How Deep Must the Bass Be?

[kemmler3D]

Coming in late but here's my take / opinion:

For real fidelity you want a system that can play flat down to 20hz in-room. This is not necessarily flat anechoic, but you need something going on down to 20hz. It should keep up with whatever listening level you prefer. 93-100dB is probably more than enough for most people with neighbors they'd like to stay friends with. EQ is a must to deal with room modes in any case.

Even if the music you listen to "doesn't have anything" below 40hz, it *should*. To produce a perfect transient, you need energy at all frequencies. This applies to any sharp transient including cymbal strikes.

Now, you don't really hear the low bass in a cymbal strike, but percussion of all kinds theoretically benefits from as much bass as you can get. This is immediately noticeable if you have good subs plays low drums that haven't been highpassed in the recording. In my opinion this obviates any arguments about the lowest notes a piano, bass guitar, or organ plays.

Personally I don't perceive much "sound" below 20hz so going to 16 seems iffy to me. With my LCD-XCs if I play a 16hz tone I just get the sense that my house is shaking, or a whooshing in my ears, but definitely not a tone. Keep in mind those 4hz are almost half an octave down there, so it's a big difference in pitch if not absolute frequency.

 

[NIN]

Whilst I have never been to a local concert I was at a sponsorship event once in Japan where there was a demonstration of these drums, including the big one.
We were indeed outside and whilst I didn’t have a sound meter with me I don’t think, at the distance we were listening, they were any louder than I can get in my music room.

The dynamic range of recordings is usually the problem, no recording distributor in their right mind would try to sell a recoring of wide dynamic range music with its full dynamic range preserved since 99% of the people buying it wouldn’t be able to replay it.

There are a few “audiophile” recordings demonstrating the dynamic range of real life but very few and usually just lollipops.

I can recommend Josef Strauss - Ein straussfest (Cincinnati Pops Orchestra) from Telarc. Well recorded and some VERY dynamic tunes, even so much that they have a warning about it.

 

[pablolie]

...

Personally I don't perceive much "sound" below 20hz so going to 16 seems iffy to me. With my LCD-XCs if I play a 16hz tone I just get the sense that my house is shaking, or a whooshing in my ears, but definitely not a tone. Keep in mind those 4hz are almost half an octave down there, so it's a big difference in pitch if not absolute frequency.

20Hz is universally called the start of the infrasonic range, which by definition means you feel it rather than hear it. My system can produce it (isn't that hard, really), as I have stated before I don't think it does anything for my music enjoyment. It's a warble rather than a tone. And I know if I boosted it for "effect" I'd hate the sound. I have been to real life events where deep bass gets inside your body, and to me it was never a pleasant experience, and most certainly not something I want to replicate at home.

Inevitably, at social events, there's some wise-ass that listens to music on my system and goes "sounds good, but it'd be SWEET with a subwoofer!" and I go "there's a carefully calibrated one behind the couch, check it out" and make sure I never invite them again.

That said, I grant everybody their preference, unless they tell me I am un-informed and am missing on some experience that renders me a third class music lover...

 

[Bugal1998]

I respect your opinion, but with those dB levels you are causing damage to your hearing.

I didn't say we needed to routinely use all that SPL; and I specifically said we shouldn't. But the reality is that bass at 30-40hz is typically 20db louder than the music at 1khz.

If someone is listening to music at 75db at 1khz, then that could easily put the 30-40hz region at 95db. Turn it up a few dB or encounter a particularly bass heavy track and you have no headroom.


If the mixer pushed lots of low frequency energy at 20 hz, one could easily you want to turn it up occasionally, or have a song that has a bit more bass, you're out of headroom.

... yes, air conditioning noise in a concert-hall. [...] That's what I meant by acoustic garbage I do not want to hear.

That's interesting. I have found recordings with audible low frequency HVAC noise in the recording can offer some of the greatest sense of 'being there'. The HVAC gives a sense of room space that overrides my own room. I've never found it offensive.

Different strokes I guess.


Edit: P.S. I somehow lost half of this post when watching my toddler, and the latter portion of the first half is now incomprehensible jibberbish. I'll repost it later tonight.

 

[pablolie]

I didn't say we needed to routinely use all that SPL; and I specifically said we shouldn't. But the reality is that bass at 30-40hz is typically 20db louder than the music at 1khz.

