Overheads

Inserts: Sonnox Oxford EQ, PSP Vintage Warmer.

The small live room in which the drums were recorded resulted in an overheads recording that had a few flaws (Track 33.7). The small-room ambiance captured on the recording was unappealing. In addition, the individual kick, snare, and tom tracks all suffered from poor interaction with the overheads. For these reasons, the cymbals-only approach was chosen for the overheads. First, they went through the Oxford EQ, where a steep HPF [400 Hz, 30 db/oct] goes as high as it can to roll off all the lows (Track 33.8). In addition, a parametric filter [702 Hz, –8.37 dB, Q 3.08] further attenuates the snare and removes an annoying mid-range noise (Track 33.9). The EQ is followed by the Vintage Warmer [drive +12.9 dB, knee 41.1 percent, speed 35.8 percent, auto release, mix 78 percent], where parallel compression tightens the sound (Track 33.10). The level of the overheads was automated to rise during the last drum bar. Also, their stereo image was narrowed to give a tighter impression and to clear space on the extremes for the distorted guitars.

Kick

Kick 1 Inserts: Toontrack EZdrummer, Digidesign Smack!, Sonnox Oxford EQ.

Kick 1 Sends: Ambiance Reverb (UAD DreamVerb).

Kick 2 Inserts: Toontrack EZdrummer, Digidesign Smack!, Sonnox Oxford EQ.

The multitrack contained both kick-in and kick-out tracks (Tracks 33.11 and 33.12). Having auditioned these tracks while working on the rough mix, it quickly became apparent that replacing the recorded kick with triggered samples would not only provide easier material to work with, but would also yield a better sound altogether. The kick on the rough mix is already triggered. The triggering process involved importing the kick-in track into Logic and converting it to MIDI using the “Audio to Score” feature. The resultant MIDI track was then loaded on an instrument track in Pro Tools, with EZdrummer as a virtual instrument. The kick sound used is the default 22” GMS Felt Beater from the Pop/Rock kit.

In the final mix, there are two kick tracks. The two never play together and the toggling between them sharpens the more powerful moments in the song, as well as adding some interest. Both kick samples are identical and both are processed using the same plugins, but the settings on these plugins are different.

Kick 1 (Track 33.13) is the more rounded, settled kick and it plays through most of the verses and during the break. It is first compressed using Smack! [norm mode, input 5, ratio 2:1, attack 4.1, release 0.8], which adds some punch and mass (Track 33.14). Then it goes through the Oxford EQ, where four parametric filters shape the sound of the kick: the first filter [83.3 Hz, 3.24 dB, Q 2.83] adds some thud (Track 33.15), the second filter [220 Hz, –5.26 dB, Q 2.57] attenuates tapping low-mids (Track 33.16), the third filter [1,003 Hz, 5.6 dB, Q 2.83] adds attack (Track 33.17), and the fourth filter [5 kHz, –2.87 dB, Q 1.93] reduces click (Track 33.18). This kick is also sent at a low level to the ambiance reverb, partly to soften it, partly to blend it into the ambiance, partly to move it farther back in the mix, and partly to add to its timbre (Track 33.19).

Kick 2 (Track 33.20) is the more powerful kick between the two, having more snap and presence. It plays during the choruses, bridge, outro, and some verse parts. It is first compressed using Smack! [norm mode, input 6.4, ratio 6:1, attack 10, release 0], which adds tightness and punch (Track 33.21). Also, the distortion facility in Smack! [odd__+ even] was used to add some grit (Track 33.22). The compressor is followed by the Oxford EQ, where four parametric filters shape the sound of the kick: the first filter [110.5 Hz, 3 dB, Q 2.83] adds thud (Track 33.23), the second filter [260 Hz, –5.78 dB, Q 1.6] reduces unnecessary mids (Track 33.24), the third filter [938 Hz, 11.45 dB, Q 4.83] adds attack (Track 33.25), and the fourth filter [5.1 kHz, 2.4 dB, Q 3.9] adds click (Track 33.26).

Snare top

Inserts: PSP Neon, Digidesign DigiRack EQ 3.

Sends: Ambiance Reverb (UAD DreamVerb).

Having tried gating, compression, and triggers, I came to the conclusion that equalization alone was the most suitable approach for the snare top. A healthy amount of hi-hats and ride spill on this track made the cymbals unstable once compression or gating had been applied, and a fully triggered snare sounded too mechanical. Not gating the snare meant that spill on the track limited equalization possibilities, mainly with regard to the highs, where the hi-hats and the ride roam. On the other hand, there was some benefit in the support the hi-hats got from the snare track.

