Historical development, sound aesthetics and production techniques of
metal’s distorted electric guitar
The sound of the distorted electric guitar is particularly important for many metal genres. It contributes
to the music’s perception of heaviness, serves as a distinguishing marker, and is crucial for the power
of productions. This article aims to extend the research on the distorted metal guitar and on metal music
production by combining both fields of interest. By the means of isolated guitar tracks of original metal
recordings, 10 tracks in each of the last five decades served as sample for a historical analysis of metal
guitar aesthetics including the aspects tuning, loudness, layering and spectral composition. Building
upon this insight, an experimental analysis of 287 guitar recordings explored the effectiveness and effect
of metal guitar production techniques. The article attempts to provide an empirical ground of the acous-
tics of metal guitar production in order to extend the still rare practice-based research and metal-ori-
ented production manuals.
Keywords: guitar, distortion, heaviness, production, history, aesthetics
With the exception of genres like black metal that explicitly value low-fidelity aesthetics (Ha-
gen 2011; Reyes 2013), the powerful effect of many metal genres is based on a high production
quality. For achieving the desired heaviness, the sound of the distorted electric guitar is partic-
ularly relevant (Mynett 2013). Although the guitar’s relevance as a sonic icon and its function
as a distinguishing marker of metal’s genres have not changed in metal history (Walser 1993;
Weinstein 2000; Berger and Fales 2005), the specific sound aesthetics of the guitar have varied
substantially. Due to that special importance, many metal bands invest a lot of energy in creat-
ing the optimal guitar sound. For example, after having recorded the guitar performances for
their album Threshold (2006), the Swedish power metal band HammerFall stated to have re-
amped their guitars for three weeks to be completely satisfied (Schwarz 2006). Such an enor-
mous effort is not a unique phenomenon of modern metal productions but emerged quickly
within a short period of time in early metal. Whereas Black Sabbath recorded their self-titled
album (1970) in just a day (Iommi 2012: 62ff), their fourth album Sabbath Bloody Sabbath
(1973) already was the result of laborious recording and production experimentation (Iommi
2012: 133ff). Following this trend, the production strategies and ways of guitar playing have
changed in metal history to improve the desired “wall of sound” (Mynett 2009a, 2009b).
Research on guitar distortion and metal production is still in its infancy. The rare literature
on the distorted electric guitar has paid particular attention to cultural identity, ethnicity and
political subversion (Waksman 1999), genre definition (Gracyk 1996), gender (Walser 1993;
Bourdage 2010), and communication (Herbst 2014). Within musicology, Walser (1993, pp.
41ff) devoted three pages to distortion while discussing the principles of heavy metal. By spec-
tral analysis of original records and experimental guitar recordings Einbrodt (1997) identified
acoustic elements having contributed to the rock guitar sound. Some work explored the percep-
tion of distorted guitar chords by theoretical and psychoacoustic analyses (Lilja 2005, 2015;
Berger and Fales 2005) and psychological experiments (Juchniewicz and Silverman 2001). So
far, the study of Berger and Fales (2005) is the only one concerned with the historical develop-
ment of the metal guitar sound. With regard to metal production techniques Mynett (2013: 18f)
notes that “academic exploration into the processes of music production can […] be viewed as
being in an embryonic phase [… and that a] comprehensive study into procedural methodolo-
gies for the production of contemporary metal music is ‘virtually non-existent’”. He continues
to claim that publications providing an empirical methodology for pre-production, engineering
and mixing approaches are still rare (Mynett 2013: 29). In the same vein, Williams (2015) ob-
serves that research on metal production techniques was behind its creative uses.
This article aims to extend the research on the distorted metal guitar and on metal produc-
tion by combining both fields of interest. At first, it will explore the historical development of
the metal guitar sound in order to broaden the exploratory results of Berger and Fales (2005).
Then, it will determine how varying production techniques affect the metal guitar sound.
One possible way of analysing sounds is to describe the listening impression. While this method
provides an understanding about the music’s effect, computer-assisted tools improve objectiv-
ity and may offer deeper insights into the acoustic features behind the perceptual effects (Cook
and Clarke 2004). Since there is no universally accepted measure for timbres (Berger and Fales
2005: 182), the acoustic properties volume, pitch, spectral content and envelope must be inter-
preted. This way, specific timbres or “colours” in the realm of metaphorical perception and
verbal description (Zbikowski 2002) like guitarist Edward Van Halen’s “brown sound” can be
explained by acoustic features.
The main approach in this study relies on spectral content as the overriding aspect of timbre,
one that is manipulated in music production. Screenshots of spectra in digital audio work-
stations are limited because they only capture one moment of the envelope, and the spectral
content of different recordings cannot be compared directly. Instead, the recordings were eval-
uated with the frequency analysis function of Audacity (Hanning window, size 16,384), and the
data was imported to Microsoft Excel to undertake statistical analysis. For improved compara-
bility, the spectral content was divided into eight frequency groups while taking into account
the spectrum of the metal guitar. Descriptive statistics of these frequency groups allowed com-
paring recordings, evaluating effects and demonstrating historical developments. Spectro-
graphic analysis with the Sonic Visualiser provided a more detailed exploration of the audio
material. Volumes were measured with Adobe Audition 3.
The first part of the study aimed to gain an understanding of the metal guitar sound from a
historical perspective. In line with Berger and Fales (2005), only original recordings with iso-
lated guitar parts were taken into account to minimize interference. The sample is the result of
a historically guided Spotify search (e. g. Black Sabbath for 1970s metal) and using the “similar
artists” function. Ten tracks each were selected for five categories: 1970s proto metal, 1980s
classic heavy metal, 1990s thrash metal, 2000s extreme metal, and 2010s modern extreme metal
(see track list in the appendix). Most of the songs were released in the early years of the decades.