If someone is listening to music at 75db at 1khz, then that could easily put the 30-40hz region at 95db. Turn it up a few dB or encounter a particularly bass heavy track and you have no headroom.

I have never ever heard that is a universal rule, even rule of thumb, and I have followed this hobby for close to 40 years. And any mix that turns the bass up artificially to me is garbage if it is acoustic instrumentation.

If the mixer pushed lots of low frequency energy at 20 hz, one could easily you want to turn it up occasionally, or have a song that has a bit more bass, you're out of headroom.

Repeat statement above about the "mixer" artificially pushing up bass in any instrumental genre I care about. I'd fire the incompetent mother&^%! And I am not saying those sad occurrences don't exist. There are thousands of examples of bathtub mixes in order to make stuff sound ok-ish on sh*tty systems. Doesn't mean it's worth further amplifying on a good system to stay "true to the (pathetic) original". Please find us correction.

 

[NTK]

I didn't say we needed to routinely use all that SPL; and I specifically said we shouldn't. But the reality is that bass at 30-40hz is typically 20db louder than the music at 1khz.

If someone is listening to music at 75db at 1khz, then that could easily put the 30-40hz region at 95db. Turn it up a few dB or encounter a particularly bass heavy track and you have no headroom.

Loud low frequency sound is not nearly as damaging as in the 1 - 8 kHz range. (Source: attachment in this Dr. Toole's post at AVSFourm)

 

[pablolie]

Loud low frequency sound is not nearly as damaging as in the 1 - 8 kHz range. (Source: attachment in this Dr. Toole's post at AVSFourm)

View attachment 324341

The problem is music is not guaranteed to come in such nice frequency domain isolation. But if you guys want to listen to stuff at 115dB, hey, go for it.

 

[Bugal1998]

I have never ever heard that is a universal rule, even rule of thumb, and I have followed this hobby for close to 40 years. And any mix that turns the bass up artificially to me is garbage if it is acoustic instrumentation.



Repeat statement above about the "mixer" artificially pushing up bass in any instrumental genre I care about. I'd fire the mother&^%!

You can find academic and industry research on the topic of energy distribution in music; it's not a rule by any means, but it does tend to be what's tonally pleasing... Of course it can vary by genre in both directions.

I try to remove my personal preferences from these types of discussions as much as is reasonably possible. Unless someone says they only listen to classical or acoustic music (if the OP did, I missed it), I assume all music is in scope (whether I like it or not), which includes rap, electronic, EDM, etc. and discuss what a system would need to do to cover it all.

And if they already know they don't like bass above a certain level, well then their limits are personal and likely don't warrant a discussion about required subwoofer output.

Oh and lastly, I wasn't offended in any way by your cautionary statement; wise words!

P.S. I somehow lost half of my prior reply when watching my toddler and the latter portion is incomprehensible jibberbish. I'll repost it later tonight.

 

[Chrispy]

The problem is music is not guaranteed to come in such nice frequency domain isolation. But if you guys want to listen to stuff at 115dB, hey, go for it.

Subs and audio systems are just used for wider use than your musical tastes?

 

[NIN]

The problem is music is not guaranteed to come in such nice frequency domain isolation. But if you guys want to listen to stuff at 115dB, hey, go for it.

Nothing is guaranteed. It is no problem to listen and to find a lot of songs/albums that have the spectral distribution that gives the possibility to play the music loud.

 

[waynel]

I have never ever heard that is a universal rule, even rule of thumb, and I have followed this hobby for close to 40 years. And any mix that turns the bass up artificially to me is garbage if it is acoustic instrumentation.



Repeat statement above about the "mixer" artificially pushing up bass in any instrumental genre I care about. I'd fire the incompetent mother&^%! And I am not saying those sad occurrences don't exist. There are thousands of examples of bathtub mixes in order to make stuff sound ok-ish on sh*tty systems. Doesn't mean it's worth further amplifying on a good system to stay "true to the (pathetic) original". Please find us correction.

What you are missing here is that due to the sensitivity of our ears as a function of frequency, much higher content at lower frequencies is what is natural and neutral in music

Please see this paper:

http://ltu.diva-portal.org/smash/get/diva2:1557981/FULLTEXT01.pdf

Here is an excerpted plot showing the spectrums of some types of music, classical and jazz are not much different in spectrum.

Here is another example:

 

[Don Hills]

...