Snare top (Track 33.27) first goes through Neon, which in linear-phase mode shapes its sound. A low-shelving filter [1.1 kHz, –5.9 dB] attenuates an excess of lows and low-mids (Track 33.28), a parametric filter [312 Hz, –4 dB, Q 1.8] attenuates a resonant tone (Track 33.29), and another parametric filter [2.5 kHz, –5 dB, Q 2.9] attenuates some cymbals spill without harming the snare timbre too much (Track 33.30). On the DigiRack EQ, a high-shelving filter [4.6 kHz, –1.8 dB] softens the snare and reduces the cymbals spill (Track 33.31). The high-shelving filter was automated to only take effect during the verses— in other sections, the snare becomes brighter and the cymbals slightly louder. The sole reason for the addition of another EQ was that automating it was more straightforward than automating the Neon.

Snare top is sent to the ambiance reverb, which gels it to the sound stage and shifts it slightly backward (Track 33.32). The level of this track was automated to balance specific hits that were either too loud or too quiet. The track is panned to approximately 10:30.

Snare bottom

Inserts: Digidesign DigiRack EQ 3, Sonnox Oxford Dynamics.

Sends: Snare Reverb (Audio Ease Altiverb, McDSP Channel G Dynamics).

Snare bottom has somewhat more importance in this track since snare top did not deliver a complete snare sound. Snare bottom complements snare top and contributes both presence and crispiness (Tracks 33.33 and 33.34). The raw track (Track 33.35) first goes through the DigiRack EQ, where an HPF [313 Hz, 24 dB/oct] rolls off much of the kick spill and the snare’s body, which is already contributed by the snare top (Track 33.36); a parametric filter [1 kHz, 5.1 dB, Q 1] adds some presence and attack (Track 33.37), and a high-shelving filter [6 kHz, 4 dB] brightens the snare and accentuates its crispness (Track 33.38). Following the EQ, the Oxford Dynamics was inserted, where both the gate and the compressor are employed. The gate [threshold –19.2 dB, range –40 dB, attack 0.01 ms, hold 30 ms, release 11.2 ms] removes spill, notably the snare rattle caused by kick hits (Track 33.39). The compressor [threshold –18 dB, ratio 2:1, soft-knee 40 dB span, attack 5.2 ms, release 130 ms] adds some weight and density (Track 33.40). Snare bottom is panned to the same position as snare top.

Snare bottom is sent to Altiverb, and the IR used is the “Plate Short” from the EMT 140 (Wendy Carlos) category (Track 33.41). The only modification on the reverb is a 9.2 dB boost on the equalizer’s treble (Track 33.42), which gives the reverb more shine. The reverb is gated by the Channel G Dynamics, with its key input set to the dry snare (Track 33.43). This gated reverb was added as an effect, but also in order to send the snare farther back.

Snare triggers

Main Track Inserts: Digidesign Smack!, Digidesign DigiRack EQ 3.

Support Track Inserts: Digidesign DigiRack EQ 3, McDSP Compressor Bank CB2.

A triggered snare track was generated using Trillium Lane Labs Drum Rehab. From its integrated library, the sample chosen was the DW1 snare from the “Perfect Drums” collection. Although ultimately the snare sound is based on the recorded snare-top track, the triggered snare is still mixed, but only as a layer underneath the recorded snare. The triggered snare is mixed in all sections of the song apart from the verses. Its main role is to add brightness and definition to the recorded snare sound (Tracks 33.44 and 33.45). Also, since the velocities of the triggered hits are nearly identical, they balance out to some extent the overall snare hits.

The raw triggered snare (Track 33.46) had a decay that was too long and yielded a fluffy overall snare sound. So it is first treated by Smack! [norm mode, input 6.3, ratio 3:1, attack 5.8, release 4.7], which adds punch using moderate attack but also attenuates the decay using moderate release (Track 33.47). The compressor is followed by the DigiRack EQ, where a combination of an HPF [262 Hz, 12 dB/oct] and a deep dip on a parametric filter [427 Hz, –14 dB, Q 1] removes all the snare body and tone (Tracks 33.48 and 33.49); a high-shelving filter [6 kHz, 1.7 dB] adds a touch of highs as part of tonality shaping (Track 33.50).

From the triggered snare track, a few hits were copied onto a new track. These provide extra support during the final sections of the song, mainly boosting the overall snare level but also causing a subliminal change to the snare tonality. These few hits appear right before the break, right after it, halfway through the outro, and at the very end. These hits (Track 33.51) are first treated with the DigiRack EQ, where an HPF [476 Hz, 6 dB/oct] rolls off much of the lows (Track 33.52); a high-shelving filter [4.61 kHz, –3.4 dB] attenuates the highs so these hits are not be over-present (Track 33.53). The equalizer is followed by the CB2 [threshold –30.6 dB, comp 7.6, knee 15, attack 70 ms, release 273 ms], which adds attack (Track 33.54).