The experimental part intended to explore production techniques of the metal rhythm gui-
tar. Metal riffs combine single notes and power chords (a fifth interval with an optional octave).
Four versions of two bars of an E note (82 Hz) and an E power chord in eighth-notes plus a
sustained power chord were recorded directly into the software (Apple Logic Pro X) to enable
re-amping the same performance. Re-amping requires the guitar signal to be recorded directly
into the computer. The performance can thus be recorded with different amplifiers, amplifier
settings, effects pedals, cabinets and microphones as often as desired. A re-amping box (Palmer
Daccapo) is needed to adjust the impedance of the recorded guitar signal to the amplifier. To
explore the effect of gear, two distinct guitar models were used: A Gibson Les Paul and a Fender
Stratocaster. Both were played with humbucker pickup at the bridge. All five amplifiers (mod-
els and settings in the appendix) were based on tube technology. The signal ran into a closed
4x12 Marshall cabinet (1960BV model) with Celestion G12 Vintage 30 speakers and was cap-
tured with a Sennheiser MD421 dynamic microphone. Some recordings were made with an
additional boosting pedal (Fulltone OCD). All files were exported using the normalisation func-
tion. Differentiating between open and palm-muted sounds, ultimately 287 recordings were
created; 140 of a single guitar, 75 of two guitars, 40 of four guitars and 32 with different miking.
Metal guitar sounds
The defining sonic phenomenon of the metal guitar is distortion. A number of elements distort
the signal: String vibration, pickups, cables, effects pedals, preamplifier and power amplifier
tube distortion, speaker distortion, and microphone choice and placement (Berger and Fales
2005: 184). A reason little known for the electric guitar’s distinct sound is the inherent inhar-
monic spectrum of the strings that is created by its bending stiffness and winding (Zollner 2014:
222ff). Frequency content a few hertz apart from the fundamental note and its partials produces
periodic “pseudo noise” creating a rough impression. The pickups’ windings and amount of
coils alter the basic signal before entering the amplifier. Overdrive, distortion and fuzz pedals
may be added to increase the voltage gain or to produce transistor clipping. Tube distortion is
achieved when the signal is amplified beyond the capability of fidelity reproduction. The signal
is pushed against the power supply causing clipping and resulting in a modified waveform
(Doyle 1993: 56). The edges are rounded off by the power valves and output transformer, pre-
dominantly intensifying even harmonics that add warmth and presence (Doyle 1993: 57). Push-
ing the signal against a barrier enhances noise, adds harmonic and inharmonic overtones, and
produces a flatter dynamic envelope of a more complex waveform (Gracyk 1996: 111ff). The
timbre becomes noisier, rougher, more compressed and present. The speaker further alters the
sound as it functions as a mechanical filter reproducing frequency mainly between 75 and 5,000
Hz (Einbrodt 1997: 198). Speakers compress the sound since the stiffness of the cone prevents
reproducing the full dynamic range (Chappell 2010: 54). The speaker breakup at the edge of
the cone further contributes to distortion (Williams 2015: 49f). The cabinet also affects the
sound; open boxes are sounding ambient and closed ones are having more low-end (Chappell
The potential of the guitar as a main distinguishing element of metal genres grounds from its
perception of “heaviness”. As Berger (1999: 58) notes, the history of metal is characterised “as
a progressive quest for ever-heavier music”. Although such heaviness is subject to individual
perception, diverse understandings within music scenes and structural elements, the character-
istic guitar timbres associated with metal genres must be considered with special care. One
necessity for heaviness is the right frequency spectrum. Low frequencies provide the powerful
sonic weight, and sculpting it for retaining transparency and note definition is challenging
(Mynett 2012). High frequencies are of equal importance because they contribute to aggres-
siveness. Berger and Fales (2005) observed the energy cut-off of guitars to have risen from
around 5 kHz in the 1970s to 8.3 kHz in the 1990s, and they concluded high frequency energy
to correlate with perceived heaviness. Two further elements determining heaviness are distor-
tion (Berger and Fales 2005; Mynett 2013) and high volumes (Walser 1993: 45; Weinstein
1991: 23; Williams 2015) as a result of distortion’s compression effect. Distortion ”simulates
the conversion of the guitar from an impulsive to a sustained or driven instrument, and this
transformation may be part of the acoustic correlate to the perceptional experience of heavi-
ness“ (Berger and Fales 2005: 194). Structurally, the power chord contributes to heaviness be-
cause distortion intensifies difference tones one octave below the root. Hence, distortion ex-
tends the guitar signal in the highs and lows (Walser 1993: 43), which makes it the central factor
Producing metal guitars is subject to a set of genre-specific conventions. Dynamic microphones
are common for recording guitar sounds because they tolerate high sound pressure levels (Chap-
pell 2010: 56ff). The microphone’s distance to the speaker is important for a number of reasons.
Close miking increases the intensity of the bass frequencies due to the proximity effect (Hall
1980: 384f); hence, the right distance is necessary for a defined low-end (Mynett 2009a: 132f).
Furthermore, close miking minimises the perceived distance to the sound source contributing
to a direct sound (Bartlett and Bartlett 2009: 105). Adjusting the microphone to the speaker
cone is of equal relevance. Pointing at the centre of the cone intensifies the higher frequencies;
off-centred microphones are capturing more middle and bass frequencies (Bartlett and Bartlett
2009: 125). Another important element of metal productions is the number of guitar tracks.
