Importance of frequency response to tone???

[SPOautos]

Can you guys give me some education on frequency response?

I read that for a single sound like a note we cant really hear above 20,000Hz but that there are secondary frequencies that create and shape tone and that these go well above 20,000Hz. They basically (from how I understand it, or maybe dont hahaha) arent heard as a single tone at a frequency over 20000Hz but that they shape the tone and sound that we do hear. That got me looking at things like equipment known for excellent sound like Sansui G series receivers and their frequency response goes MUCH higher that 20,000Hz. Even some speakers like high end Focal go up to 40,000Hz. If we cant hear over 20,000Hz they why make equipment that goes beyond that. SO *IF* this is true (big if because I dont know what I'm talking about)...does that mean I need to be looking for equipment with higher frequency responses to have more tone? A lot of equipment stops at 20000Hz so I'm wondering if this plays a role in its lack of clarity and realizm and tone...just generally not sounding as good.

Thank you for any education....and as you can tell Im no engineer so no need to super engineering talk....layman's terms and general concepts is good.

 

[RunOutGroove]

I'm not an engineer... so I can't speak with any authority. I did watch something on the web a while ago that sounded pretty convincing to me where what was being said was that those higher frequencies even if we can't hear them are perceived in other ways and can add to the realism of music. The theory is that sounds that are very high frequency aren't necessarily even perceived at all unless they're overtones of sounds that we hear. That combination of the sound that we hear with the really high overtones that we perceived but can really register with our ears is the key to it all... Supposedly.

 

[SPOautos]

I'm not an engineer... so I can't speak with any authority. I did watch something on the web a while ago that sounded pretty convincing to me where what was being said was that those higher frequencies even if we can't hear them are perceived in other ways and can add to the realism of music. The theory is that sounds that are very high frequency aren't necessarily even perceived at all unless they're overtones of sounds that we hear. That combination of the sound that we hear with the really high overtones that we perceived but can really register with our ears is the key to it all... Supposedly.

Yeah, that's the jist of how I understood it. In more detail is sounded like what we hear, a trumpet note for instance, is considered 1 frequency but that its actually comprised of several frequencies and those 'sub frequencies' (that's not the actual name but just my description) give tone and detail. But if the frequency response of the equipment doesnt go high enough, they are lost. This has me thinking more about the freq resp of everything from the source like turntable cartridge or cd player all the way to the speaker.

It's been interesting to look at the freq response of certain equipment people percieve as sounding great compared to mediocre equipment.

But I'm not any kind of expert in audio or electronics, just a newbie doing some research.

 

[stereofanboy]

I'm under the impression that the overtone and undertone patterns are what makes a trumpet sound like a trumpet rather than like a flute playing same note.

I've thought that for a long time and don't recall where I picked it up. Probably high school band class.

 

[rothwellaudio]

SO *IF* this is true (big if because I dont know what I'm talking about)...does that mean I need to be looking for equipment with higher frequency responses to have more tone?

More tone? That implies that all equipment has some tone, but better equipment has more of it. Doesn't make sense to me.
It isn't difficult to make electronics with a much wider bandwidth than 20Hz-20kHz, but I've never seen any evidence - either empirical or theoretical - that a wider bandwidth produces better sound. In fact, the opposite approach is often advocated - filter out the stuff outside the 20Hz-20kHz bandwidth to prevent it from from having nasty side-effects in the audio band.

 

[DaveVoorhis]

I read that for a single sound like a note we cant really hear above 20,000Hz but that there are secondary frequencies that create and shape tone and that these go well above 20,000Hz. They basically (from how I understand it, or maybe dont hahaha) arent heard as a single tone at a frequency over 20000Hz but that they shape the tone and sound that we do hear.

Every tone that isn't a pure sine wave consists of a series of harmonics, i.e., sine waves of different frequencies mixed together along with the so-called fundamental, i.e., the frequency of the note being played. This mix of frequencies or harmonics creates the waveshape or characteristic timbre -- along with the way the wave amplitude changes over time, called the waveform envelope -- that you recognise as a trumpet or other instrument or voice.