Hi-hats

Inserts: Digidesign DigiRack EQ 3.

Having the hi-hats sufficiently present on the overheads and snare top meant that there was little point mixing the hi-hats track. Still, the hats could use a little push during the break, so this was the only place where this track was mixed (Track 33.55). The hi-hats only go through the DigiRack EQ, where an HPF [668 Hz, 18 dB/oct] removes kick spill and low-mids that contributed little to the sound (Track 33.56), and an LPF [6.8 kHz, 6 dB/oct] eases brittle highs (Track 33.57).

Ride

As there was sufficient presence on the overheads, this track was omitted.

Toms

First Layer Inserts: McDSP Compressor Bank CB2, Digidesign DigiRack EQ 3.

First Layer Sends: Ambiance Reverb (UAD DreamVerb).

Second Layer Inserts: McDSP FilterBank E2, Digidesign Smack!

Like the kick, the toms in the mix are triggered. But at one specific section, just before the break (Tracks 33.58 and 33.59), the recorded toms did an exceptional job. The tom sound in this section is the outcome of two layers (Tracks 33.60–33.62); both involve the original tom recording (Track 33.63), but each is processed differently. It is the level balance between these two layers that dictated the final toms sound.

The first layer first goes through the CB2 [threshold –37.3 dB, comp 3.4, knee 13.7, attack 70 ms, release 40.2 ms], which was employed to condense the natural-sounding dynamics (Track 33.64). It then goes through the DigiRack EQ, where a parametric filter [3.5 kHz, 5 dB, Q 0.81] adds attack and highs that let the toms push more through the mix (Track 33.65). This layer is also sent to the ambiance reverb—although the reverb does not handle the toms very well, once mixed it adds a rich sense of warm space and a touch of chaos, which works well (Track 33.66).

The second tom layer first goes through the FilterBank E2, where a low-shelving filter [968 Hz, –12 dB] drys the lows and low-mids, leaving only the presence and attack portions of the toms (Track 33.67). The EQ is followed by Smack! [input 7, ratio 6:1, attack 8.5, release 6.1], which accents the attack and sharpens the tom’s dynamics (Track 33.68).

Tom triggers

Tom 1 Inserts: Digidesign Smack!, McDSP FilterBank E6.

Tom 1 Sends: Ambiance Reverb (UAD DreamVerb).

Floor Tom Inserts: Digidesign Smack!, McDSP FilterBank E6.

Floor Tom Sends: Ambiance Reverb (UAD DreamVerb).

With the exception of the short section above, all other tom hits are triggered. This allowes the toms to project better and provides ultimate control over their sound.

Tom 1 is used several times during the song. The tom hit (Track 33.69) was extracted from isolated hits recorded in the studio after the drums were tracked. These hits were pasted into the session manually using the tab-to-transient feature. They first go through Smack! [input 8.3, ratio 6:1, attack 3.9, release 0], where a moderate attack adds some punch while zero release lengthens the decay (Track 33.70). The distortion facility on Smack! [odd] was also utilized to add a touch of grit (Track 33.71). The compressor is followed by the FilterBank E6, where three bands are used: an HPF [135 dB, 12 dB/oct] sets the low frequency limit of the tom, essentially preventing it from overloading the lows, including those of the reverb it is sent to (Track 33.72); a parametric filter [648 Hz, –2.6 dB, Q 0.8] shapes the tonality of the tom while removing dispensable mids (Track 33.73); and another parametric filter [4.35 kHz, 2 dB, Q 1] accents the attack (Track 33.74). Tom 1 is sent to the ambiance reverb, mainly in order to detach it from the front of the mix (Track 33.75).

The floor tom (Track 33.76) only plays once just before the bridge. Like tom 1, it was also extracted from the isolated hits recorded in the studio. It is compressed quite similarly to tom 1 by Smack! [input 8.3, ratio 6:1, attack 6, release 0], which adds attack and lengthens the decay (Track 33.77). On FilterBank E6, which follows, a single parametric filter [4.9 kHz, 6.9 dB, Q 0.9] accents the attack and definition of the tom (Track 33.78). The floor tom is sent to the ambiance reverb, intentionally creating some distant thunderous thud (Track 33.79).

Tambourine

Inserts: Toontrack EZdrummer, Digidesign DigiRack Dyn 3.

Sends: Ambiance Reverb (UAD DreamVerb).

Break Tambourine Inserts: Toontrack EZdrummer, Sonnox Oxford Reverb.

