Head-Fi.org › Forums › Equipment Forums › Sound Science › What's the highest general frequency music stops at?
New Posts  All Forums:Forum Nav:

What's the highest general frequency music stops at? - Page 6

post #76 of 125
Quote:
Originally Posted by bigshot View Post


I use the equalizers in Peak.
I'm really not arguing audibility. I don't know why it keeps coming back to that. My point is that the top octave isn't a sliver of sound that makes a whole lot of difference to music compared to the rest of the audible range.
If I was EQing, and a core frequency band was out by 2 dB, that would bother me much more than the top octave being out by 6 dB. *Relative importance*"

"I use the equalizers in Peak."

Like I said, that's not enough description for a valid discussion.  We need to know Q (bandwidth), Gain/Loss, and center frequency.

 

"If I was EQing, and a core frequency band was out by 2 dB, that would bother me much more than the top octave being out by 6 dB. *Relative importance*""

 

Not sure what you mean "core frequencies", probably "important ones"?  But my point is if you EQ down 20KHz with your peak/dip, you can, and likely will, affect 10KHz as well.  It's not that the 20KHz matters all that much, it's what effect that mechanism has on the rest of the spectrum.  In my example a 6dB drop at 20KHz would result in 2.4dB loss at 10KHz. I guess if you consider 10KHz a "core frequency", then that might bother you.

post #77 of 125
Quote:
Originally Posted by bigshot View Post


Except of course for church organs, which have fundamentals beyond the abiity of many very good headphones to reproduce accurately.
By the way, the BIG difference between good headphones and mediocre ones are how they handle the fundamentals within the core frequencies. You definitely do need to talk about fundamentals because that is most of what you hear. Again, that's my point- a very basic concept which no one seems able to wrap their heads around.

It might be helpful if you define the term "core frequency" for the group.  It's not a commonly used term, and might be what nobody seems able to wrap their heads around it.

post #78 of 125
Quote:
Originally Posted by chewy4 View Post

I'm pretty sure there aren't any remotely decent headphones that drop the fundamental frequencies completely.

You keep arguing absolutes. I said the core frequencies are the most important in the spectrum, and imbalances there are MUCH more important than imbalances at the upper edge of human hearing.

To summarize my points for you... The top octave is the least important, because:

1) 10kHz to 20 kHz is only one octave out of nearly 11 octaves in the range of human hearing. It isn't a very wide band of sound.

2) Many people can't even hear much above halfway through that octave anyway, and even if they can, it's mostly sound pressure and not a musical note.

3) There are no fundamentals from musical instruments in that range. Every other octave contains fundamentals from one instrument or other.

4) The first three levels of harmonics (the most important in reproduction of music) of most instruments aren't even in that range.

5) The only musical instruments with important harmonics in that range are cymbals and triangles, which are primarily percussive.

A good way to tell for yourself what this top octave sounds like is to use an equalizer to isolate it. If you do that, you will clearly see that it's the least important octave in the range of human hearing. Number two least important is the bottom octave. The closer you get to the middle of the range, the more important the sound is to good sound quality.
post #79 of 125
Quote:
Originally Posted by stv014 View Post

There are many (especially closed) headphones with problems in the 1 to 5 kHz range, which is what human hearing is most sensitive to. Although this range is somewhat higher than (but overlaps with) fundamental frequencies.

Agreed that the 1KHz to 5KHz range is critical.

 

There seems to be a slight misapplication of the therm "fundamental" in the thread.  In the above, the term seems to refer to a range of frequencies below 1KHz.  

 

A fundamental is not in a range of frequencies.  "A fundamental is defined as the lowest frequency of a periodic waveform."  There's no frequency specific requirements for something to be a fundamental, only that it's the lowest frequency component within the waveform.  The rest of the frequency content in a wave, harmonically related or not, give it its character, makes a voice a voice, a flute a flute, etc.  

 

A fundamental can be at any frequency, even frequencies outside the audible spectrum.  

 

Reproducing only the fundamental of a complex waveform would result in a sine wave at that frequency.  A pure sine wave is a fundamental only, and has no harmonic content.  It's pointless to talk about the importance of reproducing fundamentals, they're only sine waves until harmonic character is added.  Good reproduction would require the ability to reproduce the entire waveform, fundamentals, harmonics, and non-harmonic content, all without distortion or change in relationship to each other.

post #80 of 125
Quote:
Originally Posted by jaddie View Post

It might be helpful if you define the term "core frequency" for the group.  It's not a commonly used term, and might be what nobody seems able to wrap their heads around it.