Can anyone comment on whether his tapped horn theory is for real? I once Googled this and found something by some acoustics professor somewhere supposedly proving that tapped horn theory was bs. I only found a snippet of what he published, not the whole thing, and I did not find out who he was either.

In "Acoustics" by Leo Beranek, I had read that the size of the mouth of a horn had to be really huge just to output 20Hz-about the size of a walk-in fireplace in a castle, so how could a Danley DTS-10 horn, which is not even close to being that huge, act as a horn at even lower frequencies?
...

A "tapped horn" is not a horn. It's a quarter-wave resonator, like an organ pipe. The "tap" causes it to have two resonances, one an octave above the other, thus widening the usable frequency range.

 

[pablolie]

You can find academic and industry research on the topic of energy distribution in music; it's not a rule by any means, but it does tend to be what's tonally pleasing... Of course it can vary by genre in both directions.

I try to remove my personal preferences from these types of discussions as much as is reasonably possible. Unless someone says they only listen to classical or acoustic music (if the OP did, I missed it), I assume all music is in scope (whether I like it or not), which includes rap, electronic, EDM, etc. and discuss what a system would need to do to cover it all.

And if they already know they don't like bass above a certain level, well then their limits are personal and likely don't warrant a discussion about required subwoofer output.

Oh and lastly, I wasn't offended in any way by your cautionary statement; wise words!

P.S. I somehow lost half of my prior reply when watching my toddler and the latter portion is incomprehensible jibberbish. I'll repost it later tonight.

And truly I was not trying to be contentious. I respect anyone's opinion on the topic.

What you are missing here is that due to the sensitivity of our ears as a function of frequency, much higher content at lower frequencies is what is natural and neutral in music

Please see this paper:

http://ltu.diva-portal.org/smash/get/diva2:1557981/FULLTEXT01.pdf

Here is an excerpted plot showing the spectrums of some types of music, classical and jazz are not much different in spectrum.
View attachment 324344
Here is another example:
View attachment 324345

Yes, but that kind of could possibly re-enforce the point that subsonic low frequencies are not relevant to music delivery, doesn't it?

I have seen it discussed before, and it is incredibly track dependent. I don't dispute it the least. There *is* a non-linear quality to human hearing, we all hopefully know that. Perhaps that deserves a different topic, but admittedly it is intricately weaved into this topic too.

 

[Axo1989]

The problem is music is not guaranteed to come in such nice frequency domain isolation. But if you guys want to listen to stuff at 115dB, hey, go for it.

We have different musical preferences of course, but I often check sound levels and spectra when listening to music and find the slopes shown on this page fairly representative.

Measuring z-weighted, here is a quieter moment off-beat followed by a short-term peak two seconds later. That's at LP ~2.5 metres from main speakers and represents fairly loud home listening (an a-weighted measure would be somewhat lower). While we do see peaks in the mid/treble they are typically lower SPL than the lowest octave.

We may be underestimating the lowest octave somewhat here due to limitations of iPhone app (albeit calibrated) and we still want to have some headroom beyond this (hence the desired 115 dB capability).

It's important to keep in mind that with this playback level the actual measured duration of the peaks logged by the dosimeter indicate it's safe to listen to an album at this level without hearing damage. Here we are 10 minutes in.

Also, the subjective impression in-room is balanced despite the slope. Anyway try it for the music you listen to and see how it looks, I'd be interested.

 

[pablolie]

We have different musical preferences of course, but I often check sound levels and spectra when listening to music and find the slopes shown on this page fairly representative.

Measuring z-weighted, here is a quieter moment off-beat followed by a short-term peak two seconds later. That's at LP ~2.5 metres from main speakers and represents fairly loud home listening (an a-weighted measure would be somewhat lower). While we do see peaks in the mid/treble they are typically lower SPL than the lowest octave. We may be underestimating the lowest octave somewhat due to limitations of iPhone app (albeit calibrated) but we still want to have some headroom beyond this. It's important to keep in mind that with this playback the actual measured duration of the peaks logged by the dosimeter indicate it's safe to listen to an album at this level without hearing damage.

View attachment 324353

View attachment 324355

Also, the subjective impression in-room is balanced despite the slope. Anyway try it for the music you listen to and see how it looks, I'd be interested.

Which app is this, I'd be curious?