Provided that the rhythm is accurate and synchronous, doubling results in a thicker sound due
to enhanced frequency coverage (Mynett 2012) and wider stereo image (Hamidovic 2015: 61).
Conventionally, the tracks are panned to the stereo channels to maximize the transparency and
size of the production (Mynett 2013: 51).
After recording the tracks, fine-tuning aims at balancing the frequencies with the other in-
struments to increase intelligibility (Izhaki 2013: 5) and to shape the guitar sound beyond “real”
performances. Equalising is necessary as distorted guitars cover a frequency range of more than
five octaves beginning at around 60 Hz (Izhaki 2013: 256). Mynett (2009b: 120ff) and
Hamidovic (2015: 57) suggest using a high-pass filter between 65, 105 or even 140 Hz to make
space for the bass instruments. With regard to particular frequency areas, the mid bass range
(60-120 Hz) is associated with power and boosting it may enhance the thumping sound of the
cabinet (Hamidovic 2015: 63). The upper bass (120-250 Hz) contains most fundamental notes
of the rhythm guitar in standard tuning and some resonance of the cabinet. It also contributes
to power. Containing the first partials, the lower mids (250-1,000 kHz) are decisive for the
instrument’s timbre but they may hinder transparency (Izhaki 2013: 256). According to pro-
ducer Colin Richardson the mid frequency area (1-2 kHz) is vital for the “in your face effect”
(Taylor 2011). The high mids (2-4 kHz) are responsible for volume, definition and presence
(Izhaki 2013: 266). The lower highs (4-6 kHz) are necessary for contributing definition and
intelligibility while also ensuring closeness and directness (Mynett 2012). The highs (6-10 kHz)
mainly contain high-order partials and noise that hold the risk of reducing intelligibility when
being attenuated (Hamidovic 2015: 63). The air area (10-20 kHz) mostly contains noise and
some edge (Izhaki 2013: 256).
Since distortion reduces dynamics, many metal producers like Richardson refrain from add-
ing compression to the mix (Martinelli 2008). Others like Andy Sneap utilise multiband com-
pressors to prevent the low-end build-up produced by palm-muting with a bass-heavy sound
(Hamidovic 2015: 59). Adding overdrive pedals is another trick to increase intelligibility
(Mynett 2009a: 132f). As Richardson explains, they “tighten the whole sound up. There’s a
certain place where you’ve got to wind the pre-amp up to get the gain, and putting the Tube
Screamer on allows you to use a little bit less gain. When you’re tuning in E, this isn’t so
necessary, but when you’re down to B, or C, or C sharp, it seems to put it all together“ (Marti-
Historical development of metal guitar sounds
When metal emerged around 1970, the standard guitar models Stratocaster, Les Paul and SG
were already available. More specialised models were introduced in the 1980s. Manufacturers
like Marshall were already in business, so that early metal guitarists had access to amplifiers
designed to produce distortion. During the 1970s, tonal control of the amplifiers was improved
by the extension of equalisation, adjustable pre-gain stages, and presence and resonance con-
trols (Herbst 2016: 36). The rack systems of the 1980s allowed combining different preampli-
fiers and power amplifiers. Until the mid-1990s, it was difficult to find amplifiers like the Pea-
vey 5150 that produced the desired heaviness within the spreading extreme metal genres (Mar-
tinelli 2008). In the last 20 years, the number of amplifiers designed for metal has vastly in-
In early stages of metal production, analogue multi-tracking with 16 to 24 tracks and suit-
able microphones existed. Unlike many other genres that value analogue equipment, metal has
flourished by the chances of digitalization, and metal producers like Sneap (Martinelli 2006)
and Richardson (Taylor 2011) have embraced the potential of technological innovations. Not
only was the clarity improved, the extended disk space allowed extensive layering of tracks,
and digital effects helped to manipulate the material (Williams 2015). Contrary to the move-
ment towards digitalization in metal production, most metal guitarists continued relying on tra-
ditional tube amplifiers (Herbst 2016: 78ff). Potential reasons are the musicians’ wish to main-
tain the traditional values of rock culture regarding authenticity (Moore 2002: 219) or the sonic
benefits of the analogue technology of guitar equipment (Doyle 1993: 56f).
Against the back-
drop of the developments of guitar technology, studio devices and techniques discussed, a com-
parative analysis of original recordings will now track changes of the metal guitar sound from
1970 to 2015.
Tuning down the guitar is highly effective to extend its low-end, darken the timbre and gain
power (Mynett 2013: 44). The fifty sample tracks reveal that down-tuning has been uncommon
until death metal spread in the 1990s. These findings are surprising since Black Sabbath already
experimented with this technique in 1971 to produce “a bigger, heavier sound” (Iommi 2012:
94). In the sample of 1970s proto metal, only Sabbath’s “Into the Void” (1971) has an alterna-
tive tuning. Differences in band tunings explain some variability in the 1980s classic heavy
metal sample. Six recordings are in standard tuning, three are tuned down a semitone and one
two semitones. Thrash metal in the 1990s merely differs from heavy metal as seven tracks are
in standard tuning and three are tuned down between one and three semitones. A significant
change occurred in the 2000s death metal sample. It contains tunings between three and five
semitones below standard, most of them around C or B. An indication for this trend being time-
related rather than genre-related is that most death metal (e. g. Bolt Thrower, Cannibal Corpse)
and grindcore (e. g. Napalm Death) bands of the 1990s tuned down between one and three
semitones. A rare exception was Carcass using B tunings. Consequently, the trend continued
with tunings of the 2010s modern extreme metal sample revolving between B and Ab. Tuning
down the instrument is lowering the fundamental frequency from 82 Hz (E) to 52 Hz (Ab),
which not only affects the guitar sound immensely, but also forces the bass guitar to move down
The loudness of vinyl recordings in the 1970s and early 1980s was limited because of the ana-
logue technology. Digital mastering since the mid-1980s enabled louder records with more
compression. As loudness is one of the defining characters of metal (Williams 2015: 57), the
increasing volume from 1970s proto metal to modern extreme metal is hardly surprising (Table
1). Similar to tuning, the major changes occurred in the 2000s when the options of digital re-
cording extended rapidly. The small volume differences between 1970s and 1990s recordings
may result from the remastering of 1970s and 1980s releases that are distributed nowadays. The
lower loudness of modern extreme metal (2010s) compared to 2000s death metal may indicate
that the “loudness war” in metal (Williams 2015) could be over.
