I began planning the production in early February and completed it by the end of the month. We made a few changes to the overall organisation of the recording from the time it was recorded until we arrived at the following configurations (see appendix):
The second session was dedicated to recording guitars, bass, vocals, and saxophone. Unfortunately, the saxophone did not arrive, so we had to change our plans slightly, excluding the drum booth entirely.
In addition, we scheduled a third session in which we essentially replicated the second session's setup. I personally utilised this session just to record background vocals for my song.
Production Drums “I consider it to be an abuse of a drummer to have them sit and check mics all day, and then expect them to be excited and energized when the light is on” (Massy, p. 101)
The drum session began early, which allowed us time to set up and tune the drums, with the assistance of a drummer. Jacquire King es that is “important to record the sound that you want, rather than keeping things safe and figuring out a way to make it work in the mix” (Tingen, 2008). In our case, the sound that we were looking for was impactful and aggressive. My references for the production of my song come from the album “The Hunting Party” by the band “Linking Park”. In particular the song “Guilty all the same” (see appendix page 34). I picked this style because I believe it complements the aesthetic of my track. While both Linkin Park and Steve Buscemi's songs include strong guitar and bass riffs, another thing that both bands have in common is the kind of singing that they use. Throughout Steve Buscemi, there are sections where the singer is rapping and other sections where he is screaming. Additionally, there is a version of the album that is comprised entirely of instrumentals and acapella, which was beneficial for my critical listening and comparisons. However, after I had my one-to-one session with my lecturer, I also took into consideration the band “Rage Against the Machine” (see appendix page 35), as a reference. “The Hunting Party” was engineered by the mixing engineer Andy Wallace. His style of mixing can be synthesised in two words: power and impact, as well as the creation and preservation of dynamics in the mix. "It's all part of a greater picture that I try to paint in the mix, of drama, dynamics and architecture” Wallace states (Tingen, 2014). The objective behind the drums recording was to avoid a broad stereo image while yet making them sound as impactful as possible. For this reason, we choose a Mid Side Configuration for the Overheads, which is excellent for mono compatibility and control of the stereo field. For the Mid microphone, the choice of the polar pattern should be dictated by the width of the stereo recording angle (Houghton, 2020). We opted for a cardioid Mid, which is equivalent to an XY pair of super-cardioid microphones crossed at an angle of 127 degrees (Houghton, 2020). We measured the distances between microphones; our primary objective was to create a two-mic overhead setup that met these two requirements: Each mic receives the snare sound at the same volume and at the same time (www.admiralbumblebee.com, 2018). Both Kick and Snare were recorded with a different number of microphones to provide a greater variety of colours (see appendix). We place a sub in front of the kick and an RE20 inside. Then, in order to obtain an organic sound of the kick (Massy, p. 111), we placed a D112 mic next to where the beater hits the kick drum head, and drum batter side. This is especially good for capturing drum performances in which a dynamic feel is important (Massy, p. 112). For the Snare top, we use a dynamic and a condenser microphone (see appendix), in a cardioid pattern, to avoid as much spillage as possible. In fact, directional mics, such as cardioids, exhibit different frequency responses off-axis, off-axis sounds tend to be duller.
(Massy, p. 111 )
After placing the microphones, we began attentively listening and moving them slightly, altering their position in little increments.
It was like “organising a photographic set, where the photographer keeps making small adjustments to the lighting and the way that the subject is posing until, at a certain point, he clicks the shutter and says, ‘That’s the photo I want’” (Buskin, 2014).
Then we check the phase because “the secret of recording great drums is the phasing of the microphones with each other ” (Massy, p. 108 ).