The core frequencies are the ones in the middle where the fundamentals for most musical instruments reside- the range of a piano- approximately 30Hz to 4kHz. The most important core frequencies are from about 175Hz to 1200Hz. The key fundamentals in most music lies in that range.
post #81 of 125
Quote:
Originally Posted by jaddie View Post

"I use the equalizers in Peak."
Like I said, that's not enough description for a valid discussion.  We need to know Q (bandwidth), Gain/Loss, and center frequency.

1/3 octave graphic. (Peak is a pro sound software package.)
Edited by bigshot - 11/28/12 at 12:48pm
post #82 of 125
Quote:
Originally Posted by bigshot View Post


You keep arguing absolutes. I said the core frequencies are the most important in the spectrum, and imbalances there are MUCH more important than imbalances at the upper edge of human hearing.
To summarize my points for you... The top octave is the least important, because:
1) 10kHz to 20 kHz is only one octave out of nearly 11 octaves in the range of human hearing. It isn't a very wide band of sound.
2) Many people can't even hear much above halfway through that octave anyway, and even if they can, it's mostly sound pressure and not a musical note.
3) There are no fundamentals from musical instruments in that range. Every other octave contains fundamentals from one instrument or other.
4) The first three levels of harmonics (the most important in reproduction of music) of most instruments aren't even in that range.
5) The only musical instruments with important harmonics in that range are cymbals and triangles, which are primarily percussive.
A good way to tell for yourself what this top octave sounds like is to use an equalizer to isolate it. If you do that, you will clearly see that it's the least important octave in the range of human hearing. Number two least important is the bottom octave. The closer you get to the middle of the range, the more important the sound is to good sound quality.

I'll agree with all of the above.  But just because the top octave is the least important doesn't mean I won't give it equal attention.  Least important or not, I'd rather have it than not have it, even if I personally can only hear about the first 20% of it.  And by the way, I want it flat all the way to 20KHz. But that's just me.

 

I'm surprised none of the 192KHz crowd has popped in here! You know, the DC to Blue Light wide-banders.  Seems like their war, not ours.  

post #83 of 125
Quote:
Originally Posted by bigshot View Post


You keep arguing absolutes. I said the core frequencies are the most important in the spectrum, and imbalances there are MUCH more important than imbalances at the upper edge of human hearing.
To summarize my points for you... The top octave is the least important, because:
1) 10kHz to 20 kHz is only one octave out of nearly 11 octaves in the range of human hearing. It isn't a very wide band of sound.
2) Many people can't even hear much above halfway through that octave anyway, and even if they can, it's mostly sound pressure and not a musical note.
3) There are no fundamentals from musical instruments in that range. Every other octave contains fundamentals from one instrument or other.
4) The first three levels of harmonics (the most important in reproduction of music) of most instruments aren't even in that range.
5) The only musical instruments with important harmonics in that range are cymbals and triangles, which are primarily percussive.
A good way to tell for yourself what this top octave sounds like is to use an equalizer to isolate it. If you do that, you will clearly see that it's the least important octave in the range of human hearing. Number two least important is the bottom octave. The closer you get to the middle of the range, the more important the sound is to good sound quality.

I was never arguing against any of that. I agree with that.

 

I just don't agree with least important equalling unimportant, and find the difference with these frequencies cutoff to be quite HUGE in certain songs. For example the first one I posted where it uses a noise of sorts(pink?) as a type of percussion; this noise completely changes tone when rolling frequencies off of it. And in certain electronic pieces you'll be missing sounds entirely.

 

Does a headphone need to be tuned as precisely at those frequencies? No, but it definitely helps.

post #84 of 125
Quote:
Originally Posted by bigshot View Post


1/3 octave graphic. (Peak is a pro sound software package.)

Ok, 1/3 octave graphic.  Then you're familiar with how the action of one band affects the response in adjacent bands.

post #85 of 125
The thing is, you can't give each band equal importance, because it's the top octave of most headphones that gets all spikey in te response. The reason you need to prioritize is because you can spend an awful lot of money buying headphones that perform a little better way up high, but all that money isn't going to buy you as much sound quality as getting good cans, equalizing what really matters and leaving well enough alone up where it doesn't matter.