Table 1: Average loudness (RMS) of isolated guitar tracks of original metal recordings
Average RMS and
Difference to proto
Layering of guitar tracks
The sample shows a clear trend regarding the number of rhythm guitar tracks in metal history.
Most 1970s proto metal recordings contain two guitar tracks that are panned to the sides of the
stereo channels. Some Deep Purple songs differ from that convention as one guitar is substituted
by a rock organ. Within all heavy metal tracks of the 1980s, two tracks are standard. Most
thrash metal recordings of the 1990s share this convention. According to listening impression,
Iron Maiden, Annihilator and Forbidden are exceptions; their productions are characterised by
a particularly heavy guitar sound at that time. It is also worth noting that some 1980s produc-
tions (Accept, Running Wild, Saxon) have thick layers of reverb on the guitar tracks that in-
crease the instruments’ density while negatively affecting its intelligibility. In 2000s death
metal and modern extreme metal four rhythm guitar tracks seem required for fulfilling the cri-
teria of heaviness. However, determining the actual number of tracks proved to be difficult;
some recordings might even contain more than the standard four guitars.
From the very beginning, metal artists have experimented with the number of guitar tracks
as an effect. Many of the sample recordings contain isolated guitar parts with a reduced number
of tracks on one of the stereo channels in intros and breaks. When the guitar enters the second
stereo channel, the impression of a “wall of sound” is emphasised.
Heaviness is defined mainly by spectral features (Berger and Fales 2005; Mynett 2012, 2013).
Table 2 and Figure 1 provide numerical and graphical overviews of the spectral characteristics
of the guitars of the genres. To increase the comparability, ratios between the frequency groups
will be discussed in particular because they allow comparing spectral compositions. Absolute
values have been normalized for better comparability. The upper bass (120-250 Hz) was set as
reference value because it contains the fundamental notes in standard tuning.
Comparing the spectral development reveals clear trends. The mid bass (60-120 Hz) has
increased in relation to the upper bass. In compliance with the listening impression, the major
expansion of the low-end occurred from 1980s heavy metal to 1990s thrash metal, and even
more from 2000s death metal to modern extreme metal. The lower mids (250-1,000 Hz) and
mids (1-2 kHz) are much more relevant in proto and heavy metal than in genres that are more
recent. The trend to scoop out the mids for increased heaviness might be the reason (Mynett
2013: 45f). This aesthetics started in the 1980s when in heavy metal the intensity of the lower
mids declined, both in relative ratios and in absolute values. However, some early recordings
like Black Sabbath’s “Into The Void” (1971) or Saxon’s “Never Surrender” (1981) already
have a scooped sound similar to later productions (Figure 2). The percentages of the higher
mids (2-4 kHz) and lower highs (4-6 kHz) are similar in all genres except for 1980s heavy
metal, which seems to focus on these frequency ranges. Comparing proto and heavy metal in-
dicates that the first changes in guitar aesthetics were not an emphasis on the low-end but a
slight mid-scoop as well as an extension of the higher mids and lower highs. This complies with
Berger and Fales (2005) observation of an increasing heaviness through intensified treble fre-
quencies. The highs (6-10 kHz) remained nearly unchanged while the importance of air (10-20
kHz) has significantly decreased in the last fifty years. The high standard deviations, however,
indicate different aesthetics within the genres’ high to air area, especially in productions of the
2000s and 2010s. That limitation in mind, a general guideline for a contemporary metal guitar
sound is to focus on the mid and upper bass by combining a low tuning, a powerful amplifier
and a big to oversized speaker cabinet, and to shape the frequency curve in a progressive de-
cline. Presence is not as prominent as in heavy or thrash metal for not to brighten up the sound
too much, which would lessen the bass-heavy impression. Nonetheless, the primary partials
(energy cut-off 1, see below) are loud due to high distortion levels. The sound is less scooped
but generally shifted downwards.
With regard to the consistency of idiomatic guitar sounds within the subgenres, the standard
deviations of the frequency groups are conclusive. The high values within 1970s proto metal
demonstrate that the sounds differ between the recordings and bands. 1980s heavy metal seems
to be more formulaic than 1990s thrash metal. Due to the high standard deviations within some
frequency groups, there potentially is overlapping between the thrash, death and modern ex-
treme metal sounds. Since the 2000s, productions vary greatly in their intensity in the highs and
air areas. Consequently, the difference between the genres compared to 70s proto metal reveals
the highest standard deviation in the air, followed by the upper and mid bass due to the great
differences between 1970s / 1980s and more contemporary guitar sounds.
Table 2: Development of metal guitar sounds from 1970 to 2015
Figure 1: Average frequency curves of the genres.
