This is part of my final major project: click here to return to the main menu:
Once I felt I had got enough takes of the band playing, I then proceeded to the mixing stage. First off I went through all the takes for each instrument, and listened for any slip ups with timing or if any wrong notes where played, though since the band had played this song for a while in a live environment, their musicianship paid off as they were pretty consistent.
After this I took the multiple takes of the guitars and put them on separate tracks, so I could layer them to create a denser sounding mix, before bussing all the drum tracks via 1-2 to an auxiliary input, while doing the same with the rhythm guitars with bus 3-4.
Next It was a choice of what tracks to use in the mix, ultimately deciding not to use D112 bass track, as it didn’t have much clarity to it, sounding “fuzzy” and didn’t compliment the rest of the mix. I also choice not to use the bottom snare track or the second vocal track, as the snare sounded like white noise, while the band said while listening back in the control room after we had finished vocals, that they would of preferred it with the singular vocal for the main parts.
I then decided to tempo map the song, since the band didn’t play to a click track, where the band again showed their musician ship by keeping a relatively level tempo, ranging from 80-90bpm. This allowed me to understand the structure of the song better, being able to count how long the verses where compared to the choruses etc. while also allowing me to add or take parts away and understand where they would need to come back in.
This came in handy as the song starts off with the bass playing for a bar before the song starts, but since he couldn’t play to a click track, he wasn’t able to play it in real time because he had no point of reference, but with the tempo map I was able to find where the bar started and copy it in front of the recording, making it more consistent.
After this I moved on to EQ-ing certain tracks, starting with the kick drum, where I reduced the bandwidth around 170Hz by -7.8dB, to make the kick sound less “boxy”, while also making room for the bass, while boosting the bandwidth round 3Khz to emphasise the “clicky” bass drum pedal sound, as the drummer was using a double kick pedal, which metal recordings tend to accentuate.
Next I added a compressor to both of the individual overhead tracks, rather than adding one to the buss as this way the overheads would be compressed before they are summed together with the rest of the tracks, as if it was on the auxiliary everything would be compressed.
I used a 3:5:1 ratio to add some sustain to cymbals, so that only some of the higher frequencies would get be getting pushed down below the threshold, set at -34dB, while having an Attack Release time (here to referred as A/R) of 10/80 so the compressor stays on for the duration of the track. Finally I used the makeup gain to boost the signal back up to its original strength, set at +8..3dB.
The Room Mic I also added a separate compressor to, using the plug in version of the legendary BF-76 Hardware Compressor, as this would allow me to make the drum kit sound different since during the recording it was picking up everything simultaneously. I used the 4x Ratio setting with the input set at 0dB while having the output set at 35dB, with a 1/7 second A/R so the ambiance of the studio room was being exaggerated.
Next I moved on to the Bass DI track, where I re-amped the clean signal with the SANS AMP PSA-1 amplifier plug-in to get a more “gritty” and “crunchy” bass tone to compliment the guitars, with the setting I used show below:
by pushing the higher EQ and crunch settings, it started to sound more like a lower tuned guitar on the higher strings, while the lower strings started to sound more “gainy” and have more definition.
With the Rhythm Guitars, I panned the PG52 recorded tracks to 9 and 3 o’clock, so they where far enough away from the center while still not sounding too far apart, allowing the drums to sit in that space, as well as having the SM57 recorded tracks at 2 and 10 o’clock, sitting on top of the darker sounding 52’s giving more precision to the guitar in general and giving it a slightly larger stereo image.
On the Rhythm guitar bus I EQ-ed the tracks with a High Pass Filter (referred to from now on as HPF) with a 6dB/Octave shelf at 93.6Hz while boosting the bandwidths at 1.26KHz and 2.65Khz by 4 and 6.3dB respectfully, to clear out some of the low end which would overlap with the bass, and to boost the higher frequencies to get the high gain metal sound the song needed.