The tambourine in “Hero” was never recorded—instead, it was generated using EZdrummer during mixdown. It plays through most of the song except the verses, and was added mainly to sharpen the contrast between the verses and the choruses. It was mixed as a support instrument, meaning it is not clearly defined (Tracks 33.80 and 33.81). The untreated tambourine (Track 33.82) was compressed by the DigiRack Dyn 3 [threshold –12.6 dB, ratio 5.3:1, soft-knee 24.7 dB span, attack 40.6 µs, release 5.7 ms], which balances out the loud strokes with the quiet shakes (Track 33.83). It is sent to the ambiance reverb, which sends it far back into the depth field (Track 33.84). Before the verses, the tambourine is panned left opposite the ride, which plays along with it. From the bridge onward, the tambourine plays with the hi-hats, which appear left. To prevent the hi-hats from masking the tambourine and in order to create a more balanced stereo panorama, the panning of the tambourine was automated to shift to the right from the bridge onward.

During the break, the tambourine’s nature alters, and because of this a new track was created. The only plugin inserted on this track is the Oxford Reverb [67% wet], where a modified version of the “Concert Room” preset adds a long and shiny tail (Tracks 33.85 and 33.86). This track, which is panned center, was automated in level to rise toward the end of the break, building up to the explosive outro.

Main bass

Inserts: PSP Vintage Warmer, Sonnox Oxford Dynamics, Digidesign DigiRack EQ 3, SoundHack +Chebyshev.

As I was growing to like the bass sound in the rough mix, most of the plugins and their settings were transported from the rough mix to the final mix. The raw bass (Track 33.87) first goes through the Vintage Warmer [mix 68.4 percent, drive 11.6 dB, knee 31.9 percent, speed 91 percent, auto release], which has two roles: it condenses the bass dynamics using parallel compression and adds warm saturation. This saturation yields added harmonics that sharpen the bass definition (Track 33.88). Following the Vintage Warmer is the Oxford Dynamics, but the only facility used on it is Warmth [89 percent], which adds some clarity while also enlivening the sound (Track 33.89). Then comes the DigiRack EQ, where three bands are utilized: an HPF [72.4 Hz, 24 dB/oct] was employed to remove low frequencies that jumped with very low notes, making the bass tonality inconsistent (Track 33.90); a parametric filter [206.5 Hz, –1.6 dB, Q 1] was added late in the mixing stage to ease the mids on the bass as part of frequency tuning (Track 33.91); and another parametric filter [636 Hz, 3.8 dB, Q 1] simply shapes the sound of the bass by accenting low-order harmonics (Track 33.92). Following the EQ comes the +Chebyshev, which adds even more distortion and a very distinct size (Track 33.93).

Break bass

Inserts: PSP Vintage Warmer, Digidesign DigiRack EQ 3, SoundHack +Chebyshev, Sonnox Transient Modulator.

During the second part of the break, the bass plays an important solo role, which builds up to the outro. This called for a deviation from its normal sound. So the bass during the break was moved to a new track, which was processed differently from the main bass. This track was first treated with the Vintage Warmer [mix 100 percent, drive 20.9 dB, knee 31.9 percent, speed 91 percent, auto release], which tightens the sound while also adding some warmth (Tracks 33.94 and 33.95). Then comes the DigiRack EQ, where an HPF [124 dB, 18 dB/oct] intentionally dries the lows from the bass, with the idea that these will return with the outro (Track 33.96); a parametric filter [340 Hz, –7.5 dB, Q 2.24] attenuates disturbing mids (Track 33.97); another parametric filter [497 Hz, 8.8 dB, Q 10] opens up the flat sound and adds some resonance as an effect (Track 33.98); and yet another parametric filter [2.46 kHz, 10.8 dB, Q 4.06] draws some presence (Track 33.99). Next comes the +Chebyshev, which adds an extra dimension to the sound (Track 33.100). Last comes the Transient Modulator [ratio 0.25, overdrive 90 percent], which accentuates attack and adds a subtle amount of harmonic content (Track 33.101).

Bass FX

Inserts: Digidesign DigiRack EQ 3, SansAmp PSA-1, McDSP Chrome Stack, Digidesign DigiRack EQ 3.

As part of early tryouts, I experimented with the idea of having a heavily distorted bass— an idea that did not quite work well in the context of this song (Track 33.102). Yet I had the distorted layer (a duplicated track of the bass going through the following plugins) and it worked quite well during the powerful outro of the song—adding extra power to the bass and to the overall mix (Tracks 33.103 and 33.104). The first plugin in the chain is the DigiRack EQ, which acts as a tuner for the subsequent PSA-1 (Tracks 33.105–33.107). Two bands are operational on the EQ: an HPF [194.2 Hz, 24 dB/oct] and a high-shelving filter [8.81 kHz, –6 dB]. The PSA-1 is followed by the Chrome Stack, which contributes the main effect (Track 33.108). As part of the final frequency tuning, another DigiRack EQ was inserted, where a parametric filter [829.6 Hz, –2.9 dB, Q 1] pulls some mids (Track 33.109).

