Study of Preference for Stereo Recording Techniques of a

Pelog-based of Gamelan Gender, Bonang, and Peking Ensemble

Nyoman Kieran Sebastian, Jack A. Simanjuntak

Conservatory of Music, Pelita Harapan University, Karawaci, Tangerang, Indonesia

Keywords: Stereo Recording Techniques, A Pelog-based of Gamelan Gender Bonang and Peking.

Abstract: The layout of gamelan ensemble produce stereo imaging, instrument separation, and sound clarity. Until now,

for music production purposes, gamelan ensemble recording uses spot-based microphone technique, which

produces more individual sound, while the stereo microphone technique is able to produce the characteristics

of gamelan ensemble, either in instrument separation, sound clarity, or stereo imaging. No previous study has

investigated on preference for stereo recording techniques of a pelog-based of gamelan gender, bonang, and

peking ensemble. Techniques that applied in this research were XY stereo recording technique, NOS, ORTF,

and AB. This research uses qualitative method, using statistical calculation. There were four stages, collecting

data, which consists of assign the player, analyze the characteristics of the ensemble, choose the song to be

recorded, decide the layouts of the ensemble and the microphones, and create the questionnaire. The second

stage was recording process. The third stage was sample test and interview. These findings is clearly indicate

that NOS stereo recording technique is the most prefered stereo recording techniques of a pelog-based

gamelan gender, bonang, and peking ensemble.

1 INTRODUCTION

Gamelan is a traditional musical instrument in the

form of ensemble from Java Island, Indonesia. The

instruments are made of metal, wood, or bamboo, that

creates sound by hitting it. It is also classified as an

idiophone musical instrument (Tjahyanto et al, 2011).

Gamelan has its own characteristics. Each

instrument of gamelan was created with frequency

range and with different roles (Suyatno et al, 2013).

The characteristics influence the layout of gamelan

ensemble. The layout of gamelan ensemble produces

wide stereo imaging (Suyatno et al, 2013). The

impression of wide stereo image was created by

stereo microphone techniques.

Until now, for music production purposes,

gamelan ensemble recording uses spot-based

microphone technique. No previous study has

investigated on preference for stereo recording

techniques of a pelog-based of gamelan gender,

bonang, and peking ensemble.

The stereo microphone technique is able to

produce the characteristics of gamelan ensemble,

either in instrument separation, sound clarity, or

stereo imaging. For example, Spaced-Pair (AB)

stereo microphone technique is able to produce a

clear stereo image, which makes listener able to

differentiate the position of each sound source, and

determine the boundaries of a room (Bates, 2016).

But when compared with spot-based microphone

technique, the spot-based microphone technique

produces more individual sound.

Other than that, gamelan is also still less well-

known in general, even by Indonesian people

themselves. Iswara (2010) stated that the

insufficiency of gamelan preservation is caused by

the lack of government's participation, foreign

culture, and gamelan which is still rarely known by

Indonesian society, and the lack of interest towards

the gamelan itself (Iswara 2017).

The aim of this research is to analyse four stereo

recording techniques of a pelog-based of gamelan

gender, bonang, and peking ensemble.

2 METHODOLOGY

There are four stereo recording techniques used on

this research:

▪ XY 90°;

▪ NOS 90°;

540

Sebastian, N. and Simanjuntak, J.

Study of Preference for Stereo Recording Techniques of a Pelog-based of Gamelan Gender, Bonang, and Peking Ensemble.

DOI: 10.5220/0009911105400546

In Proceedings of the 1st International Conference on Interdisciplinary Arts and Humanities (ICONARTIES 2019), pages 540-546

ISBN: 978-989-758-450-3

▪ ORTF 110°;

▪ Spaced Pair (AB).

2.1 Data Collection

2.1.1 Instrument Players

There are three players and three instruments on this

research:

▪ Gender: Roban Eko Putranto;

▪ Bonang: Rusli;

▪ Peking: Anastasia Emmanoela Putri.

2.1.2 Ensemble Characteristics Analysis

2.1.3 Song Selection

The song chosen for this study was Gundul-Gundul

Pacul, written by R. C. Hardjosubroto. Music score

courtesy of Rusli.