I also had two lead guitar tracks, which I had panned in the middle to so that they would be able to cut through the mix, while EQing the tracks each with a HPF using a 12dB/Octave Shelf set at 431.hz, and a Low Pass Filter (here on referred to as LPF) with a 18db/Octave Shelf at 10.15Khz. This cuts the low end so that it doesn’t interfere with the already HPF-ed Rhythm guitars but also takes off some of the extreme high end giving the guitar more mid range and a warmer tone.
In order to make these guitars sound bigger in the mix, I then used two slightly different delay settings on each to create this “wall of sound” to sit over the rhythm guitars, with each of the settings pictures below. It creates the wall of sound by each of the delay’s repeating at slightly different intervals so that they build up as a one constant sound.
Now that the instrumentation was done, I then addressed the main vocal, where I added a compressor to make it sit on top of the mix, with the gain set at +18dB at 3:0:1 ratio, while setting the A/R time at 10/80 to make sure that the compressor turned on quickly and stayed on for the duration of the track, this combined with the -18.9dB threshold meant that only the higher parts of the singers performance would be compressed down.
I then added a reverb to the channel, to make it sound less dry and to make it sound like she was projecting her voice more. This was achieved by having the diffusion setting at 87 percent, meaning that the density of the reflections will sound bigger. I had the wet/dry signal set at 30%, so that reverbed sound was stronger while still having the original vocal in the mix, using the room 1, small setting to let it blend in with the other elements of the track.
The backing vocals really only consisted of repetitions of certain lines at a conversation level, with the exception being towards the end of the song where the vocals overlap with each other before the singer goes into a falsetto to close the track. So in order to bring them up to the same level as the main vocal, I used a compressor with a similar setting, with the only difference being a slightly higher gain and a lower threshold.
I then decided to make these vocals in the earlier parts of the song sound darker and grim to reflect the mood of the song, so I EQ-ed them with a HPF and a LPF both with 18dB/Octave Shelves, set at 1.31Khz and 4.6KHz respectfully, to make them sound more like they were coming through a telephone or radio.
This was further supplemented with a AIR distortion plug-in with a 12dB drive and a 100% distortion setting to create almost “voices in static.” because I EQ-ed it first, I meant that the Distortion didn’t end up sounding muddy, as now it was only effecting a selecting frequency bandwidth.
Though after listening to the song back, I realised that during the bridge the vocals needed to sound the same as the main vocals, so I added both a master bypass for the EQ and the distortion, adding them to the automation list so I could have a specific point they would turn off, another reason tempo mapping the song came in handy.
In order to create a more dynamic mix, I then used panning automation on the vocals to make them sound like whispers in the listeners ears, or intrusive thoughts, further adding to the “Get out ” song theme.
I then added some LPF filtered white noise sweeps in certain sections of the mix to exaggerate certain crescendos. This was done the same way as my For the Best mix, by bussing an aux input with a signal generator to an audio channel and recording the white noise. This was then automated and flanged in the same way to get the desired effect.
The final addition to the song which I added was a siren sound I downloaded off Freesound.org, which is a creative commons licensed sound archive, allowing me to use for non profitable gain, though I could of recorded an air-raid siren if I had access to one, this was a more realistic option. The band where using this as a promotional single and demo, so I doubt they would have any copyright issues with its release.
I placed it at the start of the song, with the characteristic of the sound itself creating a slow fade in, while fading out the end so that it finishes before the bulk of the song starts.
Now it was just a case of automating some of the volume to give more attention to certain parts of the song, this started with the lead-guitar volume, having it louder during the segments it’s in the song, while fading during the second chorus, only to be boosted for the final solo segment.
I also dropped the rhythm guitar level in bridge segment to make the vocals the focal point of the song, before getting to my personal favourite edit in the song, muting everything but the vocals for a split second before everything crashes back in for the finale. This was done by selecting all the tracks and using tab to transient to make sure the section does stop midway through a bar, and then using command and M to mute the segments except the vocals.
I then added a stereo master fader and set the level at -0.3dB to make sure all my gain staging wasn’t causing the master fader to overload, adjusting levels where necessary , while adding a AIR frequency shifter plug in to try and emulate that speeding up tape sound. I used the automation controls to get the song to sweep up towards the end, adding a subtle finish to the track as the guitar rings out.