(Note 1: This research project was to satisfy my own curiosity. Consider this a collaborative document, to be updated or debunked as better evidence is available. Throw in your informed counterpoints or links below.
Note 2: I’d normally ask for permission for copyrighted work, but S&V went kaput in 1996 [1] [2]. Sin first, ask forgiveness later.)

Wrong answer: I don’t need a subwoofer. [1]

Short answer: It’s complicated, but we’re probably overthinking it: pick a reputable and reasonably large subwoofer first; sealed, ported or passive radiator enclosures are fine for movies and music. Bandpass is for SPL drag racing or non-critical listening only.

Long answer (with ancillary reading):
A. Good bass matters. According to Dr. Sean Olive et al., bass makes up 30% of the weighting in a loudspeaker’s perceived sound quality. [10, Chapter 4.4]

B. Subwoofer enclosures can be subdivided into sealed, ported, bandpass, or infinite baffle. Multiple driver subwoofers adds isobaric or push-pull configurations to these enclosure types. [2]

C. Bandpass subwoofers are the lowest fidelity design and should be avoided because they make no pretense about accuracy. Infinite baffle should be treated as a custom project of last resort due to their inability to achieve repeatably good results. [2]

D. Sealed subwoofers are favored by loudspeaker designers for less perceived overhang due to “phase rotation, lower group delay, and reduced ringing in the time domain”. [3] [4]

E. Group delay is measured in milliseconds. The threshold of audibility is frequency dependent. For the lowest music octave (20-40 Hz), we are extremely tolerant of group delay (> 14 ms), before even studying psychoacoustic preference. [9]

F. Is there a >14ms gap between sealed and ported subwoofers today? I couldn’t find any discussions on this topic. But if you’re concerned about phase, audio video receivers and pre-pros with room calibration software will measure the distances of individual room speakers for the purpose of time alignment and volume adjustment. [13]

G. Dr. Floyd Toole et al. believes that timing is a red herring. “Impulse response and time-domain ringing of resonances are already accounted for over most of the frequency range because loudspeaker transducers are minimum-phase devices - a flat, smooth frequency response indicates the absence of such misbehavior.” [10, chapter 14.1]

H. In other words, group delay and ringing issues would show up in a frequency response graph as a deviation from flat. That means good frequency response measurements matter more for “musicality” than anything inherent to port and sealed enclosures. It also means that good frequency response curves equals negligible time domain issues. Win-win. It’s interesting that most manufacturers won’t publish their subwoofers’ frequency response measurements or target curves.

I. Loudspeakers naturally roll off their bass, offset somewhat by a 12 dB/octave in-room gain. [4]

J. Sealed and ported enclosures inherently have different roll-off slopes of 12 dB/octave and 24 dB/octave respectively. [4] There are many ideas about what the ideal roll-off slope should be in relationship to in-room gain, but there seems to be a consensus among manufacturers that non-flat “movie mode” profiles are for movie playback only. Inexpensive and poorly engineered subwoofers might not even have flat frequency response. [uncited, anecdotal experience]

K. A flat frequency response down to 20 Hz or less is a recent innovation, using a combination of Digital Signal Processing equalization (DSP), prodigious amplifier power, AND large sealed or ported cabinet size. (I found examples that publish this as their performance target: the Axiom EP500, EP600, and EP800; Hsu Research ULS-15 Mk2; Monoprice Monolith M-15 V2 and M-215; Paradigm Signature SUB 2; and Rogers Speedwoofer 12S)

L. Kilowatt micro subs and pancake subs have decor-friendly compact dimensions but seldom reach below 35 Hz. They may have unsatisfyingly shallow F3 and F10 bass extension, or may be outperformed by simpler, cheaper, but bigger designs. [8] [14]

M. Anomalies in frequency response shows up as fast or slow bass (a misnomer; sound travels at the speed of sound), muddiness, or punch.

