Hearing above 20Khz & analog consoles.

[luctellier]

Originally Posted by muziekschuur

No information above 20kHz means no warm open sound

I don't agree with this... IMHO, a well balanced mix can sound open and warm without having any information above 20khz. Warmth is located well below 20khz in my book and I have many example of good open sounding songs that don't have any information above 20khz. We have to be realistic, even if you mix with a 100khz bandwitdh, what do you think will happen when people will be listening to your mix on a 16bits/44.1khz media CD with multimedia speaker or IPOD headphones? Just my 2 cents...

[Ethan Winer]

Originally Posted by muziekschuur

Taken from the D&R website ... We can even notice extremely small amounts of distortion in the order of 0.001% of the nominal signal level ... If this 12kHz sine wave is switched to a square wave we all still can hear the difference between sine and square wave ... THIS IS AMAZING.

Not amazing. Simply a load of uneducated bullshit. How could anyone discern 0.001 percent distortion when most loudspeakers are closer to 1 percent or even more?

--Ethan

[luctellier]

Originally Posted by Ethan Winer

Not amazing. Simply a load of uneducated bullshit. How could anyone discern 0.001 percent distortion when most loudspeakers are closer to 1 percent or even more?

--Ethan

I agree Ethan, this is business bullshit to sell more expensive gear that doesn't really make an audible difference in the end.

[muziekschuur]

But............... has anyone performed the test to see what happens?

[kellyrmartin]

I hear what you guys are saying.... But what about the point about the 12khz sine and 12khz square wave. It makes sense to me... Or is there some missing information in his explanation of why we can hear the difference?

[luctellier]

How can you hear any difference when your monitoring system don't go over 20 000 hz?

[kellyrmartin]

Ok. I just tried the test and I am thinking that the explanation is incorrect. I suspect your explanation of the harmonic content in a square wave is wrong because I can hear a predominat frequency that is alot lower than 12khz. Anyone?

[muziekschuur]

Luc,

Dome tweeters do go beond 20 khz. It's just, the testing equipment at manufacturers usually is OLD. And has no specifications above 20khz. The newer testing equipment (after 1998/2000) does provide information about sound above 20 khz.

Ribbon tweeters do go beond 20khz, and Tannoy sells supertweeters wich are able to extent any monitoring system you may have (up to 50 khz).

Muziekschuur

[dkatz42]

Originally Posted by kellyrmartin

Ok. I just tried the test and I am thinking that the explanation is incorrect. I suspect your explanation of the harmonic content in a square wave is wrong because I can hear a predominat frequency that is alot lower than 12khz. Anyone?

Probably digital aliasing; a pure square wave (or any other repeating waveform) has no frequency content below the fundamental.

The math on square waves is indeed correct; a pure square wave has only the fundamental plus odd harmonics. The leap of faith is that we're hearing the square wave differently from the sine because we are hearing the high harmonics; there are certainly other possibilities (see "aliasing" above.) Probably the neighborhood dogs howling gives it away.

Standing all this on its head, somebody once posted to one of these forums asking what was wrong with his digital rig; when he recorded a 10KHz square wave at 44.1, it played back as a 10KHz sine wave...

[luctellier]

Originally Posted by muziekschuur

Luc,

Dome tweeters do go beond 20 khz. It's just, the testing equipment at manufacturers usually is OLD. And has no specifications above 20khz. The newer testing equipment (after 1998/2000) does provide information about sound above 20 khz.

Ribbon tweeters do go beond 20khz, and Tannoy sells supertweeters wich are able to extent any monitoring system you may have (up to 50 khz).

Muziekschuur

I didn't know, thanks for the information!

But my point is still valid.. most of the people and consumers (read 95% of the people) will listen to your mix on multimedia speakers, headphones, TV, car, etc.

What's the point of spending so much money if the final result won't be perceived equally?

[gsilbers]

i did the test and i didnt tell the dif from a square and sine wave at 13k. but then again, i was listening to alesis monitor one speakers that i dont think go beyond 20k.

[roger]

my ears don't go to 20.......

[davidwilson]

Yeh the 20K thing is an old argument now. They have ring tones that only young people can here assuming over 20's can't here 20K. I seems to work.
So I guess unless your Brittney or Rhiannon theres no point in worrying about over 20K.


Whats got me excited in their exitment over IC,s over tubes. That's it im having my Manley Pre's modified to have the tubes removed and put in a few NE5534's YEH!!.

[drBill]

I know Duco de Rijk - the founder and owner of D&R . He knows his stuff and it's not marketing BS. Believe or not, makes no difference to me, but the fact is that overtones and transients at very hi freq's do exist. Pretty much the whole reason he makes recording consoles at all is for extremely transparent, high bandwidth, low phase shift audio. He's definately not into the "distortion", "warmth" or "Phat" camp when it comes to making audio products. But if you want transparent and linear, listen to what he says.