Audiophiles love to spend big bucks to shave THD down to tiny fractions, even when it's so far below the threshold of audibility it doesn't matter. They put lossless files on their iPod instead of lossy files that sound exactly the same. They buy SACDs with dynamic ranges they'll never turn the volume up high enough to hear. They worry about frequencies only bats can hear. The fret what they call "the last 1%"... But they buy preamps with no tone controls and refuse to equalize because of some tiny sliver of time they'd never be able to hear in a million years.

The most important thing to take away from "sound science" isn't the proper way to conduct a double blind test. It's how to APPLY measurements to WHAT YOU ACTUALLY HEAR, so you're optimizing things that matter instead of things that don't.

Without a good handle on relative differences, you're a boat adrift in an ocean going wherever fate takes you. You aren't charting a course for getting better sound.
post #86 of 125
Quote:
Originally Posted by bigshot View Post


The core frequencies are the ones in the middle where the fundamentals for most musical instruments reside- the range of a piano- approximately 30Hz to 4kHz. The most important core frequencies are from about 175Hz to 1200Hz. The key fundamentals in most music lies in that range.

Thank you for that.

 

What's interesting is that reproducing the core frequencies, especially those you've defined as "important", has been relatively easy to do for many decades.  Most of the major advances in recording and reproducing technology have historically been toward improving performance outside of the core frequencies.  And they continue to do so.

post #87 of 125
Quote:
Originally Posted by jaddie View Post

What's interesting is that reproducing the core frequencies, especially those you've defined as "important", has been relatively easy to do for many decades.  Most of the major advances in recording and reproducing technology have historically been toward improving performance outside of the core frequencies.  And they continue to do so.

In pro recording, accurate capture of the entire audible range was perfected in 1952 with the advent of high fidelity stereo recording on magnetic tape. In home playback, it's been sliding the other direction, especially in the past twenty years with a variety of colorations to cans and the compromises that come with making speakes smaller and smaller.

I've been working for 25 years on my speaker system. The work I did on EQing the middle frequencies is what sets the sound apart. I can't tell you how many headphones and speakers I've heard with big imbalances and firebreaks in the core frequencies either because of deliberate coloration or because of gaps in the crossover between the mains and subwoofer. Back in the 70s and 80s when we had big cabinet speakers, imbalances were almost always because of the room. Now it's because of deliberate design choices,

While audiophiles spend thousands on things they can't hear, they pay no attention to the important things, because they assume that's taken care of for them. It usually isn't, but until they sit down and start figuring out what each of the frequency bands sound like, they don't realize it.
Edited by bigshot - 11/28/12 at 1:18pm
post #88 of 125
Quote:
Originally Posted by bigshot View Post

The thing is, you can't give each band equal importance, because it's the top octave of most headphones that gets all spikey in te response. The reason you need to prioritize is because you can spend an awful lot of money buying headphones that perform a little better way up high, but all that money isn't going to buy you as much sound quality as getting good cans, equalizing what really matters and leaving well enough alone up where it doesn't matter.
Audiophiles love to spend big bucks to shave THD down to tiny fractions, even when it's so far below the threshold of audibility it doesn't matter. They put lossless files on their iPod instead of lossy files that sound exactly the same. They buy SACDs with dynamic ranges they'll never turn the volume up high enough to hear. They worry about frequencies only bats can hear. The fret what they call "the last 1%"... But they buy preamps with no tone controls and refuse to equalize because of some tiny sliver of time they'd never be able to hear in a million years.
The most important thing to take away from "sound science" isn't the proper way to conduct a double blind test. It's how to APPLY measurements to WHAT YOU ACTUALLY HEAR, so you're optimizing things that matter instead of things that don't.
Without a good handle on relative differences, you're a boat adrift in an ocean going wherever fate takes you. You aren't charting a course for getting better sound.

Man, did you ever speak!

 

+1x10!

 

(I knew we would eventually agree)

post #89 of 125
Quote:
Originally Posted by bigshot View Post


.... Now it's because of deliberate design choices,
While audiophiles spend thousands on things they can't hear, they pay no attention to the important things, because they assume that's taken care of for them. It usually isn't, but until they sit down and start figuring out what each of the frequency bands sound like, they don't realize it.

So, then I can't interest you in a great deal on a platinum over silver plated anti-static, anti-magnetic, gas filled neutron irradiated six foot 10ga power cord?  Not even with the hospital-grade connectors?  Darn. What about if it's braided in corn rows?

post #90 of 125
Does it come in lilac?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › What's the highest general frequency music stops at?