A central criterion of heaviness in Berger and Fales (2005) exploratory study was the energy
cut-off. Most recordings of the present study have two cut-off areas regardless of genre belong-
ing. The first cut-off marks the end of the fretted notes and primary partials, the second marks
the fading of the quieter upper partials. Some recordings only have one clear cut-off, which
could result from excessive limiting. Overall, the sample does not support Berger and Fales
(2005) finding that in early metal productions the guitars’ frequency ends abruptly, contrary to
a gradual fading in later releases. In 1970s proto metal, the main frequency range stops around
4.66 kHz, which is slightly lower than the reproduction capability of most guitar speakers. Over
the fifty years, the main frequency spectrum has extended by almost 40% to 6.45 Hz, although
with an increasing variance. The second cut-off frequency has increased even more, however,
not linear. The biggest extension was from 1970s proto to 1980s heavy metal by 60%, from
8.20 to 13.31 kHz. Heavy, thrash and death metal hardly vary; modern extreme metal shows a
slight decrease in high-energy content.
Figure 2: Scooped sound in Black Sabbath’s (1971) “Into The Void” (left) and Deeds of Flesh’s (2003)
“Disintered Archaic Heap” (right); 2048 window; 21-9,000 Hz.
70s proto metal 80s classic heavy metal 1990s thrash metal
2000s death metal 2010s modern extreme metal
Experimental exploration of production techniques
The historical analysis has demonstrated major changes in metal guitar sound aesthetics. Many
of today’s metal productions require highly professionalised routines to keep developing the
quality and heaviness of the guitar sound. Engineers and producers can draw upon production
manuals (Owsinski 2006; Izhaki 2013) that provide general guidelines. Beyond printed litera-
ture, advice is available in online music production forums. Nonetheless, hardly any profes-
sional literature on metal production exists (Mynett 2013). Rather than trying to present formu-
las, the experimental exploration will investigate the effectiveness and effect of various pro-
duction techniques as an orientation for creating the desired sounds. The focus is on the devia-
tion that can be achieved (Table 3) along with qualitative changes in the spectrum.
Table 3: Standard deviation of gear choice, playing techniques and miking on frequency
Guitars (n = 20)
Amplifiers (n = 20)
Playing technique (n = 10)
Equaliser (n = 20)
Overdrive pedal (n = 20)
Distortion level (n = 80)
Microphone direction (n = 12)
Microphone distance (n = 20)
The small standard deviation of 0.19 indicates that different guitars are the least effective vari-
able for increasing frequency coverage. The largest variance between the Stratocaster and the
Les Paul is in the bass and treble registers. Without the averaging effect of the frequency groups,
Figure 3 displays that both guitar models do not fundamentally differ. In compliance with the
experimental studies of Einbrodt (1997) and Herbst (2016), the distorted Les Paul produces
more overtones, which results in a greater intensity above 2 kHz. More distortion also leads to
an assimilation of both models (Einbrodt 1997: 141). Yet, the guitars react differently to dis-
tortion level (SD = 0.70), with greatest effect in the middle and upper bass, lower highs and
Interpreting the small effect of different guitars needs considering a number of aspects. To
begin with, the attack phase is not represented adequately, which is crucial for the distinct per-
cussiveness of the Stratocaster compared to the mellow sound of the Les Paul (Jauk 2007: 282f).
Moreover, the Les Paul is 4 to 5 dB louder in the bass and treble registers while the Stratocaster
is more dominant in the 200 to 600 Hz area. Since 6 dB are perceived as twice as loud (Hall
1980: 114), these differences do result in distinguishable sounds that contribute to dimension-
ality of the sonic wall of guitar. It also must be noted that due to methodical reasons the effec-
tiveness of the other variables (Table 3) is greater than the guitar because they produce changes
with a greater bandwidth. The instruments’ formants are much subtler, and differences in the
first six to eight partials affect the guitar sound considerably (Roederer 1973: 128).
Figure 3: Frequency response of a Stratocaster and a Les Paul guitar with a medium distorted sound.
With a standard deviation of 1.78, combining different amplifiers is an effective way to shape
the sound of the guitar and to enhance frequency density. The results reveal amplifiers to differ
increasingly in higher frequencies. Furthermore, the devices vary in the mid bass area, which
is important for the low-end of contemporary metal sounds. This variance yet is mainly a result
of the Peavey 5150 model with its bass-laden sound. The upper bass hardly varies. As the ratios
show, the lower mids differ considerably, which may be explained by the frequency reduction
between 200 and 500 Hz as a characteristic feature of amplifier models and manufacturers
(Lemme 1995: 34).
Table 4: Spectral features of amplifiers
Although open and palm-muted playing depends highly on the composition, with a standard
deviation of 3.50 playing technique is very effective for controlling the sound (Table 5). As
with most variables, the upper bass is affected least, but palm-muting can effectively improve
82 829 1,576 2,323 3,070 3,817 4,564 5,312 6,059 6,806 7,553 8,300 9,047 9,794
Stratocaster Les Paul
the low-end bass. In combination with the reduced middle and treble frequencies, a powerful
sound can be achieved.
Table 5: Spectral features of playing technique
Equaliser and overdrive pedal
Equalising and adding an overdrive pedal to the signal are common techniques to shape the
guitar sound. Both effects were compared to the original recording. The EQ was set to slightly
scoop the sound by boosting 100 Hz (+5 dB; Q 0.60), cutting the mids around 850 Hz (−4 dB;
Q 0.30), and intensifying the highs around 4 kHz (+3.5 dB; Q 0.64). The overdrive pedal had
volume on 6, drive on 3, tone on 5.5, and high pass selected. The equaliser had a medium overall
effect (SD = 1.11) and altered ratios by ca. 4-5% (Table 6). By comparison, the overdrive pedal
had a large effect (SD = 2.57) altering the ratios by 7-8%. By decreasing the low-end and con-
siderably boosting the middle and higher frequencies, the pedal improved note definition and
Table 6: Effect of equaliser and overdrive pedal compared to original sound
SD original and EQ
SD original and
Setting the optimal distortion level is crucial for heaviness and transparency. With a standard
deviation of 2.50, the gain also effectively modifies the sound. The results indicate that increas-
ing distortion levels in the same amplifier channel tends to scoop the mids (Table 7). Creating
heavy distortion with additional pedals results in a slightly different, less scooped sound that is
rather focused on the middle frequency groups with less low-end and more treble.