Each harmonic in the mix of harmonics has its own amplitude, which is sometimes described (in this context) as energy level or power of the given harmonic. The energy level of harmonics outside of roughly 20hz - 20,000hz is relatively small. Yes, harmonics above 20,000hz do contribute to the wave shape below 20,000hz, but only negligibly. To what degree they're audible -- i.e., to what extent you'd hear the difference between two otherwise-identical amplifiers where one has a bandwidth of DC - 100,000hz and the other has a bandwidth of 20hz - 20,000hz (or in other words, to what degree we can hear slight differences in waveshape due to supersonic harmonics) is to my knowledge not definitively known.

 

[Guest159]

My amp has generated a good bit of discussion about whether things going on beyond 20KHz make a difference. Incidentally, I'm listening to Focal speakers, but not with Be tweeters. The amp is a Pass F5 with a flat bandwidth of 0Hz to 1,000,000Hz. No two ways about it; it's an extremely detailed amp. Is that kind of bandwidth why it's as detailed as it is? It's a subject more of speculation than fact based debate in everything I've read. Nobody has ever done measurements of any sort and been able to point to something that makes an argument for it being the case I've ever seen. Certainly it's nice that you'll never come across anything that comes close to challenging it's bandwidth, but there are practical problems with 1MHz amps which is why nobody makes them. For instance, some interconnect cables can be problematic.

In playing with it's distortion character, it's tonal quality comes from it's distortion. If you like warm and wooly and euphonically rich, you can do that with the F5. If you want it clean, tight, and sterile, you can do that too. I can't say I've ever heard anything in it's sound that suggests it's crazy bandwidth is why it sounds as good as it does. All I know is that I'm not physically able to hear most of what it can do.

 

[DaveVoorhis]

... No two ways about it; it's an extremely detailed amp. ...

I do wonder what the difference is -- strictly in terms of the input to the amp vs the acoustic output from a given set of speakers connected to that amp -- between an extremely detailed amp and a non- extremely detailed amp.

 

[twiiii]

The sounds we hear are not simple tones. The sounds we hear are the additions and differences between the tones. The question is how much signal to noise and freedom from distortion is necessary and along with frequency response needed to keep the tones in the original form that were originally being reproduced. We know what happens and what we hear when plying analog tapes at 3 3/4 ips, versus, 7 1/2, versus 15 ips. Each step is an improvement in sound quality. But is it all about frequency response and of course the answer is no. With higher speeds you get more particles of oxide on the tape going over the heads giving more frequency response, low er distortion and in some cases more signal to noise because the dynamic range has become wider. At what point do the diminishing returns make moving the tape faster unnecessary. Some consumers would say 7 1/2 ips. Cassette owners with elaborate noise reduction circuits say 1 7/8 is plenty. If you have ever heard 15 ips, 1/2 inch 2 track masters, you know we consumers are missing the boat. So its not all about frequency response. Mathematically its been proven all you need is 22,000 hz band width. Personally I prefer a little more . But its not so much about the response but the freedom from noise and distortion at what ever bandwidth that influences the sound, too. I would much prefer to have a system capable of reproducing 90 db dynamic range than one that was limited 55 or less that we experienced back in the 60's. Though with careful attention to detail the final result in capturing the sound in the 60's can still be very enjoyable today. So don't limit yourself to only one parameter. Its like the guys that walk in and want a 1000 watt car stereo or 1000 watt speaker for their home. Its not just about watts, or cycles , or db, or timing errors, or distortion . Its all those things. Why have speakers that can produce 40,000 hz. Most likely because by the time you get back down to 20,000 hz, they now have a more linear movement, without resonating or cone break up, the dispersion over the frequency we can hear is more uniform not to cause attention any one small group of frequencies and the distortion is much lower. Why have speakers that can produce 16 hz. When we know the ear isn't cap able of hearing that low till the level becomes very loud. Its the same old story when is enough enough and when is it to much?