Intro/break rhythm guitar

Inserts: Digidesign DigiRack Dyn 3, SansAmp PSA-1, Digidesign DigiRack EQ 3, Sonnox Oxford Reverb.

Sends: Ambiance Reverb (UAD DreamVerb).

The intro/break guitar, a DI recording (Track 33.113), first goes through the DigiRack Dyn 3, where the compressor [threshold –26.4 dB, ratio 9.7:1, soft-knee 30 dB span, attack 1 ms, release 15.1 ms] evens out level fluctuations (Track 33.114). It is followed by the PSA-1, which adds some color to the pale DI sound while also adding some crunch (Track 33.115). Next comes the DigiRack EQ, where a low-shelving filter [230.6 Hz, –7.2 dB, Q 1] attenuates an excess of lows (Track 33.116), then a wide parametric filter [2 kHz, 7.4 dB, Q 1] adds essential highs to fabricate the final guitar sound (Track 33.117). The Oxford Reverb was inserted in series to add some stereo size to the mono recording. The reverb was programmed to only output early reflections with very little (8 percent) of the wet signal mixed (Track 33.118). In addition, this track was sent to the ambiance reverb in order to place it in the mix space (Track 33.119).

The intro guitar plays for four bars alone before other instruments are introduced—an introduction that in my opinion did not have sufficient power. So the level of the intro guitar was set intentionally at a level low enough to allow some level impact once other instruments are introduced (compare Tracks 33.120 and 33.121).

Main rhythm guitar

Inserts: Digidesign DigiRack EQ 3, Digidesign DigiRack Dyn 3.

Sends: Guitar Chorus (UAD Delay Modulator DM-1), Ambiance Reverb (UAD DreamVerb).

By the time all the tracks had been brought up and coarsely mixed, there was very little space for the rhythm guitar. Its definition is in the high-mids—an area already occupied by virtually all other instruments but the bass. Luckily, a rhythm guitar need not be in the limelight as it plays a rather supportive role in the mix.

The raw track (Track 33.122), a DI recording, went through aggressive equalization using the DigiRack EQ. An HPF [648.7 Hz, 18 dB/oct] rolls off a good portion of the lows and low-mids, which contributed little to the guitar sound yet cluttered a valuable range of frequencies (Track 33.123). A parametric filter [1.1 kHz, –8.3 dB, Q 1] in combination with another parametric filter [1.83 kHz, –9.5 dB, Q 3.79] tunes the guitar into the busy high-mids (Tracks 33.124 and 33.125). Then a high-shelving filter [6 kHz, –2.7 dB] treats exaggerated highs (Track 33.126). The EQ is followed by the DigiRack Dyn, where a compressor evens out level fluctuations (Track 33.127). The guitar is sent to the DM-1, which by adding chorus effect sends the guitar back in the depth field and also creates a wide curtain of sound (Track 33.128). The guitar is also sent to the ambiance reverb, which gels it into the sound stage (Track 33.129).

Support rhythm guitar

Inserts: McDSP FilterBank E6.

Sends: Ambiance Reverb (UAD DreamVerb).

The main rhythm guitar was panned to the right, yet, despite the stereo effects added, something was still missing on the left. To combat this, a different take—an SM57 recording—was mixed panned left. Combined with the main rhythm guitar, this additional track creates an even wider and richer impression (Tracks 33.130 and 33.131). The raw track (Track 33.132) was only equalized using the FilterBank E6, where an HPF [1.23 kHz, 6 dB/oct] rolls off dispensable lows and low-mids (Track 33.133). It is sent to the ambiance reverb, which places it on the same line as the main rhythm guitar (Track 33.134).

Distorted guitars

When it came to mixing the distorted guitars, a few things had to be considered. First, during the verses, there are intensity changes between sections that involve the vocal and those involving the lead guitar. To support these changes, the distorted guitars are split into two pairs: one pair, which I term “curtain guitars,” play during the relaxed sections of the verses; the other pair, simply termed “power guitars,” play during all other sections. The second consideration was that the power guitars in the choruses, bridge, and outro should stand out more than those during the verses.

The distorted guitars were recorded using the same DI, SM57, and U87 setup as the rhythm guitar, and here again the U87 was not used. The DI recordings that were used were processed by amp simulators—a process that allowed ultimate control over their final sound.

Curtain guitars

Guitar 1 Inserts: UAD Preflex.

Guitar 2 Inserts: Digidesign DigiRack Mod Delay II, Digidesign DigiRack EQ 3.

Sends: Ambiance Reverb (UAD DreamVerb).

