Jump to content

Does 'hi fi' always have to be 20 Hz to 20 Khz in audio bandwidth?


Recommended Posts

G'day all, I ask that because a lot of program sources that I frequently listen to, whilst of good audio bandwidth, are generally not 20 Hz to 20 KHz in audio bandwidth, yet still sound great!  I appreciate that reasonable HF extension does sound good but is a response to 20 KHz (on the high side) really necessary to be 'hi fi'?  Regards, Felix. 

Edited by catman
  • Like 1
Link to comment
Share on other sites



short answer ...  no

 

eg. most of the classic tube amps of the 70s, 80s ( arc, cj, jadis et al ) barely got beyond 15khz

the famous quad esl57's bandwidth is well within 20-20.

 

no one ever said they were not hi-fi.

 

the point of going beyond 20-20 is that what happens outside this envelope affects the linearity of the 20-20 band.

 

 

 

 

 

  • Like 3
Link to comment
Share on other sites

  • Volunteer
1 hour ago, michaelw said:

 

the point of going beyond 20-20 is that what happens outside this envelope affects the linearity of the 20-20 band.

 

Interesting. So even though I may only hear up to 15kHz, the sound I can hear is affected by the sound that's above my hearing range ?

thats a very good point :thumb:

Link to comment
Share on other sites

26 minutes ago, Sir Sanders Zingmore said:

 

Interesting. So even though I may only hear up to 15kHz, the sound I can hear is affected by the sound that's above my hearing range ?

thats a very good point :thumb:

 

Above AND below. As a very broad generalisation, (assuming a 6dB/octave roll-off) it is desirable for an audio device to possess a frequency response (+/-3dB) at approximately 10 times the highest and lowest frequencies of interest. That means; 2Hz ~ 200kHz is the desired aim. The reason is to ensure that phase shift errors are kept to a minimum. Digital and fast foll-off systems are different and the above truism is not necessarily required. Most 16/44 digital systems, for instance, employs a 'brickwall' fiter system, which can maintain excellent phase response all the way up to around 20kHz. Obviously speakers cannot deliver +/-3dB response down to 2Hz. 

Link to comment
Share on other sites



For me, as long as it is a fairly wide bandwidth, then it is how well it handles itself within that bandwidth that counts.  

 

To explain, I believe the weakest point is probably the speakers.  I love listening to fullrange single driver based systems, where the top end is usually rolling off at 17 or 18kHz.  However, the smooth high frequencies from these systems sound better to me than the more brittle sound of a small dome tweeter (for example), even when I boost the highs in the amplification to compensate for the roll-off.

Link to comment
Share on other sites

G'day all, interesting stuff.  The question of the audibility of possible harmonics/response on either side of the audible spectrum is contentious.  I recall that years ago that NAD made a point/feature of not broadening the audio bandwidth of their amplifiers excessively for good technical reasons (filtering out unwanted rubbish outside of the audio spectrum). 

 

To be honest I have mixed feelings about all of this.  It is worth considering live unplugged acoustic and vocal music as a comparison of real music spectra.  Interesting indeed.  Regards, Felix.

Link to comment
Share on other sites

This topic is also affected by each persons hearing ability as most middle aged will have lost the higher frequencies and it has been covered in other SNA discussions like 

 

  • Like 1
Link to comment
Share on other sites



3 minutes ago, davewantsmoore said:

 

Typical content doesn't contain very much (if anything) in the top octave.     When it does, then this can be a relevant factor.

Sorry to digress.

 

In terms of hearing ability at some late stage in most audiophiles lives whether that be for some right now with premature hearing loss in their mid age or beyond 70years the hearing can be dramatically reduced and/or may only hear down to around 8khz, and around there higher freq hearing is likely to be noticeabally affected with the music content.

 

No disrespect to our fellow SNAers who might have this issue already. Some have already posted about this and I myself have bought gear from such people quitting their audio gear because of this.

 

Also, it has been claimed on many occasions that hearing acuity at age some how makes up for this, but I could not find any medical hearing Health articles that support it.

 

Link to comment
Share on other sites

Guest m_james
12 minutes ago, Al.M said:

Sorry to digress.

 

In terms of hearing ability at some late stage in most audiophiles lives whether that be for some right now with premature hearing loss in their mid age or beyond 70years the hearing can be dramatically reduced and/or may only hear down to around 8khz, and around there higher freq hearing is likely to be noticeabally affected with the music content.