Tab 7: Ratios of different distorted sounds
Heavy distortion amp
Heavy distortion pedal
Similar to the guitars, the distortion level changes the variance between amplifier models (Table
8). More distortion reduces the distinctive features of amplifiers in many frequency groups like
the mid bass, higher mids, lower highs and highs. More deviation is only possible in the lower
mids where most amplifiers have their distinctive mid-cut. The amplifiers also seem to be af-
fected differently by distortion; this may explain why some models are better suitable for much
distorted sounds. Moreover, the models differ mainly in the low-end and higher frequencies
when boosted by a pedal.
Tab 8: Deviation between amplifiers with regard to distortion
Heavy distortion amp
Heavy distortion pedal
Although easy to neglect, miking can hugely affect the guitar sound. Using various microphone
distances shapes the sound considerably (Table 9). Moving the microphone away from the
speaker reduces the low-end linearly. Regarding the proximity effect, the ratios show that an
increase of bass frequencies up to 15 dB, as claimed in studio literature (Albrecht 2010: 23), is
likely to be unrealistic. With more distance, the frequency curve increases from the lower mids
upwards. Ambience is better captured (Bartlett and Bartlett 2009: 105); it makes the sound less
artificial but reduces the directness required for heavy guitars (Mynett 2012).
Tab 9: Effect of microphone distance
The microphone position is more common in studio literature (Albrecht 2010; Bartlett and Bart-
lett 2009). Altering the position affects frequencies beginning with the lower mids while leaving
the low-end unaltered (Table 10). The overall effect of positioning (SD = 1.40) is similar to the
distance (SD = 1.21) but affects the frequency groups differently. Hence, a mixture of both
improves control over the intended sound.
Tab 10: Effect of microphone position
Between Centre and Cone
The historical analysis has shown the increasing number of guitar tracks as a central element of
contemporary metal guitar sounds. Methodically, it is difficult to capture the layering effect
with analytic tools (Herbst 2016: 228ff). Varying guitars, amplifiers and sounds only led to
small average standard deviations for doubling (SD = 0.21) and quad-tracking (SD = 0.19).
Even the detailed spectrogram (Figure 4) fails to capture all distinctive features of the wall of
sound. For both the open and palm-muted recordings it demonstrates less defined partials in the
area between 500 and 5,000 Hz in the quad-tracked recording. With an increasing number of
tracks, there may be periodic noise accompanying the partials (Zollner 2014: 222ff). This sug-
gests that reduced definition and extended ambience contribute to heaviness. Regardless of
playing technique and in line with the exemplary spectrograms, the differences in the quad-
tracked recordings lie in the increased mid bass, high mids, lower highs and highs. For power
chords, the effects are even larger but more difficult to spot in a spectrogram (no Figure). These
findings demonstrate a tendency of more tracks leading to a scooped sound, which complies
with the scooping tendency of higher distortion levels observed before. In addition, the fretted
notes of the two-track recording, without taking into account the number of tracks, appear more
dynamic so that the greater compression of more tracks may be a contributing factor of heavi-
ness as concluded by Berger and Fales (2005: 194).
Figure 4: Comparison of double and quad tracking. Top: two tracks Les Paul with Laney amplifier; bottom: two
tracks Les Paul with Laney and two tracks Stratocaster with Marshall amplifier; left: eighth-notes open E2 string;
right: eighth-notes palm-muted E2 string. 1024 window; 43-10,700 Hz.
The spectrographic inspection could not detect rhythmic irregularities of double and quad
tracked recordings. Nonetheless, these unavoidable small variations must be considered a cru-
cial element of the heaviness of layered guitars. Whereas combining different kinds of equip-
ment results in greater frequency coverage making the signal larger in the vertical dimension,
slightly varying rhythmic phrasing extends the duration of the phenomenon in the horizontal
dimension. If the performances are similar enough not to be perceived as asynchronous tracks,
the sound is felt as a single entity (Albrecht 2010: 47) with more heaviness and less definition.
The tempo, playing technique, structure and stereo separation determine when asynchronous
playing becomes apparent.
It is a matter of personal preference whether the sounds on each stereo channel should be
different. Similar sounds create a balanced impression that is beneficial for the wall of sound.
Different sounds may not provide this effect to the same degree, yet they add a vertical dimen-
sion to the stereo width. For example, separating a Stratocaster and a Les Paul guitar, or a bass-
heavy and a presence-rich amplifier, to the stereo channels creates a diagonal spatial impres-
sion. This technique produces a real stereo sound rather than what Owsinski (2006: 21f) de-
scribes as “big mono”, a sound with pseudo-stereo sources panned to the sides.