Massy, p. 99, states that “If your drummer’s performance is not good, it is very difficult to fix afterwards”. With this in mind, We gave some time to the band for rehearsing and we recorded an average of 6-7 takes for each song, making sure that the drummer’s performance was acceptable, and the band was feeling more confident. For the Room sounds: A hoe (see appendix page 21) was set up around the kick going into an SM57, and the result is the sound of a drum room without the “ irritating cymbals that can ruin your room sound” (Massy, 2017). In his article on sound on sound, Rogers, 2015, describes several variations to recording the room sound, one of them is called “True Grit”. This involves positioning a Beyerdynamic M160 ribbon mic above the kick drum. On the other hand, We placed it on the back of the drummer. Guitar, Bass and Vocals. “By blending mics checking phase and moving mics when needed it's easy to record a great guitar sound without ever engaging EQ” (Massy, p. 132) The topic of how far away microphones should be placed from amplifiers is very controversial. While Chuck Ainlay's 'just off the grille' seems to represent the majority opinion, Steve Albini often begins from a distance of roughly ten to twelve inches. On the other side, Alan Parsons avoids near placements (Senior, 2019). Parsons, 2019 states: "On a guitar amp, I usually use condenser microphones; dynamics are just too telephony. That is an unjust assessment of a really excellent dynamic microphone, but in comparison to a condenser, you will not receive the low end." Producers often seem to be aiming for a wider frequency range and a more diffuse sound produced by the large diaphragm. Additionally, these mics enhance the 5-15kHz frequency range, however, this enhancement soon fades as you go off-axis. On the other hand, small-diaphragm condenser microphones often have flatter frequency curves and a more regulated off-axis response, resulting in a more focused sound (Senior, 2019). For our session, we prefer to conceive in terms of combining 'bright' and 'dark' mics. This was the idea I had in mind while I blended the sounds of the various microphones. The SM57 and MD421 added body to the overall sound (see appendix, page 34). Barresi, for example, relies heavily on the SM57 and MD421 combination “often both close up, placed at the edge of the speaker, where the speaker centre meets the cone” (Senior, 2019). The SM57's frequency response suits guitar recording (see appendix, page 34). For a start, the sub-200Hz response roll-off reduces low-end cabinet 'thumps', which might otherwise conflict with the kick drum and bass in the mix. However, there's also a slight attenuation at 300-500Hz, an area where muddiness can easily occur, and a broad 2-12kHz
presence peak, which adds bite and helps the guitars cut through the rest of the track (Senior, 2019).
Sennheiser's cardioid MD421 has a wider frequency response, none of the low mid-range attenuation, and an even heftier sensitivity boost upwards of 1kHz. This microphone also has a larger diaphragm than the SM57, and the off-axis response anomalies of the larger diaphragm, in particular, give a different character to the sound (Senior, 2019).
On the other end the KSM 32 (as we can notice in Figure 28) and the DI, brought more mid/upper-mid prominence to the mix.
Steve Albini (2009), usually set up close to cabinet two microphones: one dark and one light, such as a ribbon and a condenser, or two separate condensers with distinct personalities. In our recording, the Beyerdynamic 160, which was placed in front of the cabinet, added a “darker” tone to the blend (Senior, 2019).
The same principle has been applied to the choice of bass microphones. While the Audix D6 was covering more the lower part of the spectrum, the RE20 represented the mid information of the bass, and the Shure Beta 52 has a boost on the high mid (see appendix, pages 35,36).
After careful consideration, I chose to remove the MD421 and the Shure Beta 58 from the bass and guitar blends since they were not contributing so much to the mix.
For the Vocal, I wanted something comparable to the Neuman U67, and our lecturer advised that the Soundelux U195 was the closest match. The Soundelux is a well-balanced and extremely smooth vocal mic in Fat voice mode, while Normal mode slightly tightens the lows and adds a little sensation of air to the high end (White,2017).
Post Production Editing: The band requested that the recording needed to be done without a click because there is a significant shift from a slower BPM to a quicker BPM at the finish of the song.