 

No disrespect to our fellow SNAers who might have this issue already. Some have already posted about this and I myself have bought gear from such people quitting their audio gear because of this.

 

Also, it has been claimed on many occasions that hearing acuity at age some how makes up for this, but I could not find any medical hearing Health articles that support it.

 

 

It's a really good point. One that is often ignored.

 

Could this also be part of the reason why young people often prefer a bassier or less trebly sound in their music and stereo equipment?

Link to comment
Share on other sites

3 hours ago, Zaphod Beeblebrox said:

 

Above AND below. As a very broad generalisation, (assuming a 6dB/octave roll-off) it is desirable for an audio device to possess a frequency response (+/-3dB) at approximately 10 times the highest and lowest frequencies of interest. That means; 2Hz ~ 200kHz is the desired aim. The reason is to ensure that phase shift errors are kept to a minimum. Digital and fast foll-off systems are different and the above truism is not necessarily required. Most 16/44 digital systems, for instance, employs a 'brickwall' fiter system, which can maintain excellent phase response all the way up to around 20kHz. Obviously speakers cannot deliver +/-3dB response down to 2Hz. 

Interesting question and statements. What frequencies do you use to set up your soundstage and adjust it? What frequencies create a 3d image with space and air? What frequencies create the harmonics difference between Hi pitched Brass, Wind/wood and string instruments? If someone had an understandable answer to these questions, I would listen. The Brain is an amazing thing. I feel at 62 I can hear the same as when I was 32.I know my ear ole if tested would prove me wrong.

Emotion, what frequency is that? It must be well outside 20-20,000 as I cant hear it in CD the way I hear it in Vinyl. (I did not use the analog/ Digital comparison).

Edited by Wimbo
  • Like 1
Link to comment
Share on other sites

I'm 62, had hearing tested last year, typical loss of highs. I am a carpenter. Thing is you don't notice it, if you boost the highs to compensate it sounds too bright. BTW sweetest midrange I ever heard was a jukebox playing thru stage coloms in a small bar(4x6inches x2) go figure.

Sent from my ASUS MeMO Pad 7 using Tapatalk

  • Like 1
Link to comment
Share on other sites



1 hour ago, Wimbo said:

Interesting question and statements. What frequencies do you use to set up your soundstage and adjust it?

 

All of them.

 

1 hour ago, Wimbo said:

 

What frequencies create a 3d image with space and air?

 

All of them.

 

1 hour ago, Wimbo said:

 

 

What frequencies create the harmonics difference between Hi pitched Brass, Wind/wood and string instruments?

 

All of them.

 

1 hour ago, Wimbo said:

 

If someone had an understandable answer to these questions, I would listen.

 

Perhaps you should wait for the answer. 

 

1 hour ago, Wimbo said:

 

 

The Brain is an amazing thing. I feel at 62 I can hear the same as when I was 32.I know my ear ole if tested would prove me wrong.

 

I'm 63 and I know that I can't hear stuff I heard in my 20s. I am not amazed, just sad that I can't hear the stuff I once could. 

 

1 hour ago, Wimbo said:

Emotion, what frequency is that? It must be well outside 20-20,000 as I cant hear it in CD the way I hear it in Vinyl. (I did not use the analog/ Digital comparison).

 

Emotion is created by the performers, not the reproduction system. 

  • Like 4
Link to comment
Share on other sites

From the OP, speaker manufacturers shouldn't have to consider the average hearing loss in the population because that is out of their control so the 20 - 20 is relevant. In practice, however the average customer that can afford larger amounts of audio tends to be older and more hearing loss associated.

 

Subjectively, when considering speaker specs in a product and depending on what the product is I feel comfortable that it claims about:

 

Conventional box speaker-

* Large full range three way design 30-40 to 17000hz

 

* medium size standmount two way 50-17000hz

 

* small monitor 70 - 17000hz, which needs sub asssitance

 

Full range dual cone type speaker (Lowther, Fostex etc) 70 - 15000hz, which needs sub assistance

 

Panel speaker (ESL, ribbon) 50 - 12000hz (e.g. Martin Logan, Ambience, Magnepan, Acoustat, Quad 57) and often needing sub assistance.

 

Whether those specs above are actually being achieved or can I hear that in my listening room acoustics is another matter. 

 

It depends what floats your boat the most as in the above types of speakers, IMO the panels can be most satisfying despite their lack of the 20 - 20 specs. The midrange clarity, imaging and coherence of panels can outweigh a large full range box speaker. Whereas for others it will be the opposite.