2.1.4 Ensemble Layout Planning

There are five ensemble layouts used throughout the

recording process. One layout uses the standard

ensemble layout (as 1), and the other four layouts are

adjusted to the characteristics of each stereo recording

techniques (as 2).

2.1.5 Microphone Layout Planning

For the research, the microphone that was used is a

pair of Neumann KM184 as a stereo microphone and

also stand mics. There are 24 microphone layouts that

were used on the recording process (three of each

ensemble layout and stereo recording technique).

Three microphones were placed to the midpoint of all

the ensemble layout with a distance of:

▪ Two point five meter (2,5 m) as A;

▪ Three meters (3 m) as B;

▪ Three point five meter (3,5 m) as C.

So, there were six names for each stereo recording

techniques, namely 1A, 1B, 1C, and 2A, 2B, 2C.

2.1.6 Questionnaire Making

The questionnaire sheet was divided into several

section. The sections consist of application section,

‘statement of being a respondent’ section, explanation

and interview section, as well as the interview

questions (can be found in appendix B essay). The

questionnaire was divided into three stages:

▪ First stage, is to find out the respondents'

preferences for microphone distance towards

gamelan ensemble for all ensemble layouts

(first and second);

Respondents had to choose one out of the three

given audio samples of first layout and one of the

three audio samples of second layout, therefore

respondents had to choose eight out of 24 given audio

samples.

▪ Second stage, is to find out the respondents'

preferences for gamelan ensemble layouts for

all stereo recording techniques;

Respondents had to choose one of two audio

samples from the previous selected audio samples,

therefore respondents had to choose four of eight

audio samples

▪ Third stage, is to find out the respondents'

preferences for stereo recording techniques of

gamelan gender, bonang, and peking ensemble.

Respondents had to choose one of four audio

samples from the previous selected audio samples.

2.2 Recording Process

2.2.1 Recording Process

Devices being used on the recording process:

▪ Universal Audio Digital Apollo 8 audio

interface, by using first and forth channel, and

the amplitude of 40 dB;

▪ Mac Pro Server, mid 2010, 2.8 GHz Quad-Core

Intel Xeon processor, 8 Gb 1066 MHz DDR3

memory, OS Sierra version 10.12.3;

▪ Yamaha HS8 speaker;

▪ Logic Pro X version 10.4.1;

▪ Two pieces of microphone stands;

▪ One pieces of carpet with 159 cm length and

123 cm width.

Audio samples have been recorded and bounced

with 48 kHz 24 bits sample rate. The audio that was

produced from the microphone was directed to the

left and panned as much as minus 64 degrees (-64°)

and the audio that was produced from the microphone

was directed to the right and panned as much as plus

63 degrees (+63°).

2.2.2 Preparation of Audio Samples

The devices being used on the recording process:

▪ MacBook Pro Retina, Retina, 13-inch, mid

2014, processor 2.6 GHz Intel Core i5, memory

8 Gb 1600 MHz DDR3, OS High Sierra version

10.13.5;

▪ Headphone Sennheiser HD 280 Pro 64 ohm;

▪ Logic Pro X version 10.4.1.

Study of Preference for Stereo Recording Techniques of a Pelog-based of Gamelan Gender, Bonang, and Peking Ensemble

541

Format of the audio samples is wav, with 48 kHz

24 bits sample rate.

2.3 Samples Test

2.3.1 Questionnaire

The respondents were divided into two categories:

▪ Professional, gamelan instrumentalist and

sound engineer with minimum five years

experiences;

▪ Non-Professional, musician and non-musician

with age range 16 until 30 years.

The quantitative data collection was done by

letting respondents listen to the audio samples that

have been recorded. After respondents listen to the

audio samples, respondents were asked to fill the

questionnaire. After that, respondents were

interviewed to find out the reasons of their choices

before.

The devices being used to listen on the recording

results:

▪ Focusrite Scarlett 2i2 interface, amplitude on

12 o'clock;

▪ MacBook Pro Retina, Retina, 13-inch, mid

2014, processor 2.6 GHz Intel Core i5, memory

8 Gb 1600 MHz DDR3, OS High Sierra version

10.13.5;

▪ Speaker Sennheiser HD 280 Pro 64 ohm;

▪ QuickTime Player version 10.4 (928.13).

