Distortion in bass..undetectable when it stays under 20%..?!?

[retro]

What do you guys think of this..?

https://www.axiomaudio.com/blog/distortion

And under 40hz, undetectable until it reaches 100%..!!

So why do we chase low dist woofers and servo control etc,, if we cant hear it anyway..?!?

Apologies if this has been posted before.

 

[oivavoi]

It's more or less in line with the scientific papers I've read on the issue. The actual level of detectability seems to vary between tests and procedures and listeners. Sometime listeners are able to distinguish lower levels of distortion, sometimes higher. But in all of the tests I've seen, we are much less sensitive to distortion in the bass than in the higher frequencies.

Just think of this: Imagine a violin not being played in tune. Doesn't sound good, right? Now imagine an upright bass being played somewhat out tune. Does the thought of that scream in your ear? Probably not. It's as simple as that.

Another issue is that the authors are way off the mark with what they write on TIM and Ottala, perhaps our resident expert @Bruno Putzeys can fill in on that... Also, THD has in general been shown to have a fairly low level of correlation with perception, whereas some other metrics have faired better - such as IMD and two other distortion metrics, one developed by Geddes and one developed by Harman.

EDIT/ADDENDUM: I think the research on bass and perception shows that there are two things that are important:
1) To actually have bass extension
2) To avoid strong room modes, or bass "inertia" and overhang in the room. Methods to avoid that can be through active or passive bass traps, eq, multiple subwoofers, and possibly through the use of dipole bass or sealed subs with little overhang (which is a different issue from distortion).

I think @andreasmaaan also knows a lot about perception of distortion, if I remember correctly.

 

[Ron Texas]

Users of servo control sub's report they sound great. Besides Rythmik subs are competitively priced.

 

[retro]

Users of servo control sub's report they sound great. Besides Rythmik subs are competitively priced.

Sure. I have had lots of servosubs and systems. But the point was that maybe we cant hear those improvements anyway..?!?

 

[Ron Texas]

Sure. I have had lots of servosubs and systems. But the point was that maybe we cant hear those improvements anyway..?!?

There are a lot of things we can't hear. Do we really need amplifiers with SINAD over 80db just because cheap DAC's can do 105db? Remember with room noise and distortion even the best speakers are in the 60db range. I think as long as you don't have to pay extra for it, don't worry.

 

[pozz]

A low-distortion sub will have have shorter decay times and smoother frequency response. There are other benefits to improving design besides a lower THD figure.

 

[DonH56]

A lot of people think a sub with higher distortion sounds better. Why? The harmonics are more readily heard than the fundamental tones and so the distortion makes it sound "richer", "fuller", "louder". That said, IME/IMO, most folk appreciate the sound (or lack of -- more the "feel") of a good sub.

 

[Kvalsvoll]

The article can be misleading, if you do not read and understand what was tested. They did not test for audibility thresholds for harmonic distortion, they tested for audibility of an added signal at various frequency, when playing music.

For harmonic distortion, the threshold is close to what it is for higher frequencies; the lower h will be masked, and higher harmonics approach threshold of hearing. That means for a 80Hz pure sine wave played at 80dB the 5h. can be heard at around 0.04%, 2h around 2%.

But music is not a single pure sine wave. Two properties of typical music changes this audibility limit quite significantly:

- Most low frequency instruments - even computer-generated sounds - has a very high level of harmonic content, adding another 5% on top of 50% makes little difference.
- Other instruments and sound at higher frequencies will mask the distortion.

In this context, the findings described in the article makes sense. But there is music with very loud, clean bass at very low frequencies, where lack of masking sound at higher freqs easily reveals distortion in the bass-system.

When testing for audibility limits at low frequencies, you also need a capable bass-system with distortion levels low enough not to compromise the experiment. The system described in the article does not satisfy this criteria.

 

[RayDunzl]

Most low frequency instruments - even computer-generated sounds - has a very high level of harmonic content, adding another 5% on top of 50% makes little difference.

Electric Bass guitar, single note, raw electrical signal.

Had to try several times to not have the 2nd harmonic come in higher than the fundamental, so the THD measurement would not be 100%.

 

[Julf]

A low-distortion sub will have have shorter decay times and smoother frequency response.

What would that correlation be based on?

 

[pozz]

What would that correlation be based on?

