High frequency response, how important is it?

[Xcaliber]

While I was researching and comparing different amps and speakers, there are a lot of attention puts in the high frequency response range in general, let's take the TPA3255 chip amps for an example, the upper top end frequency range have quite a bit of load dependencies in the utmost top region, and a lot of attention put in that aspect, same goes for speaker reviews, some speakers showed either spike, or roll off in the upper region.

Out of curiosity, i went to listen to a few sinewave sweeps to see where my hearing limit is, at the age of 54....it's sad to find out that my hearing isn't what it used to be, I can only hear up to around 13khz. Basically everything above 14khz is becoming completely irrelevant to my relatively old ears.

How sensitive is your hearing? it would be interesting to see where everyone's limit is in relate to age as well.

note: i played back those tones through home speakers and phone, didn't make any difference, my hearing hit the brick wall at 13khz. My 6yr old dog however, she could hear 20khz no problem, her ears perked up every time at 20khz. lol

 

[DVDdoug]

let's take the TPA3255 chip amps for an example, the upper top end frequency range have quite a bit of load dependencies in the utmost top region,

The "theory" is that high-fidelity should reproduce the full audio range.

IMO there's "no excuse" for an amplifier that isn't flat over the "traditional" 20-20kHz range. It's not rocket science, although it's more difficult to get high frequencies with class-D.

The TPA3255 datasheet says it goes up to 100kHz, but there may be compromises in the final circuit design.

Speakers aren't as easy.

On the other side of things... The people who develop lossy compression (MP3, etc.) discovered that even if you can hear "loud" 20kHz tones in a hearing test, you don't (normally) hear the highest frequencies in the context of music where they are weak harmonics (no regular instruments can play a 20kHz "note"), they are masked (drowned out) by other not-so-high frequencies, and our hearing is less sensitive at the frequency extremes. With the standard settings, MP3 usually cuts-off above 16kHz. Other lossy compression formats often go higher. (The main trick to lossy compression is throwing-away frequencies that are masked by other-louder nearby frequencies).

 

[Pinkynator]

Man, this seriously annoyed my cat...

I topped out just slightly above 14 kHz, but it required increasing the amp's volume to what would otherwise be too loud when listening to music. I spent too much time in my youth going to raves with no hearing protection (though a couple of times I had to make earplugs out of tissues).

Anyway... If you can't hear above a certain frequency, what happens if you lowpass it? Do speakers have an easier time? Does it have a negative impact on music that you *can* hear?

 

[Xcaliber]

Anyway... If you can't hear above a certain frequency, what happens if you lowpass it? Do speakers have an easier time? Does it have a negative impact on music that you *can* hear?

That would depend on the order of xover being used, the steeper the slope the less bleed through you get from the frequencies you want to filter out, but in general i think speakers are far more sensitive or strained in the lower frequency range.

 

[Tassin]

Out of curiosity, i went to listen to a few sinewave sweeps to see where my hearing limit is, at the age of 54....it's sad to find out that my hearing isn't what it used to be, I can only hear up around 13khz.

Not too bad, I think. My limit, at the same age, is 13.9-14 kHz. My neighbour just turned 60. I tested his hearing last year: limit ±9.5 kHz.

 

[Pinkynator]

That would depend on the order of xover being used, the steeper the slope the less bleed through you get from the frequencies you want to filter out, but in general i think speakers are far more sensitive or strained in the lower frequency range.

There was a discussion on the Wiim Amp review thread where people complained that things may go berserk after 20 kHz. So is that even an issue if you can't hear it and if you can lowpass it without damaging hearable (that isn't a word, I know) part of the music? And no stress on the tweeters in hypersonic frequencies. If there is a stress. I don't know.

 

[DWPress]

above a certain frequency, what happens if you lowpass it?

I've played with such a filter around 16Khz. I couldn't hear a difference but younger listeners could.

 

[sonitus mirus]

I assumed this test was meant for a normal listening level, which for me drops to inaudible between 12kHz and 13kHz at about 75dB, with approximately 40dB of background noise. As others may have experienced, it's possible to increase the volume in a very quiet environment or use headphones with good isolation to hear higher frequencies. However, this is not recommended as it could jeopardize the safety of your hearing health.

To calculate how many theoretical musical notes exist between 12kHz and 20kHz, I get about 8.84 discrete musical notes in Western music.

n=12×log2(f2/f1)

where:

• n is the number of semitones
• f1 is the lower frequency (12kHz in this case)
• f2 is the higher frequency (20kHz)
• log⁡2 denotes the base-2 logarithm

Not even a full chord on an extra-wide hypothetical piano with 125 keys.

 

[AwesomeSauce2015]

Doing the test, I have a hard time hearing anything above 14khz, with my headphones. However, I happen to know that the old wireless (not bluetooth) headphones I have are basically useless above that.
I do hear differences when I experiment with a "low pass" to the treble. I would say that my hearing is probably in line with what is expected.

But yes, I do believe that having full-range reproduction is important. Ideally, for music, you need 30hz to 20khz. However, if you have a quality speaker (ie, low distortion, and good 100hz-400hz performance), and factor in psychoacoustics, you can "get away with" something around 80hz-16khz.

 

[Audiofire]

30 years old, RME ADI-2 Pro FS R, AKG K702, ambient noise is very low: I detect faint 20 kHz at +22 dBu in Audacity (sine wave set to full amplitude at 1) at around -10 dBFS and easily hear it (faint) at 0 dBFS. Same with 18-19 kHz. 17 kHz detected at around -20 dBFS.

With regard to 20 Hz, I detected it around -45 dBFS.

 

[valerianf]

After hearing tests with my MD, conclusion is that my eating is going up to 12khz included, but not a lot above.
In my youth, I was listening at very high level and going very often to nightclubs that we're not regulated regarding the sound level...

Anyway, I am modifying the crossover of my speakers, if needed, until I am hearing clearly the tweeter.
It is my choice.

 

[sigbergaudio]

Anyway... If you can't hear above a certain frequency, what happens if you lowpass it? Do speakers have an easier time? Does it have a negative impact on music that you *can* hear?

Your speakers won't care much, there is very little energy up high and very little energy required to produce it. A low pass filter won't be brickwall (it will be a slope), so it will affect frequencies below the crossover frequency.

 

[sigbergaudio]

The frequencies north of 15khz don't matter much. When one can barely hear a pure sine 14-15k at loud volume, you have pretty much zero chance of hearing it in music content. That being said, it shouldn't be too much to ask of electronics to be able to reproduce 20-20khz in a linear fashion.

 

[Audiofire]

Effective upper frequency limit of almost all vinyl records is around 17 kHz (except attenuation distortion and noise). Would make sense if someone decided this was scientifically the most feasible option for producing and marketing physical music back then. 17 kHz is the effective upper frequency limit for most of music consumers probably.