you can always head on over to my Mixcloud page here: Gregg Hermetech on Mixcloud

Mastering Compression

One of the questions I see a lot is how to use a compressor in mastering. Of course, nothing beats a lot of time spent using a particular compressor on many different tracks to get a feel for it, but there are some techniques that can help to speed up the process. This is not the only way, but it’s one that works for me, after a lot of trial and error, and if you are mastering for punch/clarity/”glue” etc., as opposed to just squash. I will say upfront that it is almost completely paraphrased from Michael Paul Stavrou’s excellent book, ‘Mixing WIth Your Mind’, which despite some now outdated sections on digital audio, is one of the best recording books I have ever read.

So, here’s the technique: First choose a basic stereo compressor and an unmastered, highish tempo, steadily rhythmic track (e.g. something Disco, House or Techno etc).

1) Set the main controls like this: Attack as short as it will go. Release as short as it will go. Ratio as high as it will go. Threshold/Input Gain depends on the compressor type, but basically be smashing it as hard as you can. It sounds really bad, right?

2) Now start tweaking the Attack control and listen to the attack of the transients on the kicks, snares, bass line, arpeggios or whatever. You should be able to find a nice spot where it really lets the transients ‘thwap’ through whilst clamping down on everything else. Use your ears. Where does it sound best? Less snap and more smash = shorter Attack, more snap and less smash = longer Attack. I can usually find a sweet spot very quickly. I usually err on the side of a little too short rather than a little too long. Don’t touch the Attack again, you are done there. It still sounds really rubbish though, doesn’t it?

3) Move on to the Release control. We are now at the real magick. What you are looking for here is an indefinable certain something that grooves. Again, use your ears. Might be best left at minimum, might sound good thinking about the tempo of the track, so it clamps down and swings back just before the next kick (although it might not…), it might want even longer than that for a volume wave/RMS/loudness levelling like function that just “feels” right and rides the groove. It’s the ineffable control, but the most important for adding feel. So use your ears, not your eyes, not your screen, not your meters, not the compressor markings, just your ears. When you are happy, don’t touch the Release again, leave it, you are done there.

4) If the comp has a HPF in the sidechain, I usually play with it here. On full band tracks I usually find somewhere between about 60-300Hz will work to stop those kicks and bass lines pulling things down too much, but I’ll also usually go back and tweak this again later. I almost always have the HPF in the sidechain for mastering compression, very rarely off altogether. It really adds to compression transparency. ‘Suck, Thwap, Suck, Thwack’, it still sounds bad, even though it’s now “bouncing” along nicely with the music.

5) Move on to the Ratio control. Bring the Ratio control down as far as you possibly can, all the way if possible, until that groove you have established becomes almost subliminal. Mastering ratios are often very low, like 1.1:1 or 1.5:1, very rarely more than 2.0:1. You’ve now got that groove sounding really nice!

6) Threshold or Input Gain should be set so that the amount of Gain Reduction is never clearly audible. Try not to do more than about 1.5-3dB of Gain Reduction on the highest peaks. You have added a subliminal certain something to the snap and groove that the artist will hopefully appreciate. If they had wanted an obvious ‘Suck & Thwap’, they would have done it themselves, or asked you to do it. If they did ask you, that’s great, go ahead and slam it a little more by lowering the threshold, pushing the input gain harder, or raising the Ratio back up.

7) It should hopefully now sound much better, have a certain ‘je ne sais quoi’, glue, warmth, togetherness blah blah blah when you A/B it with the original track. You are done!

Notes: Play with the compressors’ extra controls, as they all work and sound slightly different. My Chandler Germanium compressors are great because they also contain two input and two output transformers you can gain stage for more or less saturation, and an output gain with feedback control to adjust the type of ‘colour’. They also offer six different transfer curves/knee types, I usually use the two or three widest knee types as they are the most transparent/least “grabby”. I also have a “Clean/Dirty” switch which changes the amount of harmonic distortion the gain reduction causes. Usually Clean, but sometimes Dirty can be just what the doctor ordered.

I hope that was somewhat helpful.