Curtain guitar 1 is a DI recording (Track 33.135) processed by the UAD Preflex with a modified version of the “Foxy Gravy” preset (Track 33.136). This track was panned hard-left.

Curtain guitar 2 is an SM57 recording of the same take as guitar 1. It is panned hard-right. The raw track (Track 33.137) goes through the DigiRack Mod Delay [mix 50%, 24 ms delay, no feedback or modulation] for the sole purpose of sending it backward without adding stereo width (Track 33.138). On the succeeding DigiRack EQ, a low-shelving filter [175.5 Hz, –8 dB] attenuates an excess of lows (Track 33.139).

Both tracks are routed to an audio group, which is sent to the ambiance reverb. The reverb shifts the guitars farther back in the mix and also detaches them from the extremes (Track 33.140). This creates a nice contrast between the curtain guitars and the power guitars— the curtain guitars appear farther back and more centered; the power guitars appear farther in front and their image is bound to the extremes. The level of the curtain guitars was automated so they only appear in the relaxed section of the verses.

Power guitars

Guitar 1 Inserts: SansAmp PSA-1, SansAmp PSA-1, Digidesign DigiRack Mod Delay II.

Guitar 2 Inserts: Digidesign DigiRack EQ 3, Digidesign DigiRack Mod Delay II.

Power guitar 1 is a DI recording (Track 33.141). The first plugin in the inserts chain is the PSA-1, which is bypassed during the verses (Track 33.142). It is followed by another PSA-1, which operates throughout (Track 33.143). While the second PSA-1 plays alone during the verse, in all other sections it is combined with the first PSA-1— a combination that creates a richer, brighter, and more powerful sound (Track 33.144). Next comes the Mod Delay [36 ms delay on right channel only], which, by applying the Haas trick, turns the mono track into stereo while adding a noticeable size (Track 33.145).

Power guitar 2 is an SM57 recording of a different take than the one used for power guitar 1. The raw recording (Track 33.146) goes through the DigiRack EQ, where a low-shelving filter [142.4 Hz, –4.4 dB] attenuates overemphasized lows (Track 33.147) and a parametric filter [2.5 kHz, 2 dB, Q 1] adds a touch of presence (Track 33.148). The guitar then goes through the Mod Delay [57.23 ms delay on the left channel only], which turns the mono track into a wide stereo one; modulation on the delay [depth 3%, rate 1.26 Hz] adds further dimension and size (Track 33.149).

Distorted guitar group

Inserts: Sonnox Oxford EQ.

All distorted guitar tracks were routed to an audio group. On the group track, the Oxford EQ is responsible for tuning the guitars to the frequency spectrum. On the EQ, four bands are employed: an HPF [288.2 Hz, 18 dB/oct] sets the low-frequency limit of the guitars (Tracks 33.150 and 33.151); a parametric filter [600 Hz, –2.75 dB, Q 6.25] attenuates some mids that the guitars could live without and other instruments could use (Track 33.152); for the same reason, another parametric filter [3.7 kHz, –2.75, Q 2.22] clears some highs (Track 33.153); and then a high-shelving filter [9.4 kHz, –6.28 dB] pulls some harsh highs (Track 33.154).

Lead guitar

Inserts: Digidesign DigiRack Dyn 3, Digidesign DigiRack EQ 3, PSP Nitro.

The selected source for the lead guitar was the U87 recording (Track 33.155). It is first compressed using the DigiRack Dyn [threshold –48.9 dB, ratio 3:1, hard-knee, attack 272.7 µs, release 13.6 ms] in order to balance out level fluctuations and also to increase sustain (Track 33.156). The compressor is followed by the DigiRack EQ, where a steep HPF [448.6 Hz, 24 dB/oct] rolls off dispensable lows and low-mids (Track 33.157); a parametric filter [2.09 kHz, 4.5 dB, Q 1.38] adds some flesh (Track 33.158); and another parametric filter [5.96 kHz, –4.1 dB, Q 3.23] eases some harshness, resulting in an overall smoother, more rounded sound (Track 33.159). The sound of the lead guitar is mostly the outcome of the Nitro and its “No-Fi” preset—essentially a combination of bit reduction and sample rate reduction followed by an LPF (Tracks 33.160 and 33.161). The lead guitar is sent to the ambiance reverb, mainly in order to shift it backward in the depth field and to blend it into the sound stage (Track 33.162). The lead track level was automated at various sections of the song, for example to rise during the intro.

Lead guitar FX

Inserts: Digidesign DigiRack EQ 3, Digidesign Smack!, PSP 84.