Synopsis and conclusion
Creating a heavy guitar sound is vital for the powerful effect of many metal genres. This study
confirms the still ongoing quest for ever-heavier guitar sounds (Berger and Fales 2005). Inno-
vations in guitar equipment and studio technology as well as structural changes have contrib-
uted to different guitar aesthetics in fifty years of genre history. Each genre and decade has an
individual sound, yet there are some general trends. Proto metal of the 1970s stood for experi-
mentation, and the sounds can be ascribed to a band rather than to a genre. Classic heavy metal
of the 1980s already had a distinctive metal sound but yet different from most of the later re-
cordings of the sample. Compared to proto metal, the energy cut-offs were substantially higher
for more aggressiveness, and despite the intensive middle range, classic metal guitarists already
started scooping the sound. 1990s thrash metal marked a substantial change in metal guitar
history. Although still relying on early metal conventions like standard tuning and two guitar
tracks, the sound had considerably more low-end, less mids and more distortion making it heav-
ier than most metal before. Extreme metal of the last fifteen years continued to extent the bass
range by using lower tunings along with seven string guitars, new amplifiers and more track
layering. It further developed the metal guitar sound by exploiting heaviness while combining
traditional practices with new production techniques.
The experimental part of the study closer investigated production techniques. It demon-
strated how the instrument’s sound could be shaped by equipment and production choices to
improve frequency coverage for a dense sound, or to achieve the desired balance between the
frequency groups. Even though the basic techniques have hardly changed since early metal,
producing a contemporary metal guitar sound requires much balancing at the recording and
mixing stages due to the high expectations within the genre and the powerful amplifiers having
to be tamed. Focussing on the low-end only is likely to lead to muddiness and lack of power.
Although the study cannot provide a formula or replace experience, the results may be a starting
point for advancing artists helping them to produce the sound of one of the subgenres discussed,
to fix problems, or to shape the sound to the desired direction. Depending on the song and
intended sound aesthetics, for instance, it may be more effective recording with a less bass-
heavy amplifier, and deliberately setting the microphone position and distance to get the
warmth without the amplifier’s characteristic emphasis on the bass range; or combining a bass-
heavy device with an overdrive pedal and to equalise the missing spectral content. All these
combination possibilities and their special effects interrelate with layering. At this point, aca-
demic study on guitar production has reached its limit, and individual experimentation must
As Williams (2015: 61) noted, “timbral measurement of specific guitar-related timbres for
metal“ is a field of research that has been hardly studied so far (Mynett 2009a, 2009b, 2012,
2013). Similarly, only a few production manuals include metal genres (Chappell 2010;
Hamidovic 2015). Research and production literature are based mainly on practical experience
or are exploratory with small case numbers (Berger and Fales 2005; Williams 2015). The pre-
sent study contributes to research by providing insights into the development of the metal guitar
sound based on a greater sample size, bigger time-span, and more variables such as frequency
groups, ratios, tuning and layering. In addition, the experimental investigation makes the first
attempt to establish an empirical ground useful for guitarists, engineers and producers. The
academic field of metal music studies may benefit from the deeper understanding of the issue,
as well as from the acoustic evidence for theoretical and practice-based research, and a new
methodical approach for studying metal sounds.
The findings are subject to a number of limitations. Genre analyses bear the risk of over-
simplifications, and therefore the results cannot be definitive. A methodological challenge has
been the comparative analysis of sound phenomena. Whereas the experimental part could de-
pend on the same structure, the historical analysis was based on different guitar riffs that inev-
itably affected the frequencies. Methodically, the focus on spectral features meant a limitation
on the sustain phase in which instrument timbres differ little. On top of that, dividing the spec-
tral content into frequency groups caused a loss of information. Moreover, the bass guitar and
the kick drum as crucial instruments for the genre’s perception of heaviness were not consid-
ered. To gain deeper insights into the metal guitar, future research should be taking greater
account of the differentiation between time and genre, and include the transient design and
dynamics. It could also explore the development of aesthetics, playing and production of other
instruments aiming at a holistic understanding of metal genres. In conformity with Williams’
(2015) and Mynett’s (2013) claims, further research is needed to develop methods for research-
ing metal productions and sounds.
Albrecht, C. (2010), Der Tonmeister. Mikrofonierung akustischer Instrumente in der Popmusik,
Berlin: Schiele & Schön.
Bartlett, B. and Bartlett, J. (2009), Practical recording techniques. The step-by-step approach
to professional audio recording, 5. ed., Amsterdam: Elsevier Focal Press.
Berger, H. M. (1999), Metal, rock, and jazz. Perception and the phenomenology of musical
experience, Hanover, N.H.: University Press of New England.
Berger, H. M. and Fales, C. (2005), ‘“Heaviness“ in the Perception of Heavy Metal Guitar
Timbres. The Match of Perceptual and Acoustic Features over Time’, in P. D. Greene and
T. Porcello (eds.), Wired for sound. Engineering and technologies in sonic cultures. Mid-
dletown, Conn: Wesleyan University Press, pp. 181-197.
Bourdage, M. (2010), ‘”A Young Girl’s Dream”. Examining the Barriers Facing Female Elec-
tric Guitarists’, Journal of the International Association for the Study of Popular Music,
1(1), pp. 1-16.
Chappell, J. (2010), The Recording Guitarist. A Guide to Studio Gear, Techniques, and Tone,
Winona, MN: Hal Leonard.
Cook, N. and Clarke, E. (2004), ‘Introduction: What Is Empirical Musicology?’, in N. Cooke
and E. Clarke (eds.), Empirical Musicology. Aims, Methods, Prospects, Oxford: Oxford
University Press, pp. 3-14.
Doyle, M. (1993), The Sound of Rock. A History of Marshall Valve Guitar Amplifiers, West-
port: The Bold Strummer.
Einbrodt, U. D. (1997), Experimentelle Untersuchungen zum Gitarrensound in der Rockmusik,
Frankfurt am Main: Peter Lang.
Gracyk, T. (1996), Rhythm and noise: An aesthetics of rock, Durham: Duke University Press.