 

 

Edited by Al.M
Link to comment
Share on other sites

G'day all, well just on harmonics and such please consider the question of RIAA equalisation that calls for an 'ongoing' 6 db/octave roll off starting at around 2.1 KHz.  Are harmonics beyond the audio spectrum relevant here?  Also, many popular modern op amps don't go too far beyond 20 KHz in absolute terms before running out of steam.  In that respect 'discrete' devices fare far better, but in all honesty does it really matter assuming a linear 20 KHz audio bandwidth in any case?  Regards, Felix.   

Link to comment
Share on other sites

3 hours ago, muskrat1954 said:

I'm 62, had hearing tested last year, typical loss of highs. I am a carpenter. Thing is you don't notice it, if you boost the highs to compensate it sounds too bright. BTW sweetest midrange I ever heard was a jukebox playing thru stage coloms in a small bar(4x6inches x2) go figure.

Sent from my ASUS MeMO Pad 7 using Tapatalk
 

 

It's possible the frequency range of what you were listening to at age 32 may still be within the range of what your hearing at 62 years. For example, a lot of music both then and now contains information mostly between 20 - 10000hz, so at age 62 the average male can hear around 12hkz so no problem.

 

The other things to note with normal age related usable hearing loss is not only a reduction in high frequency, but about a 20-40dB reduction overall loss and differences between left and right ears. So the brain and/or hearing organ can normally adjust for that range of reduction with its inbuilt loudness adjustment mechanism, but less so at higher rates of dB loss, where an audiologist will start to declare you are partly deaf and need hearing aids. So at my age 53 I follow the normal population statistical high freq reduction around 14khz and also have about 25dB overall loss. The overall loss if not severe can be adjusted at the amplifier volume leve to how you like it but will not fix the high frequency loss in a normal listening room. I can still enjoy listening to hi fi.

 

Also, note the difference between males and females who have significantly better hearing at age. 

 

And another issue about left and right ear dominance and hearing differences:

http://www.soundtherapyperth.com/overview/rightear.php

 

In the animal kingdom Homo sapiens pretty much suck at hearing as we have evolved from the trees for vocalisation, best around the mid frequencies and limited to only 20 - 20000hz, with an odd shaped A-weighted hearing response (we don't hear flat response, we hear reduced lows, pronounced mids and reduced highs), whereas dogs, bats and dolphins can hear much better. Dolphins up to 150hkz, so the audio industry should be doing stem cell research on dolphin gene cure for audiophiles if they want to sell more stuff.

 

So apart from technical truths and our own personal tastes in audio, you can see some additional physiological reasons why on these forums it's so difficult to reach agreement on what sounds good.

 

Edited by Al.M
Link to comment
Share on other sites

4 hours ago, catman said:

G'day all, well just on harmonics and such please consider the question of RIAA equalisation that calls for an 'ongoing' 6 db/octave roll off starting at around 2.1 KHz.  Are harmonics beyond the audio spectrum relevant here?  Also, many popular modern op amps don't go too far beyond 20 KHz in absolute terms before running out of steam.  In that respect 'discrete' devices fare far better, but in all honesty does it really matter assuming a linear 20 KHz audio bandwidth in any case?  Regards, Felix.   

 

Despite the negative vibes coming out of WA, Felix, I suggest you - as a DIY-competent fellow - should do your own experiment.  :)

 

Rig up a line level passive LP filter for each channel of your power amp - I'm sure you know the formula.  Select the component values to give you 'knee frequencies' (ie. the -3dB point) of:

  • 100Khz
  • 120Khz
  • 140Khz
  • 160Khz
  • 200Khz.

Now, a 6dB LP filter with a -3dB point of 'Xhz' has a hearable effect a couple of octaves down from 'Xhz'.

 

But I suggest when playing vinyl (which doesn't have a brick-wall filter at 20Khz), you will be able to hear the effect of inserting a 150Khz LP filter.  Possibly even 200Khz (I haven't tried that).  If so, you've proved to yourself that you can actually pick up stuff way beyond 20Khz.

 

EDIT:  This assumes your power amp has a wide enough bandwidth.  People with cheap tube amps (which use cheap output transformers) won't get anything out of these tests as their bandwidth is probably well down by 30Khz.

 

4 hours ago, Zaphod Beeblebrox said:

 

Emotion is created by the performers, not the reproduction system. 

 

 

Sure, Trevor - emotion is created by the performer(s).