Another element that was transported from the rough mix was the lead guitar effect. To prevent correlation to the lead track, the original lead track (Track 33.163) was duplicated and processed as follows: first, it is equalized by the DigiRack EQ, where an HPF [623.3 Hz, 6 dB/oct] rolls off dispensable lows (Track 33.164) and a high-shelving filter [6 kHz, 3.6 dB] adds some spark (Track 33.165). Smack! [norm mode, input 5, ratio 4:1, attack 2.9, release 5] is then in charge of balancing level variations and lengthening the sustain (Track 33.166). The wah-wah-like effect is the outcome of a modified “Sarawak” preset on the PSP 84 (Track 33.167). This track is also sent to the ambiance reverb (Track 33.168).

Lead vocal

Inserts: Pultec EQH-2, McDSP MC2000 MC2, SoundHack +Chebyshev, Sonnox Oxford Dynamics.

Sends: Ambiance Reverb (UAD DreamVerb), Vocal Reverb (UAD DreamVerb).

There was a conceptual issue with mixing the lead vocal. On the one hand, “Hero” has its share of aggressiveness; on the other, it has its pop characteristics and involves a catchy melody. So the main question was: Should the lead vocal be sweet and beautiful, or should it be bad and dirty? It ended up somewhere in between.

The first plugin in the inserts chain, the EQH-2, was employed to shape the basic tonality of the vocal. It removes dispensable lows [100 Hz, attenuation 4.9] and adds a touch of spark [10 kHz, boost 1.5] (Tracks 33.169–33.171).

Following the EQ is the MC2000— a multiband compressor. I tried two different compressors (Smack! and the Oxford Dynamics) prior to trying the MC2000. Smack! gave an obvious character that I found was too much in the context of the song; Oxford Dynamics did a fair job, but not a perfect one. Dan’s voice fluctuates noticeably during some sections of this song. Whatever settings I tried, the level of the vocal never seemed stable throughout the song. The fundamental problem was that the frequency content of the recording changes quite noticeably in relation to the pitch sung. Lower-pitch notes seemed to disappear in the mix—making the vocal come and go. This issue is not uncommon, but on this production it seemed extremely tricky. For this reason, I picked the MC2000. Perhaps the most important setting on the compressor was the crossover frequency, which ended up at 1.6 kHz. Both bands are heavily compressed, with the threshold of the low band set to –24 dB and the threshold of the high band to –44 dB. In addition to their standard use, the independent gain controls of each band were used to further shape the tonality of the vocal (Tracks 33.172–33.174). The MC2000 added quite some character to the vocal, which I liked.

Following the MC2000, the vocal sounded balanced but still did not stand out as much as it should have. This was mostly the outcome of missing vitality on the mids. The +Chebyshev did a perfect job of filling this missing range. It also added some aggression in the form of distortion that at points becomes rather obvious—an effect that I thought was for the better (Track 33.175). The last plugin in the chain is the Oxford Dynamics [threshold –12 dB, ratio 2:1, soft-knee 40 dB span, attack 0.52 ms, release 5 ms], which acts as a catch-all compressor to perfect the balancing of the distorted voice (Track 33.176).

The lead vocal is sent both to the ambiance reverb and to a dedicated vocal reverb. The ambiance reverb blends the vocal into the sound stage, while the vocal reverb—a slightly modified “PLATE 140 Vocal Plate” preset—adds a rich effect, life, and size (Tracks 33.177–33.179). The send level to the vocal reverb was automated to rise for the duration of a few sustained vowels (Track 33.180).

Vocal FX

The lead vocal involves a few effects. First, there is a narrowed ping-pong delay after the first chorus. This delay was not produced by a plugin, but by crossed bus routing in Pro Tools—left bus sent to right bus, right bus sent to left bus. On each bus there is 830 ms of delay, which, despite not being perfectly synced to the tempo, sat quite right. In addition, the echoes were treated by the DigiRack EQ 3, which was inserted on each bus with a combination of an LPF and a parametric filter that creates a resonant filter. The send level on these buses was automated so only three echoes are produced—a fourth echo would clash with the verse vocal. The echoes are blended into the depth field by the ambiance reverb to which they are sent (Track 33.181).

The ping-pong delay closing the first chorus fills some space between the chorus and the second verse. The second chorus is followed by an additional vocal phrase and then the lead guitar immediately takes the stage, so there was little place for the same ping-pong effect. The only processing on the vocal phrase after the second chorus is the SansAmp PSA-1 (Track 33.182).

Halfway through the bridge, the phrase “deep within us” is also processed, this time using four plugins. The first plugin, which is the main effect contributor, is the PSP Master Q. The settings involve an HPF at 96 Hz, an LPF at 3.5 kHz, and a mad boost of 24 dB at 981 Hz. The actual effect is also the outcome of the hard saturation on the saturation facility (Tracks 33.183 and 33.184). The Master Q is followed by the Lexicon PSP 42, which adds 172 ms delay with feedback (Track 33.185). Then comes the Digidesign D-Verb set to large hall and 19% wet (Track 33.186). Finally, the McDSP Chrome Tremolo adds a subtle tremolo effect (Track 33.187). This specific phrase is sent to the ambiance reverb (Track 33.188).