I started with editing the drums, following this workflow: Phase alignments Beat Identify Beat detective Then I phase-aligned all the other instruments, and I used elastic audio for the bass and the vocals. Because of the style of the singing, I applied Melodyne just to 4 bars at minute 0.51s. I did a comp of all the vocals and create 3 different busses for the main vocal, the backing vocals, and the screams. I also have to use a plugin called “Clarity” to remove some noise floor from the Urei 1178’s track. Because the sound was too dull, I added a sample reinforcement for the snare. Andy Wallace is another example. As in Linkin Park's record, he usually replaces drums with his own samples (Tingen, 2014b). I concentrated a great deal of my attention on the kick; I wanted the kick to sound punchy and big like Andy Wallance's, so I used a clipper and transient shaper. Around minute 2.03 of “guilty all the same”, Andy Wallance introduced a low pass filter on the solo guitar, similarly, I added a low pass filter but on the drums at 0.53 seconds. Mixing : "That first stage, when the mix is dry and I'm going for punch and clarity, is done while monitoring at higher volumes, generally speaking on Genelec nearfield monitors. I listen loud to get a sense of the general impact feeling, and to work with the low end”. (Andy Wallace) I divided the mixing procedure into four distinct stages: digital, analogue, digital, and analogue: 1. ITB (In the box) I use Pro Tools’s stock plugin for Gating the Kicks, Snares, and Toms. For the subtractive eq, I used the digital EQs, such as the ProTools stock’s plugin EQ7 and the ProQ3, because are cleaner for this type of operation (Addeo, n.d.). The digital Eqs were in linear phase mode in order to not alter the phase relationship of the source (Silverstein, 2018). For the Vocals, I use a few plugins: Vocal Rider, which is an automated clip-gain controller; the SSL Vocal strip Plugin, but just using the functions of de-essing and anti-pop-up; then I employed, a plugin called, de-breath. This procedure assisted me in balancing the voice levels and removing undesirable sounds from the recording. 8
2. ANALOGUE Specific to 'Guilty All The Same', Andy Wallace, used a Distressor on the vocals, but also the SSL compressor. The reverbs he used were the Lexicon 480 for a longer reverb, the PCM70 for smaller rooms, and the PCM42 for delay lines. All these effects would be sent on the board (Tingen, 2014b). 24/03/2022 From 9.30 till 15.30 SSL G+ Console Organization of the console and patching for the SSL G+ 4000: Group Output 1 into GoldMike, and return into the console on track 1. I used the Gold Mike as a “parallel saturator” for the vocals, making use of the High Flair function which “uses the pleasant sound characteristics of the coils in combination with the tube to emphasize the presence range. It is ideally suited to intensify the overtone range of voices and acoustic instruments, thus improving the ability to cut through in the mix” (spl.audio, n.d.). Below is a comparison of the voice signal before and after the SPL:
(Laurito, 2022)
(Laurito, 2022)
It's interesting to see how, in the second image, the SPL, especially around 4 kHz, where speech is intelligible, enhances the signal's harmonic content.
I insert the Neve 8801 channel strip in the vocal, using his compressor and EQ.
For the bass and guitar, I mixed the microphones I used for the recording (see chapter 1) to various levels, using the different characteristics of the microphone in order to cover all of the instruments' frequency ranges.
I created a blend of the bass tracks in Protools and output them into track 7, then insert the compressor Neve 2254. This conferred a rounded sound to the bass, (Abbey Road Institute Amsterdam, 2016). In regard to this compressor Robjohns, 2009, states that “the sound is pleasingly smooth and full but not obviously coloured, and the dynamic control is sublime and almost creamy”.
Same for the guitars, then I inserted the Neve 1073 preamp, Pre EQ and use the EQ onboard.
From 15.30 till 10.30 Neve VR Console. Organization of the console and patching for the Neve VR :
(Laurito, 2022) After I sent all the drum’s tracks to the consoles, I grouped them into channels 23-24, and, after pressing CH.OP, I send them, through the small fader, to the matrix and so to the group1, which I previously assigned on the patch bay, to the Urei 1178, then return into channels 1-2. The Urei 1178 is FET Compressor and is one of the most famous, widely used compressors of all time (Messitte, 2018). FET stands for “Field Effect Transistor” (Poole, 2015). Because parallel compression can help shape transients while still maintaining a consistent output level, this technique is most often used for helping “glue” drum kits together (Kody, 2022). Senior (2009), states that despite the reputation the Urei's 1176 and 1178 have for a fast attack, Lord‐Alge was actually using an 1178 more for its sonic flavour with slower attack settings. Rich Costey's preference for a parallel processing setup with this compressor also suggests that the tonality of the Urei, especially when driven hard, was of more importance than its ability to catch transients (Senior, 2009). The Drum Bus was sent into aux 1-2, patched into the Lexicon 480, which then was returning into tracks 25-26. 11
Effects: “I always imagine the listener being in the same space as the performers. If things are very dry and clean, it can sound unnatural and lightweight”. (Andy Wallace)
(Laurito, 2022) I added some room reverb to the vocals, in order to give colour and dimension, and the Yamaha Spx as delay to add some tridimensionality. Finally, I put the Bricasti M7, at the end of the effect chain, in order to blend all the effects and obtain more space coherence. The sound of the Bricasti is it is extremely dense and detailed, adding air around a sound, giving it a sense of place without making it sound splashy or gritty (White, 2008).