Then there is the psychology of sound. Ever walk in and turn your system on one day and it sounds fine, go away for an hour and return and swear there is something wrong. But yet for the next 20 times you listen it sounds fine. Yes our mind can change the experience of listening to sound depending on other factors that influence our daily lives. There is a rule out there. If you are happy with your sound system, very happy in fact, don't go out and audition other systems. Because once you hear another system that is perceived to sound better than yours you can never go back and be happy with what you have. It doesn't matter if the amps cost more or less or the speakers are bigger or smaller, have higher or lower distortion. Its the end result. Then you buy all this stuff and drag it home and find out your old system is sounding marvelous, and the new stuff is less so. I have seen it happen many times. Being in the business was quite a revelation. Whats your thrill. Do the 1/4 mile in under 10 seconds. Being able to follow the curves of a twisty road at twice the speed limit. Be able to drive 500 miles in less than 6 hours in absolute comfort. Same goes for your high fi system. Its not about quantity for me, but quality. How smooth and enjoyable is the ride and can I listen to the system all day without distress. Can I enjoy the view out the windshield picturing the images of the musicians playing the music I am hearing. Or is the image thru that windshield colored all wrong with ripples and discontinuities? If the highs are blue in color and the bass is red and the middle green is the over all balance white or does the picture have a blue or red tint, with freedom from hot spots ripples and other distortions. A slight color shift we human can handle. We go from in door lighting, to theatre lighting to outdoor sun with out even thinking about it. Its like that with sound. So some of prefer a blue tint while others a red, big deal. Some of us like a brighter presentation, others more muted. Its about balance and freedom from distortions. Why by amps that are all about being capable of producing great blues, does that mean they can automatically produce great reds, far from it. Do they get the balance correct and produce a white, or does violet sneak in to the picture?

 

[Grenadeslio]

If an amplifier gives a rating like 10 - 20,000hz there's not a brick wall that limits the amp to this range. This is just "the range" the amplifier is rated to meet it's stated specs within, nothing more. Using this to select an amplifier has to be about the most ridiculous idea I've heard. Hell, I have amps with frequency response numbers like 5 - 100,000hz, going with this theory they should be my best amps, they are not.

So, unless the amplifier manufacturer states frequency range specs like 100 - 10,000hz I wouldn't sweat the petty stuff but pet the sweaty stuff eh?

 

[Guest159]

I do wonder what the difference is -- strictly in terms of the input to the amp vs the acoustic output from a given set of speakers connected to that amp -- between an extremely detailed amp and a non- extremely detailed amp.

The best way I can think of to test that is with an F5. It's very simple to chop off it's top end. Just add some caps in the feedback loops. It doesn't change the distortion character.

 

[nj pheonix]

Clearly only observed, so perhaps I'm interpreting my observation wrong.
I've heard more pedestrian tube gear with certainly less bandwidth, present more naturally than some more refined(? For lack of a better word) SS amps rated from DC to 60khz and other bandwidths).
So solely based on this , tone isn't largely determined by FR (perhaps poorly phrased. Intent FR not the dominant reason for accuracy of tone)
A couple of qualifiers in the statement.
I believe accurate reproduction requires some reasonable quality frequency response.
Any amp with a F R rating doesn't stop producing tones beneath or above the rating . They generally roll off slowly.
There is a lot more to accurate reproduction than FR it is only one component to the equation. I wouldn't fixate on it when trying to choose gear.

 

[Guest159]

Rich tonality usually doesn't arise from either frequency response or high accuracy. It usually arises from rather gratuitous distortion. There are very expensive amps out there making 2-3% distortion that make voices and instruments sound absolutely magical.

Distortion isn't as bad as most people presume, and it doesn't sound like a lot of people think it does. When you inject high levels of 2nd, 3rd, and 4th order distortion into a signal, it adds a lot of richness, warmth, and dimensionality. If the even orders are positive phase, the give a little more punch and bite to some sounds like guitar strings and cymbals. If it's negative phase it tends to lay the sound back and expand it around you.

Most amps that do that well have acceptable bandwidth, but nothing very impressive. A gain device needs to be pretty nonlinear throughout it's power band to get that kind of euphonic distortion and that usually means it's useful bandwidth will drop off sharply at the top and bottom.