Hagen, R. (2011), Musical style, ideology, and mythology in the Norwegian black metal, in J.
Wallach, H. M. Berger and P. D. Greene (eds.), Metal rules the globe. Heavy metal music
around the world, Durham: Duke University Press, pp. 180-199.
Hall, D. E. (1980), Musical Acoustics. An Introduction, Belmont: Wadsworth.
Hamidovic, E. (2015), The Systematic Mixing Guide, Melbourne: Systematic Productions.
Herbst, J.-P. (2014), Netzwerk Sound, Augsburg: Wißner.
Herbst, J.-P. (2016), Die Gitarrenverzerrung in der Rockmusik. Studien zu Spielweise und
Ästhetik, Muenster: LIT.
Iommi, T. (2012), Iron Man. My journey through heaven and hell with Black Sabbath, London:
Simon & Schuster.
Izhaki, R. (2013), Mixing audio. Concepts, practices and tools, 2. ed., Oxford: Focal Press.
Jauk, W. (2007), ’Der Sound des hedonisch-performativen Körpers und das Spiel der
Elektrogitarre’, in F. Krieger and B. Hoffmann (eds.), Jazz Research, 39, Graz:
Akademische Druck- und Verlagsanstalt, pp. 273-289.
Juchniewicz, J. and Silverman, M. J. (2011), ‘The influence of progression type and distortion
on the perception of terminal power chords’, in Psychology of Music, 41(1), pp. 119-130.
Lemme, H. (1995), Gitarren-Verstärker-Sound, Muenchen: Pflaum Verlag.
Lilja, E. (2005), Characteristics of Heavy Metal Chord Structures. Their Acoustic and Modal
Construction, and Relation to Modal and Tonal Context, Licentiate Thesis, University of
Lilja, E. (2015), Dealing with the 3rd: Anatomy of distorted chords and subsequent composi-
tional features of classic heavy metal, in T.-M. Karjalainen and K. Kärki (eds.), Modern
Heavy Metal – Markets, Practices and Cultures, Helsinki: Aalto University press.
Martinelli, R. (2006), Interview: Andy Sneap, Maelstrom, 45, http://www.mael-
stromzine.com/ezine/interview_iss45_208.php. Accessed 16 August 2016.
Martinelli, R. (2008), Interview: Colin Richardson, Maelstrom, 61, http://www.mael-
stromzine.com/ezine/interview_iss61_254.php. Accessed 16 August 2016.
Moore, A. F. (2002), Authenticity as authentication, Popular Music, 21(2), pp. 209-223.
Mynett, M. (2009a), ‘Extreme Metal. The SOS Guide to Recording & Producing Modern
Metal’, Sound on Sound, November 2009, pp. 120-133.
Mynett, M. (2009b), ‘Mixing Metal. The SOS Guide to Extreme Metal Production: Part II’,
Sound on Sound, December 2009, pp. 118-126.
Mynett, M. (2012), ‘Achieving Intelligibility Whilst Maintaining Heaviness When Producing
Contemporary Metal Music’, Journal on the Art of Record Production (6).
Mynett, M. (2013), Contemporary Metal Music Production, PhD thesis, University of Hud-
Owsinski, B. (2006), The Mixing Engineer’s Handbook, 2. ed., Boston: Thomson.
Reyes, I. (2013), Blacker than Death: Recollecting the “Black Turn” in Metal Aesthetics, Jour-
nal of Popular Music Studies, 25(2), pp. 240-257.
Roederer, J. G. (1973), Introduction to the Physics and Psychophysics of Music, New York:
Schwarz, F. (2006), ’Hammerfall’, Guitar, 11/06, pp. 26-28.
Taylor, M. (2011), ‘Interview: Metal production guru Colin Richardson’, Musicradar,
son-400260. Accessed 16 August 2016.
Waksman, S. (1999), Instruments of Desire. The Electric Guitar and the Shaping of Musical
Experience, Cambridge: Harvard University Press.
Walser, R. (1993), Running with the Devil. Power, gender, and madness in heavy metal music,
Hanover, NH: University Press of New England
Weinstein, D. (2000), Heavy metal. The music and its culture, Cambridge, Mass.: Da Capo
Williams, D. (2015), ‘Tracking timbral changes in metal productions from 1990 to 2013’, Metal
Music Studies, 1(1), pp. 39-68.
Zbikowski, L. M. (2002), Conceptualizing music: Cognitive structure, Theory, and Analysis,
New York: University Press.
Zollner, M. (2014), Physik der Elektrogitarre, Regensburg: Zollner.
Amplifier settings (all in lead channel)
Laney GH50L: Bass 5, Middle 5, High 5, Presence 5, Gain 7, Drive 6, Level 3
Marshall JCM2000 TSL: Bass 5.5, Middle 7, Treble 5.5, Presence 5, Volume 4, Gain 8
Mesa Boogie Triaxis: Gain 8, Treble 9, Middle 8, Bass 1, Lead Drive 8, Master 9, Presence 9,
Orange Dual Terror: Tone 7, Volume 10, Gain 8.5
Peavey 5150: Bass 8, Middle 8, High 5, Vol 2.5, Resonance 8, Presence 4
A recent empirical investigation studying the use of equipment of 413 guitar players indicated
significant differences between subgenres of rock and metal (Herbst 2016). Whereas hard rock
and heavy metal players are still preferring traditional 100W tube amplifiers with separated
head and cabinet, new metal and metalcore guitarists tend to rely on transistor or hybrid tech-
nology and rack designs.
The frequency range ”air“ (10-20 kHz) is not included in the experimental part due to the