N. Informed audio journalists believe bad subwoofers are the exception today. [6] So what to make of subwoofer head-to-head comparisons with clear winners and clear losers (eg. The Wirecutter, Erin’s Audio Corner)? The commonality I see is that testing is done at ear-bleeding volumes that favor power, robust construction, and size. The absence of distortion then takes priority over frequency response. Distortion is a non-issue at normal volume when operating within its intended performance window. Mediocre subwoofers distort early and often.

O. We are forgiving of up to 6 dB clean amplifier clipping and up to 1% total harmonic distortion (THD), but object to >8% THD. Manufacturers can economically use limiters and compression to reduce THD. There are entire audiophile subcultures offended by these equipment, warranty, and hearing preserving techniques, with CEA-2010 tests designed to identify the culprits. Frequency response performance appears to be irrelevant in these tests. [8] [10, Chapter 4.4] [11] [14]

P. Infrasonic frequencies below 20 Hz are felt rather than heard. There seems to be disagreement whether they should be reproduced or removed using an infrasonic filter. It’s a non-issue for lesser subwoofers unable to reach infrasonic territory. [uncited]

Q. Configuring a subwoofer for your room and listening positions is easy to mess up and time consuming to optimize. [6] Multiple subwoofers and room correction software may be necessary for multiple seating positions. [5] [7] [12] But the results are worthwhile. [1]


Daniel Ho

Sources

1: “An Audio Revelation”, Alan Lofft, Sound & Vision Magazine (Canada), November/December 1994
https://www.dropbox.com/scl/fi/5qpc...rlkey=q0xe04blsckeynh0dzyhsqbt8&dl=0

2: “Choosing a Car Subwoofer”, Derek Lee, Sound & Vision Magazine (Canada), July/August 1994
https://www.dropbox.com/scl/fi/1eeq...rlkey=dhfg2c666ahehx5gh7k0lnmgz&dl=0

https://www.dropbox.com/scl/fi/yhsa...rlkey=rak6atnlr501lti4dwwxhiktt&dl=0

3: “Sealed vs Ported Subwoofers”, SVS, https://www.svsound.com/blogs/subwoofer-setup-and-tuning/75367747-sealed-vs-ported#

4: “Sealed vs Ported Subwoofers: Which is Right for You?”, Steve Munz, Audioholics, https://www.audioholics.com/loudspeaker-design/sealed-vs-ported-subwoofers

5: “Subwoofers: Optimum Number and Locations”, Todd Welti, Harman International, https://www.harman.com/documents/multsubs_0.pdf

6: Topic: why audiophiles don’t own subwoofers. Brent Butterworth, Dennis Burger, “Episode 18”, Audio Unleashed Podcast, (time cued), https://youtu.be/hQuRnxZYdPQ?t=3215

7: “Subwoofers: 4, 2, or 1?”, Brent Butterworth, Sound & Vision Feb 2011 https://www.soundandvision.com/content/subwoofers-4-2-or-1-0

8: “CEA-2010: A Good Baseline for Bass?”, Brent Butterworth, Sound & Vision Sept 2011 https://www.soundandvision.com/content/cea-2010-good-baseline-bass

9: “Audibility of Group Delay for Sound System Design”, Scott Hinson, Audio Science Review forum

10: “Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms Third Edition”, Floyd E. Toole, ISBN 978-1138921368

11: “Hearing Distortion”, Ian G Masters, Masters on Audio, April 1 2004, https://web.archive.org/web/20050210050742/http://www.mastersonaudio.com/features/20040401.htm

12: “Loudspeakers and Rooms for Multichannel Audio Reproduction Part 3 – Getting the Bass Right”, Floyd E Toole, Harman International,
https://www.harman.com/documents/LoudspeakersandRoomsPt3_0.pdf

13: “Home theater calibration 101: speaker levels, distance, and subwoofer phase”, Nyal Mellor, Acoustics Frontiers Blog, Nov 2011 https://acousticfrontiers.com/blogs...ker-levels-distances-and-subwoofer-phase

14: “Can technology produce small subwoofers with big bass?”, Andrew Welker, Axiom Audio, https://youtu.be/-N72-xoRImc