2.3.2 Interview

After filling in the questionnaire, the respondents

were also going to be interviewed.

Qualitative data collection was done by preparing

interview's questions, sound recording device, and

stationery. Interview was done with semi structured

format, with two questions. First question, "why did

the respondents choose the option?". Second

question, "what are the factors that influence the

respondents choices?"

Devices being used on the recording process:

▪ Iphone 6S, 2015, iOS 11.4.1 handphone;

▪ Stationery.

2.4 Data Analysis

2.4.1 Questionnaire

The questionnaire results on the first stage

questionnaire is compiled and analysed on Table 2.1

𝐹𝑂𝑅𝑀𝑈𝐿𝐴:

𝑇𝑜𝑡𝑎𝑙

𝑇𝑜𝑡𝑎𝑙 𝑅𝑒𝑠𝑝𝑜𝑛𝑑𝑒𝑛𝑡𝑠

 100%

The questionnaire results on the second stage

questionnaire is compiled and analyzed on Table 2.2

𝐹𝑂𝑅𝑀𝑈𝐿𝐴:

𝑇𝑜𝑡𝑎𝑙

𝑇𝑜𝑡𝑎𝑙 𝑅𝑒𝑠𝑝𝑜𝑛𝑑𝑒𝑛𝑡𝑠

 100%

Table 2.1: Analysis table from first stage questionnaire.

ICONARTIES 2019 - 1st International Conference on Interdisciplinary Arts and Humanities

542

Table 2.2: Analysis table from second stage questionnaire.

The questionnaire results on the third stage

questionnaire is compiled and analyzed on Table 2.3

𝐹𝑂𝑅𝑀𝑈𝐿𝐴:

𝑇𝑜𝑡𝑎𝑙 𝐿𝑖𝑘𝑒

𝑇𝑜𝑡𝑎𝑙 𝑅𝑒𝑠𝑝𝑜𝑛𝑑𝑒𝑛𝑡𝑠 𝐿𝑖𝑘𝑒

 100%

𝐹𝑂𝑅𝑀𝑈𝐿𝐴:

𝑇𝑜𝑡𝑎𝑙 𝐷𝑖𝑠𝑙𝑖𝑘𝑒

𝑇𝑜𝑡𝑎𝑙 𝑅𝑒𝑠𝑝𝑜𝑛𝑑𝑒𝑛𝑡𝑠 𝐷𝑖𝑠𝑙𝑖𝑘𝑒

 100%

Table 2.3: Analysis table from third stage questionnaire.

2.4.2 Interview

The interview results (qualitative data) have been

processed with three stages:

▪ Evaluation Coding;

Evaluation coding is a systematic data collection

process to increase the effectiveness of the research.

This method focuses on patterned observation or

respondents' answer from detail's quality. This

method also allows the researcher to know which

participator has given positive or negative response.

▪ Data Presentation;

Reduced data was grouped and took the action to

be presented at data conclusion. This stage allows the

researcher to take further actions if there were less

information to understand the interview data analysis

▪ Data Conclusion.

On this stage, all data preparation should be done,

so that data can be presented by bar chart.

3 ANALYSIS AND DISCUSSION

3.1 Preference Analysis of Stereo

Recording Techniques based on

Professional Preferences

3.1.1 First Stage

The first stage most liked stereo recording technique

by professional category preference is XY 2A (50%),

NOS 2C (50%), ORTF 1A (50%), and ORTF 2C

(50%), followed by XY 1B (42.9%), XY 1C (42.9%),

and AB 2C (42.9%).

At the same time, the first stage most disliked

stereo recording technique by professional category

preference is NOS 1A (35.7%), NOS 1C (35.7%), AB

1A (35.7%), and AB 1C (35.7%).

3.1.2 Second Stage

The second stage most liked stereo recording

technique by professional category preference is NOS

2 (85.7%), followed by AB 1 (71.4%), and ORTF 2

(64.3%).