(Laurito, 2022)
For the drum reverb, I employed the Lexicon 480L. I started from bank 2, prog 6 "Large Wood Room”, and then change different parameters like shape and size.
The key to the Lexicon sound is to be found in the algorithms they use, and the powerful hardware needed to run them. This algorithm incorporates Lexicon's Spin function, which modulates certain reverb parameters to prevent modal resonances predominating at certain frequencies during long decay times (White, 1996).
(Laurito, 2022) On the guitar Solo, I applied a reverb flange.
(Laurito, 2022)
I also used a few more plug-ins for the effects, mainly on the vocals. The “Valhalla Vintage reverb” and the “H-Delay”.
I returned all the effects to the console, in order to process them with the onboard EQ. However, I also used the filters, located in the aux section on the SSL Console, in order to tame low and high frequencies.
(Laurito, 2022)
After Patching everything, I started with careful listening, tweaking the EQs and compressors on board and AB the signals to check if the sound was improving or not. I did all this always in the contest of my reference track (which was occupy the last two channels of the consoles). After this process, then I printed everything in Pro Tools.
3. ITB
After I had all my analogue processed tracks, then I made the last corrections and all the automation. In this stage, I made use of different transient shapers plugins, on the drums in order to give more punch and eventually let the drums cut through the mix. I used a transient shaper also for the bass, this allowed me full control over the envelope of the bass (Kody, 2022).
4. ANALOGUE
“ I am very concerned with preserving the dynamics of each mix, as it is part of a greater picture that I try to paint. Riding the master fader during the mix is part of my emphasis on the sonic architecture of the mix and of maximising the feel that I get from it.”
(Andy Wallace)
The last stage was to come back into the SSL G+ and print the Mix of the Stems, going through the SSL’s Master Buss Compressor.
Mastering:
Perkins (2018), suggests that In most cases, following these 2 steps will provide a mix that is ideal for mastering:
1. Don’tallowpeaklevelstoreach0dBFS
2. Don’tuseapeak-limiterorothersimilarpluginsthatpreventpeaksthatwouldotherwise
reach or clip 0 dBFS by creating a hard ceiling at or below 0 dBFS.
On top of that, the ideal headroom for the mastering engineer should be 6dB.
The main issue that I found in this process, was to achieve a correct gain staging in order to obtain -6dBSPL on the Master Bus.
Before the final print, I checked the crest factor.
By comparing the difference between peak and average levels, crest factor measurements give us a very good way to judge the balance between transient and sustained elements in a mix and provide an indication of how to approach compression and limiting treatment (Stewart, 2020).
Higher low-end crest factor readings can suggest the balance between low-end percussions, such as a kick drum, and sustained elements like basses or synths may be biased toward the percussive transient content. Conversely, a low crest factor reading points to sustained elements that are dominating the percussion. In my specific case, the main issues were coming from the area around 6/8 kHz.
After my one-to-one session with my lecture I understand that I could address this by adjusting balances in the mix, but also through a few ways to address it in the master:
Compression: Frequency-specific compression with a fast attack and release can reduce the crest factor; targeted dynamic EQ: Cuts can reduce the crest factor (Stewart, 2020).
The picture below shows an improved distribution of the frequencies across, the High-Mid.
