The audio files available on this page accompany the Mix Rescue column for SOS July 2012 (www.soundonsound.com/sos/jul12/articles/mix-rescue-0712.htm), featuring the instrumental track ‘Ying Yang’ by SOS reader Santi Vega. The filenames should be fairly self‑explanatory, but the descriptions below should help you understand a little more about what you’re hearing. In addition to these demonstrations, you can also download both the raw recording and my full Cockos Reaper remix project from the ‘Mixing Secrets’ Free Multitrack Download Library at www.cambridge-mt.com/ms-mtk.htm#SantiVega.
Here’s a section of the original solo guitar recording, without any processing, as sent in to Mix Rescue by SOS reader Santi Vega.
Because the basic sound of the guitar was fundamentally pretty good (on account of decent playing, a good instrument, and high‑quality recording gear), I applied very little EQ directly to the guitar channel in the final remix. However, a shelving boost of 3.5dB at 120Hz provided some flattering additional warmth at the low end (in combination with a safety high‑pass filter at 20Hz to catch any stray subsonics) while a two‑octave‑wide cut of 2dB at 235Hz addressed a touch of woolliness.
The main trouble‑shooting task for this remix was to tame some distractingly overprominent mechanical noises. I achieved this by separating out the high‑frequency region of the spectrum using Cockos Reaper’s Jesusonic 3BandSplitter and 3BandJoiner plug‑ins and then limiting that using an instance of Stillwell Audio’s The Rocket compressor running at a 20:1 ratio, operating at a 5 microsecond attack time and a 9ms release time. The threshold control was adjusted using automation, such that the processing is only active during the following sections of this example file: 0:01‑0:03, 0:07‑0:09, and 0:15‑0:17.
This example file demonstrates what the parallel compression channel added to the guitar tone and sustain. The processor I used for this was a IK Multimedia’s Fairchild 670 emulation (from their TRackS 3 plug‑in suite), set to its fastest‑acting Time Constant setting (the leftmost switch position) and registering around 7dB of gain reduction on the strongest signal peaks.
Although the parallel compression showcased in my 04_Paracomp example file does a lot of good things to the overall sound, I wasn’t happy with the way it was bloating the low end, which had already been fattened up more than enough by my initial EQ, so I inserted an instance of DDMF’s LP10 equaliser to shelve 2.6dB off the low end from around 280Hz. In addition I inserted SPL’s Transient Designer to balance the parallel channel more in favour of sustain elements than attack transients, so that the parallel channel’s return level would give me better individual control over their relative balance in the mix. In this file you can hear the results of these tweaks.
The first proper send effect I added to this remix was a version of the classic Harmonizer pitch‑shifted delay widener, implemented using Reaper’s internal ReaDelay and ReaPitch plug‑ins. This audio example shows what this effect sounded like in isolation. Although I sometimes filter out the low end of this effect in busier mixes, here I didn’t feel the need for any EQ at all.
This audio example illustrates what the widener effect soloed in the 06_ClassicWidenerSolo file sounded like in context, balanced around 15dB lower in the mix than the dry guitar signal. Compare this with 05_ParaCompTweaked to get a sense of how much this affected the stereo width of the overall panorama.
I used Universal Audio’s emulation of Lexicon’s venerable 224 digital reverb hardware unit to thicken the sustain of the guitar and further fill out the stereo picture. For this file I’ve soloed that effect return so you can hear its characteristics more clearly. The sound is based on the ‘Small Hall A’ program, but with some adjustments to the reverb decay time, and tonal modifications to both the high end (via the 224’s Treble Decay slider) and the low end (via some sub‑320Hz cuts from Reaper’s internal ReaEQ plug‑in).
The Universal Audio Lexicon 224 effect I soloed for the 08_Lexicon224Solo file can be heard in context here, and you can compare this example with 07_ClassicWidenerContext to hear the impact of adding it in. Notice the comparatively high level of this effect in the mix, which was why I carved away some of the less useful aspects of it with EQ — it felt too overbearing without this.
Although the emulated vintage Lexicon reverb demonstrated in the last two audio files was great for warming up and widening the sound, it wasn’t very good at providing a nice sense of acoustic space. For that I turned to an impulse response from TC Electronic’s System 6000 outboard processing mainframe, running in Christian Knufinke’s SIR2 convolution engine — an effect that I’ve soloed in this example. An instance of SPL’s Transient Designer knocked 9dB of transients feeding the effect, to prevent unpleasant high‑frequency ricochets in the reverb tail, while ReaEQ shelved 6dB out of the low frequencies at around 145Hz.
This example shows what the impulse‑response reverb from the 10_SpaceReverbSolo file sounded like in context. Line this up against the 09_Lexicon224Context file to hear a with/without comparison — listen particularly for the sense of height and depth the second reverb adds.
In the final stages of the remix, I did some detailed fader rides on the track to even out some phrasing moments. By comparing this audio example with the 11_SpaceReverbContext file, you can hear the effects of this automation data. The main gain changes during this example occur at 0:02‑0:09 and 0:21‑0:23, so focus on those moments.
Here’s a section of the final remix with all my processing in place except the six narrow peaking EQ cuts which I used to tame various out‑of‑balance harmonics. The most problematic frequencies for me in this section are 1062Hz (eg. 0:18‑0:20), 1188Hz (eg. at 0:32‑0:34), 1337Hz (eg.0:29‑0:31), and 1600Hz (eg. at 0:35‑0:39).
Although it’s tempting to think that the overblown pitched harmonics of the AutoEQOut file are simply unwanted processing artefacts of the multiple layers of sustain‑enhancement applied during the remix process, if you listen carefully to the same section of the raw recording, you can hear them clearly there also.
This audio example shows the same section of the final remix you heard in the AutoEQOut file, but this time four of my six automated peaking EQ cuts are targeting the four problematic pitched harmonics to bring them back into a suitable balance within the context of the instrument as a whole. Each band covered a range of around 8‑12dB, and usually only one or two of the bands were active at any one moment.
The original mix of Santi Vega’s solo guitar piece ‘Ying Yang’, which he sent in to Mix Rescue.
My remix of ‘Ying Yang’, based on the same single‑track acoustic guitar recording. I carried out the mix in Cockos Reaper with the help of some additional third‑party plug‑ins from DDMF, Christian Knufinke, Stillwell Audio, Universal Audio and IK Multimedia. Note that the remix is slightly shorter than the original, because Santi decided to cut out a small section while the remix was underway. 0