I can share an experience that nearly did in my ears... effing Matthew Herbert in his appearance in Deutsche Grammophon's "Recomposed" series, I seem to recall it was a Mahler composition (notorious for dynamic range as is) (I just checked and yes, it is Mahler's post-humous and unfinished X), but he exaggerated it... so you turn up the volume quite a bit wondering what they were thinking engineering the sound, and then a few minutes into it they criminally BANG up the volume. If one likes extreme DR this is one for you. But I was so upset I deleted the album from my collection, which is something that I never do.

PS: Sir Simon Rattle's Mahler X album on EMI is a good thrill too, albeit not in any way as extreme as the Herbert re-composition.

 

[waynel]

And truly I was not trying to be contentious. I respect anyone's opinion on the topic.

Yes, but that kind of could possibly re-enforce the point that subsonic low frequencies are not relevant to music delivery, doesn't it?

No, I think you are drawing the wrong conclusion, it reinforces they point that in order to be audible and sound balanced subjectively , low frequencies exist in music and need to be reproduced at objectively high amplitudes relative to midrange and high frequencies. This is why good subwoofers have a ton of headroom and can hit 100dB plus.

 

[Axo1989]

Which app is this, I'd be curious?

That one is Decibel X on iPhone, there are a few similar. I think I've paid for Pro features so not sure how much functionality you get for free. Anyway, like you I am certainly interested in keeping my hearing, so I keep an eye on it.

I can share an experience that nearly did in my ears... effing Matthew Herbert in his appearance in Deutsche Grammophon's "Recomposed" series, I seem to recall it was a Mahler composition (notorious for dynamic range as is), but he exaggerated it... so you turn up the volume quite a bit wondering what they were thinking engineering the sound, and then a few minutes into it they criminally BANG up the volume. I was so upset I deleted the album from my collection, which is something that I never do.

Oh, tricky. Actually cool to have such recorded DR but it should come with a warning on the wrapper.

While I expect the most sensitive/damaging frequencies are weighted to the mid-range, it is of course possible to do damage with bass. I had a car system a while back that I gave up on, for that reason. But my baby sister introduced me to the best of screamo driving to school so worth it.

 

[pablolie]

No, I think you are drawing the wrong conclusion, it reinforces they point that in order to be audible and sound balanced subjectively , low frequencies exist in music and need to be reproduced at objectively high amplitudes relative to midrange and high frequencies. This is why good subwoofers have a ton of headroom and can hit 100dB plus.

Honestly I don't quite follow. So you are saying to hear the *natural* levels, lower frequencies need to be boosted? Odd then that we enjoy live venues if the lower frequencies are so wimped out, which I don't think is what you mean.

My point is that I want that natural delivery. I don't want anything artificially boosted. When I still had a Sansui or Luxman or Accuphase amp with a "loudness" button, I never ever pushed it. But that is my personal preference only and I respect other's choices. Based on that preference, and even though my system can reach down to 20Hz (I think by spec it is at -3dB), I regularly turn off the sub, and the xover point on the DAC-amp is 70Hz (but we know there is a roll-out that probably means it still reaches to 40Hz, albeit pushed down) and while I can tell something is a bit less lively with some tracks, I don't enjoy it any less.

Why do I turn the sub off? My GF is a constant study person, and the sub sometimes bugs her in learning even with noise canceling headphones on.

 

[waynel]

Honestly I don't quite follow.

yes

So you are saying to hear the *natural* levels, lower frequencies need to be boosted?

No, but lower frequencies are objectively higher amplitude in almost all music even though they don't sound louder due to the way the ear works.

Odd then that we enjoy live venues if the lower frequencies are so wimped out, which I don't think is what you mean.

No, It's not what I mean, If you measure the sound power at whatever live venue/performance that you are enjoying, you will see that the low frequencies are higher amplitude and the spectrum is downward sloping just like the plots that I posted above. Therefore, when played back on a neutral system the spectrum will also slope downward in the same manner and the measured spectrum will have higher amplitude at lower frequencies.

 

[pablolie]

yes

No, but lower frequencies are objectively higher amplitude in almost all music even though they don't sound louder due to the way the ear works.

No, It's not what I mean, If you measure the sound power at whatever live venue/performance that you are enjoying, you will see that the low frequencies are higher amplitude and the spectrum is downward sloping just like the plots that I posted above. Therefore, when played back on a neutral system the spectrum will also slope downward in the same manner and the measured spectrum will have higher amplitude at lower frequencies.

Appreciate your reply and clarification, and have zero objections to your point.