It's not an exact correlation, but if the sub is low distortion then the better FR and decay will result from the effort to control cone velocity, excursion and hysteresis, and particularly voice coil temperature.

 

[tmtomh]

There are a lot of things we can't hear. Do we really need amplifiers with SINAD over 80db just because cheap DAC's can do 105db? Remember with room noise and distortion even the best speakers are in the 60db range. I think as long as you don't have to pay extra for it, don't worry.

This is an interesting point. Hasn't @amirm noted elsewhere that noise and distortion are cumulative in the system chain? If so, wouldn't it still be important to try to get each component to the lowest SINAD of which that type of component is capable?

 

[Ron Texas]

This is an interesting point. Hasn't @amirm noted elsewhere that noise and distortion are cumulative in the system chain? If so, wouldn't it still be important to try to get each component to the lowest SINAD of which that type of component is capable?

It is cumulative, but the question is how good does it have to be? However, Class D amps with decent power and extremely good SINAD are under $1000 now, and transparent DAC's are cheap.

 

[Eirikur]

https://www.axiomaudio.com/blog/distortion

"Two selections of rock/pop music of limited dynamic range typical of modern-day rock recordings were selected from Phil Collins's Hits and Barenaked Ladies' Gordon albums. The limited dynamic range of +/- 5 dB allowed us to introduce the noise without having to follow the music level up and down throughout the song."

Brilliant!
You might as well listen to the Barenaked Ladies Ballad of Gordon on youtube then (pay attention to the particularly nice middle piece @ 0:32).

 

[Julf]

This is an interesting point. Hasn't @amirm noted elsewhere that noise and distortion are cumulative in the system chain? If so, wouldn't it still be important to try to get each component to the lowest SINAD of which that type of component is capable?

It is cumulative but not linear. If one component has significantly lower noise and distortion than the other, the latter one dominates completely.

 

[Julf]

It's not an exact correlation, but if the sub is low distortion then the better FR and decay will result from the effort to control cone velocity, excursion and hysteresis, and particularly voice coil temperature.

Maybe, maybe not. If a more linear suspension or voice coil doesn't necessarily imply better FR or decay.

 

[Wombat]

It's not an exact correlation, but if the sub is low distortion then the better FR and decay will result from the effort to control cone velocity, excursion and hysteresis, and particularly voice coil temperature.

In pro audio, maybe. In home systems, not so much.

 

[Frank Dernie]

I can easily believe people can't tell the difference between distortion and a slight change to the genuine timbre defining harmonics of bass notes, like Sal's bass.
All the components of any harmonic distortion in the bass will be in the audible frequency range whereas with harmonic distortion at 10kHz none will be.

 

[Sergei]

So why do we chase low dist woofers and servo control etc,, if we cant hear it anyway..?!?

The article illustrates the following:

(A) Typical pop music, if played loud enough, sounds pretty much the same regardless of how high the distortions are. It evolved to be that way, so that anyone can listen to it anywhere at any time. Some other genres are very different in this regard - even slightest distortions take their magic away.

(B) Sound objects localization effects are not always taken into account by experiments like that one. For instance, a highly-distorting subwoofer may become locatable in an unwelcome way during "ominous rumble" passages in movies.

A highly-distorting subwoofer may generate hilarious illusions sometimes. For instance, bass guitar player may be perceived as shorter than he or she actually is, because the sound source perceptual location could be shifted downward.

To illustrate: let's say a bass guitar plays open A string at 55 Hz, on a system which includes a subwoofer crossed-over at 80 Hz with 24 dB/octave filter. A low-distorting subwoofer would reproduce the fundamental, some of the second harmonic, and a little bit of the third; the rest of harmonics would not be audibly reproduced.

A highly-distorting subwoofer would add tons of harmonics at 110 Hz and above, which could be readily detected via the shift of the perceived guitar origin closer to where the subwoofer is. Since the effect is frequency-dependent, the bass guitar may appear jumping around or stretching and contracting while playing.

So, one of the reasons for buying low-distorting servo subwoofers is preservation of the bass guitarists dignity. Without good subwoofers, they may audibly appear to be short fat gnomes wildly jumping around the stage, while playing an instrument inexplicably made of rubber

 

[Julf]

Without good subwoofers, they may audibly appear to be short fat gnomes wildly jumping around the stage, while playing an instrument inexplicably made of rubber

I think you justs described the Eurovision Song Contest...