 

But some components let this emotion through better than others.  :thumb:

 

Andy

 

Edited by andyr
  • Like 1
Link to comment
Share on other sites



Short answer - yes.

 

And the hearing limitation of specific individuals is utterly irrelevant. Including yours, Felix, since I assume I correctly read your question as a general one.

 

Beyond 20-20k is a completely different matter. Preserving inaudible energy at either end can actually be harmful, and trying to reproduce it can raise challenges that the playback system might not meet. Filtering out is not harmful unless the filter reaches below 20k (or above 20 Hz) in a harmful way.

 

Harmonics reaching beyond 20k don't need to be preserved. Same for fundamentals.

Link to comment
Share on other sites

Surely emotion is also created by the listener? We all have our own individual emotional responses to certain pieces of music.

 

On the subject of hearing loss, I'm quite certain that I have had some high frequency hearing loss in my right ear for a long time now, noticeably more so than in my left. I remember once going to a hi-fi show where they had a hearing test booth. It was quite simple, you put on a pair of headphones, then there was a frequency sweep from low to high in each channel individually, and you pressed a button when you could no longer hear the tone. I can't remember what my actual high frequency cut-offs were, but it was definitely lower in my right ear compared to my left. That was when I was in my early 20s, I am in my early 60s now. I don't know if that comes from going to too many loud concerts in my youth, or maybe its always been like that. In any case, it doesn't really bother me, and certainly doesn't interfere with my enjoyment of music.

 

Another factor that nobody appears to have mentioned yet is the dynamic range of a recording. Not every recording will use the full frequency range, but may still have good dynamics and sound great.

 

 

 

 

Link to comment
Share on other sites

2 hours ago, Zaphod Beeblebrox said:

 

All of them.

 

 

All of them.

 

 

All of them.

 

 

Perhaps you should wait for the answer. 

 

 

I'm 63 and I know that I can't hear stuff I heard in my 20s. I am not amazed, just sad that I can't hear the stuff I once could. 

 

 

Emotion is created by the performers, not the reproduction system. 

No answers here. You need information on the soundstage to get everything right. I once did an A/B between an Ipod and a CD player. Both the customer and myself heard a lot of things missing on the Ipod that the CD had. This was using a K.D. Lang album. You have heard the phrase,"Flat and two dimensional." So just plonk the speakers down in an approximate position and that will do. No need to set up, it all sounds the same. With high Res gear, the story is a lot different. Hi Rez-more audilble info- more Frequencies. Specific frequencies make three dimensionality "Air" and "Focus".

Link to comment
Share on other sites

35 minutes ago, Wimbo said:

No answers here.

 

The answers related to the banality of the questions posed. Here is what is required to preserve accurate sound staging is a reproduction system (in approximate order of importance):

 

* The room

* The room

* The room

* The phase accuracy of the speaker system.

* The phase accuracy of everything else in the system.

* Correct input/output impedance matching (minimum 100:1).

* Correct setup of LP playback if used.

 

The frequency response, per se, is not a particularly important parameter. hence my response to your question.

 

35 minutes ago, Wimbo said:

 

You need information on the soundstage to get everything right.

 

Of course. 

 

35 minutes ago, Wimbo said:

 

I once did an A/B between an Ipod and a CD player. Both the customer and myself heard a lot of things missing on the Ipod that the CD had.

 

Of course. An iPod is not hi fi. It is, at best, medium fi. Way too much missing. 

 

35 minutes ago, Wimbo said:

 

This was using a K.D. Lang album. You have heard the phrase,"Flat and two dimensional." So just plonk the speakers down in an approximate position and that will do. No need to set up, it all sounds the same. With high Res gear, the story is a lot different. Hi Rez-more audilble info- more Frequencies. Specific frequencies make three dimensionality "Air" and "Focus".

 

It is less to do with frequency response, per se, and MUCH more to do with phase relationships. 

 

And again: It is the performers who insert emotion into music. NOT reproduction equipment. 

Link to comment
Share on other sites

"It is less to do with frequency response, per se, and MUCH more to do with phase relationships."

Confusing.

 

 

"And again: It is the performers who insert emotion into music. NOT reproduction equipment. "

Dont agree with this. The reproduction Equipment must have the ability to reproduce the emotion of the performer.

Original CD players couldnt do this as their reproduction of low level detail dynamics was very low.

  •  
  • Like 1
Link to comment
Share on other sites



  • Recently Browsing   0 members

    • No registered users viewing this page.




×
×
  • Create New...
To Top