At the same time, the second stage most disliked

stereo recording technique by professional category

preference is XY 1 (50%) and XY 2 (50%).

3.1.3 Third Stage

The most liked stereo recording technique by

professional category preference is NOS (71.42%),

on Figure 3.1. There are several factors:

Study of Preference for Stereo Recording Techniques of a Pelog-based of Gamelan Gender, Bonang, and Peking Ensemble

543

▪ The sound of the gamelan blends together;

▪ Enough reverb;

▪ Clear and balance stereo image;

▪ Clear separation instruments;

▪ Clear gamelan's sound;

▪ Balance loudness;

▪ Cleaner sound;

▪ Comfortable to hear;

▪ Optimal harmonic content;

▪ Balance of gamelan's sound;

▪ The performance;

▪ Tempo;

▪ Stable tone.

Figure 3.1: Diagram value on the third stage of professional

respondent percentage.

The most disliked stereo recording technique by

professional category preference is XY (64.29%), on

Figure 3.1. There are several factors:

▪ Too much noise;

▪ Too much reverb;

▪ Narrow stereo image;

▪ The sound of gamelan does not blend together;

▪ The worst clarity sound;

▪ Less high frequency;

▪ The sound cracked;

▪ The sound was too smooth;

▪ Harmonics at certain frequencies was

annoying;

▪ Less balance;

▪ The performances are not stable;

▪ The dynamics doesn't sound comfortable.

3.2 Preference Analysis of Stereo

Recording Techniques based on

Non-professional Preferences

3.2.1 First Stage

The first stage most liked stereo recording technique

by non-professional category preference is NOS 2C

(64.2%), followed by XY 1B (50%) and ORTF 2A

(50%). At the same time, the first stage most disliked

stereo recording technique by non-professional

category preference is NOS 1B (35.7%) and NOS 1C

(35.7%), followed by XY 2A (42.8%), ORTF 1B

(42.8%), AB 1A (42.8%), AB 2B (42.9%), and AB

2C (42.9%).

3.2.2 Second Stage

The second stage most liked stereo recording

technique by non-professional category preference is

ORTF 2 (78.5%), followed by XY 1 (71.4%) and AB

1 (71.4%). At the same time, the second stage most

disliked stereo recording technique by non-

professional category preference is NOS 1 (57.1%).

3.2.3 Third Stage

The most liked stereo recording technique by non-

professional category preference is NOS (71.42%),

on Figure 3.2. There are several factors:

▪ Comprehensive reverb;

▪ Perfect melody and accompaniment separation;

▪ The sound of gamelan blends together;

▪ Clear melody and accompaniment;

▪ Clear transient;

▪ Balance range frequency;

▪ Sound of the reflection effect from hitting

gamelan;

▪ Balance melody and accompaniment loudness.

The most disliked stereo recording technique by

non-professional category preference is XY (64.29%)

and ORTF (64.29%), on Figure 3.2. There are several

factors:

▪ Too much noise;

▪ Less separation balance;

▪ The sound of gamelan does not blend together;

▪ Too wide stereo image;

▪ Less direct and reflected sound balance;

▪ Annoying resonance;

▪ Less melody sound;

▪ Too much high frequency;

▪ Too rough, sharp, and piercing sound;

▪ Too much noise.

ICONARTIES 2019 - 1st International Conference on Interdisciplinary Arts and Humanities

544

3.3 Preference Analysis of Stereo

Recording Techniques Based on All

Respondent Category Preferences

Figure 3.2: Diagram value on the third stage of non-

professional respondent percentage.

3.3.1 First Stage

The first stage most liked stereo recording technique

by all category preference is NOS 2C (57.1%). At the

same time, the first stage most disliked stereo

recording technique by all category preference is

NOS 1C (35.7%) and ORTF 2A (35.7%), followed by

ORTF 1A (39.2%), AB 1A (39.2%), AB 2C (42.9%),

XY 1B (46.4%) and XY 2A (46.4%).

3.3.2 Second Stage

The second stage most liked stereo recording

technique by all category preference is ORTF 2

(71.4%) and AB 1 (71.4%), followed by NOS 2

(64.2%). At the same time, the second stage most

disliked stereo recording technique by all category

preference is XY 1 (60.7%).