The “deep within us” phrase shortly after repeats. This second instance could have been processed in the same manner, but to create some variation and contrast a different effect was used: the lead vocal sends to the ambiance, and vocal reverbs are muted, causing forward shifting of the voice (Track 33.189).

Fake second voice

Inserts: Lexicon PSP 42, Digidesign DigiRack EQ 3.

Sends: Vocal Reverb (UAD DreamVerb).

One of the last overdubs to be recorded, at around 6 a.m., was the second voice. By that point, Dan’s vocal cords were a few takes away from snapping. Second voice harmony was recorded for all but the verses, but the performance during the choruses and the outro was pretty unconvincing. As a replacement, the lead vocal track was duplicated, trimmed in the right places, and processed with ADT settings on the PSP 42 [15 ms modulated delay, no feedback] (Tracks 33.190–33.192). Following the PSP 42 comes the DigiRack EQ, where an HPF [239.4 Hz, 18 dB/oct] rolls off dispensable lows (Track 33.193) and a parametric filter [1.79 kHz, 7.6 dB, Q 2.81] boosts some presence (Track 33.194). The track is sent to the vocal reverb, which sends it backward in the depth field and also correlates it to the lead vocal (Track 33.195). To distinguish it from the lead vocal, it was panned left to around 9:00. As part of a gradual buildup, the second voice is only introduced during the second verse.

Real second voice

Inserts: PSP 84, Sonnox Oxford Dynamics.

Sends: Vocal Reverb (UAD DreamVerb).

The only section in which the recorded second voice is featured is the bridge. The original recording (Track 33.196) was first processed with the PSP 84, but no delay is actually applied. Instead, the filter, drive, and plate reverb were used to create a unique effect (Track 33.197). Following the PSP 84 is the Oxford Dynamics, where a compressor [threshold –26.5, ratio 3.48:1, soft knee 30 dB span, attack 0.52 ms, release 5 ms] contains level fluctuations (Track 33.198). The vocal is sent to the vocal reverb (Tracks 33.199 and 33.200).

Ahh BV

Inserts: Digidesign DigiRack EQ 3, McDSP Compressor Bank CB1, PSP Nitro.

Audio Group Inserts: Digidesign DigiRack Mod Delay II, McDSP FilterBank E2, SoundHack +Chebyshev.

Two backing vocals tracks were recorded for the bridge (Track 33.201) and, like the second voice performance, they are not the greatest. A conscious decision was made to bury these in the mix. The two tracks were processed identically and then routed to an audio group for additional processing. The first plugin is the DigiRack EQ, where an HPF [260.6 Hz, 12 dB/oct] rolls off dispensable lows (Track 33.202). It is followed by the CB1 [threshold –40 dB, comp 3.2, knee 4.7, attack 5 µs, release 204 ms], which flattens the levels (Track 33.203). The succeeding PSP Nitro was employed to create an autopan effect (Track 33.204).

On the audio group, the Mod Delay [23.75 ms delay of the right channel only, 44% depth and 0.74 Hz modulation] produces the Haas trick (Track 33.205). A low-shelving filter [1.7 kHz, –5.1 dB] on the FilterBank tunes the vocals into the mix (Track 33.206). Then the +Chebyshev adds some edge to what was a somewhat numb sound (Track 33.207).

Break BV

Inserts: Digidesign DigiRack EQ 3, Digidesign DigiRack Dyn 3.

Audio Group Inserts: PSP 84.

Three backing vocal tracks were recorded from the break onward, and these also conclude the production. Two of the tracks were performed by Dan and one by Lish. Lish’s voice was mixed to be the leading voice out of the three, panned a notch to the left; Dan’s vocals were each panned nearly fully to opposite extremes. The three voices were first equalized using the DigiRack EQ 3, then compressed using the Dyn 3, before being routed to a bus for additional processing. Perhaps the most interesting processing is the EQ on Lish’s voice, where an HPF [505.6 Hz, 12 dB/oct] rolls off an excess of lows (Tracks 33.208 and 33.209), a parametric filter [2.93 kHz, –4 dB, Q 1] softens some edge (Track 33.210), and a high-shelving filter [3.67 kHz, 7.4 dB] boosts a generous quality of air and definition (Track 33.211).

The effect applied on the audio group is a slightly modified version of the PSP 84’s “Gasherbrum Four” preset, set to 50% wet (Tracks 33.212 and 33.213). The group on which the PSP 84 was loaded is sent to the ambiance reverb (Track 33.214).