(Laurito,2022)
The meters we see in our digital mixer show the values in dBFS, and the maximum level possible in a digital system is 0dBFS. However, before this signal can be played out on speakers or headphones, it must be reconverted from digital to analogue, which may produce inter-sample clipping (McAllister, 2021). For this reason, the suggested target for the True Peak is a maximum of -1 dB.
However, during the first mastering session I did, there was a true peak of -0.9 at a specific point in the song that was caused by an overhead transient. I returned to the mixing session and lowered the gain of that particular section by 2 dB.
The second target to achieve was -14 LUFS with 2 dB of tolerance. This was considerably simpler to do, given the song's loudness level was already floating around that value.
I had planned to master the track in the master suite, but when I played it via the outboard gear, I noticed a white noise coming from the right speakers. I opted to complete the mastering ITB at this point since I already had Ozone 9 installed on my laptop.
This was my mastering chain:
● EQ, in Mid Side, I attenuate on the sides everything below 200 Hz
● A dynamic EQ, to attenuate harsh frequency around 6-8kHz, Is this the area that was more problematic, I also compress this range.
● Some warm saturation, especially from 200 Hz above, and The Ozone Imager plugin, give more stereo spread in the mid-high range.
● Then I used a parallel compressor to add colouration. Following mastering, my LUFS and True Peak levels looked as follows in the image below:
(Laurito, 2022)
Consideration on Compression As Kody (2022) states, learning where to draw the line for each compressor takes a decent amount of trial, error, and experimentation. By splitting compression targets out over a couple of different compressors, is possible to let them operate in a way that better handles a large workload. In “Steve Buscemi” the dynamic range of the vocal is very wide. So initially We used the compressor on board the SSL, in a very subtle way, just 1 dB of reduction, in order to control a bit the enormous fluctuations caused by the rap/screaming performance of the singer.
After that, I used the compressor present in the Neve 8801, again 1 dB of reduction and after the R-vox compressor plugin. As a result of that, I could achieve more intelligibility, and the voice was more upfront in the mix (recording revolution.com, 2016).
Appendix
First session 10/03/2022
This was the set-up we had planned for the first session, where we were going to record the drums:
Console: Neve VR
INPUT1: SUBKICK_Infront of the kick
INPUT2: RE20_ inside or side of the kick
INPUT3: D112_ inside or side of the kick (we will try which sounds best)
INPUT4: SM57_ snare top_ pointing to the edge
INPUT5: AKG C4000 B (on cardioid) second SNARE TOP, pointing to the centre:
In order to maintain phase coherence between SM57 and the C4000, we gonna use the 3:1 ratio for the placements.
INPUT6: C414 snare bottom ( eventually with pad -16 dB)
INPUT7: MD421_ tom (If the mic point at the centre of the tom there is more low end, more is pointing to the edge, more overtones)
INPUT8: AUDIX D6 or MD421_ floor tom
INPUT9: AKG 414 sideL_ OH (Fo8)
INPUT10: AKG414 sideR_ OH (Fo8)
INPUT11: AKG 451 B Mid_ OH (cardioid)
INPUT 12: BeyerDynamic M160_ room mic back
INPUT13: SE Electronics Z5600_ (on OMNI) room mic front
INPUT16: DI GUITAR
INPUT 17: DI BASS
Press the “ground” button on the boxes
INPUT 18: TALKBACK MIC
Second Session 12/03/2022 Console: SSL G+ 4000
DRUMBOOTH
INPUT1: MD421(SAX) INPUT2: RE20 (SAX) INPUT3: SPIRIT (room)
LIVE ROOM:
GUITAR
INPUT1:
INPUT2:
INPUT3:
INPUT4:
BASS
INPUT5:
INPUT6:
INPUT7:
INPUT8:
ROOM MICS (this will change in the relation to the amps set up) INPUT7: SE Electronics Z5600
INPUT8: TLM ROOM L
INPUT9: TLM ROOM
Vocals:
INPUT10: Soundelux U195 (vocals) DIs:
INPUT11: AKGC414 (B vocals) INPUT12 and 13: DIs
SM57 MD421 KSM 32
BeyerDynamic M160
AUDIX D6 BETA 58 BETA 52 RE20