3.3.3 Third Stage

The most liked stereo recording technique by all

category preference is NOS (71.42%), on Figure 3.3.

There are several factors:

▪ The sound of gamelan blends together;

▪ Enough reverb;

▪ Clear gamelan's sound;

▪ Balance loudness;

▪ Cleaner sound;

▪ Noise were not disturbed by the gamelan's

sound.

The most disliked stereo recording technique by

all category preference is XY (64.29%), on Figure

3.3. There are several factors:

Figure 3.3: Diagram value on the third stage of all

respondent percentage.

▪ Too much noise;

▪ Less clear sound;

▪ Less high frequency;

▪ Very less balance;

▪ The performance are not stable.

4 CONCLUSIONS

This research has shown that preference for stereo

recording technique of a pelog-based of gamelan

gender, bonang, and peking ensemble is NOS stereo

recording technique. NOS stereo recording technique

has high definition. However, it has long reverb time.

A long reverb time is the characteristic of gamelan.

There is no sound of gamelan instruments that is

covering one another. Frequency ranges that are

relatively balanced, there are no dominant or less

frequency ranges. Gamelan performances are smooth

and stable. In the majority of NOS samples, there is

no high noise and sound of other disturbances that

disturb the sound of the gamelan. This shows that the

respondents likes the sound of the gamelan which

replicated the room image factor, instrument

separation, and the clarity of gamelan sound during

the live performance.

There is a focus difference between the categories

of professional gamelan respondents, professional

sound engineer respondents and non-professional

respondents. The professional gamelan respondents

focus into the quality of the performance while the

sound engineer and non-professional respondents

categories tends to focus accordance with the research

Study of Preference for Stereo Recording Techniques of a Pelog-based of Gamelan Gender, Bonang, and Peking Ensemble

545

topic, namely the analysis of stereo recording results.

This is because gamelan's professional respondents

plays the gamelan from five years to 40 years,

different from sound engineer professional

respondents who have at least sound experience for

five years, and non-professional respondents who do

not have any experience for gamelan musical

instrument and are not sound engineers for at least

five years. Vocabulary and responses from gamelan

professional respondents are also based on

perceptions, instead of sound engineer professional

respondents who are based on technical matters.

The result of this research indicates that space,

timbre, and quality factors are the aspects which the

professional gamelan respondents focuses, while

space, timbre, quality, and defects are the main focus

of non-professional respondents.

Further research might explore room acoustics

parameter measurements for gamelan's ensemble,

because the room acoustics parameter measurements

affects the gamelan's sound and the recording results.

ACKNOWLEDGEMENTS

Thank you for Bapak Jack Simanjuntak, Kevin, Olin,

and Cung.

REFERENCES

Aris Tjahyanto, Yoyon K Suprapto, dan Diah Puspito

Wulandari, 2011. Model Analysis-By-Synthesis

Aplikasi Pembangkit Suara Gamelan Sintetik. In

Seminar Nasional Aplikasi Teknologi Informasi 2011,

1907-5022.

Suyatno et al., 2013. Karakteristik Akustik Gamelan Jawa

Studi Kasus Gamelan Milik PSTK ITB. In Seminar

Fisika dan Aplikasinya 2013, 2086-0773.

Suyatno et al. 2013. Pengaruh Tata Letak Instrumen

Gamelan Jawa di Panggung Pendhapa ISI Surakarta

Terhadap Parameter Akustik Bagi Penggendang.

Tom Bates, 2016. Coincident or Near-Coincident Mic

Placement Techniques, DPA Microphones Online

Publications.

Noor Hidyat Iswara, 2017. Dinamika Kesenian Gamelan

pada Fungsi dan Pelestarian Kesenian Gamelan dalam

Sanggar Budaya Singhasari di Kecamatan Singosari,

Kabupaten Malang, Jawa Timur. In Universitas

Airlangga, Skripsi Sarjana 1.

ICONARTIES 2019 - 1st International Conference on Interdisciplinary Arts and Humanities

546