Duco is a very smart electrical engineer and entrepeneur who's been manufacturing consoles for close to 40 years and is independantly owned, still in business and doing great. (How many guys are out there that can say that???) As you read what he's writing, keep in mind, English is his THIRD language and when he says "warm and open", he most likely means open. He's well aware that the freq range that we describe as 'warmth" resides in the lower mids. He's also well aware of the Nyquist theory and the brick wall filtering that digital employs.

Also keep in mind that many botique audio manufacturers (like Duco) believe in 100-300k bandwidth. Why on earth would they do that when the only delivery medium we really have is 16b/44.1k digital???

Having extended bandwidth definatly helps in the delivery of overtones and transients. Having a console whose -3dB down points are somewhere approaching 200K (my D&R OrionX) makes a difference to me, and I'm sure I can't hear much over 20k. But I can "feel" the "punchy-ness", "open-ness" "or "transparency" (for lack of a better term) that good analog makes.

But if all you've ever heard has been an analog mackie or soundcraft.......you might think me silly. No matter to me. Just filling in a few cracks.

[ryanojohn]

Beyond the fact that most speakers can't do it... 90% of input and output transformers only pass 20-20k... unless it's the custom transformers that Millenia uses...

[roger]

and most mics won't go any higher than 20 to capture these 'super high frequencies' either..........i have no doubt that these frequencies do exist in music - it's just that you and i can't hear 'em to enjoy beauty!
i don't agree that 'openness' or 'transparency' are phenomenoms that require >20 hearing or >20 gear.....

[drBill]

So I guess 96k - or even 48k are completely useless innovations the, eh?

[ryanojohn]

96k and 48k don't change the cutoff frequency of the analog LPF built into every A/D, they just reduce the slope, and thus reduce the rippling effect that does affect the frequencies IN our audible range...

[NLP]

http://images.industryclick.com/file...p_wideband.pdf
and
There's life above 20 kilohertz! A survey of musical instrument spectra to 102.4 kHz

we dont hear only with ears

[ryanojohn]

Originally Posted by NLP

http://images.industryclick.com/file...p_wideband.pdf
and
There's life above 20 kilohertz! A survey of musical instrument spectra to 102.4 kHz

we dont hear only with ears

that is some phenomenal reading... thanks!

[mohthom]

I don't know whether I'm completely misguided on this whole thing, but thought it a good opportunity to offer another opinion on the wideband audio phenomenon.

It's pretty much agreed that we can't hear too much above 20k (even us young'uns who look after our ears, use our chest cavities to the full etc.), but working at higher samplerates and with wideband equipment seems to improve my sound (as does higher bitrates etc.). My theory argues that while it's impossible to hear such high frequency sounds, the ability of the equipment to work at that high a frequency indicates that the electronics are 'fast' enough to be able to reproduce a 150k oscillation etc. - therefore a reproduction of a 20k (or more usefully 2k) tone is going to be far more accurate as the speed of the electronics to adjust to the 'real' frequency will be much higher.

It's a pretty simplistic view - I guess a good analogy would be a sports car. It's possible to drive a Nissan Micra at 110mph, but it's going to struggle quite badly - if you use a Ferrari Enzo to do the same thing, it's going to have a lot more control because of the increased speed available to it. It's been designed to be operated at the bleeding edge - at the limits of speed available to us, so when used in a fairly 'normal' 70mph motorway drive, it's going to have incredible response simply because of the increased control required for the high speeds 200mph+ that the car was designed for.

Am I way off the mark with this one, or does this float?

MohThoM

[thenoiseflower]

http://www.earthworksaudio.com/tech/...yond_20khz.pdf


why not open your mind to the idea that we dont have control/perception/understanding over everything in this universe, sound has many many other properties we cannot perceive, or even begin to speak to.

why think that the book on hearing and kHz is closed?

[thenoiseflower]

http://www.earthworksaudio.com/tech/hf_sound.pdf

[Duardo]

The math on square waves is indeed correct; a pure square wave has only the fundamental plus odd harmonics. The leap of faith is that we're hearing the square wave differently from the sine because we are hearing the high harmonics; there are certainly other possibilities (see "aliasing" above.)

Also the possibilith that that higher-frequency signal is producing audible distortion that is within the spectrum of frequencies we can hear. Especially since, as has already been pointed out, most equipment isn't designed to pass those higher frequencies.

Pretty much the whole reason he makes recording consoles at all is for extremely transparent, high bandwidth, low phase shift audio. He's definately not into the "distortion", "warmth" or "Phat" camp when it comes to making audio products. But if you want transparent and linear, listen to what he says.

I think the "what" of what he says has merit, but not necessarily the "why"...the fact that people can hear a difference in audio in equipment that's flat up to 100 kHz or so doesn't necessiarly indicate that people can hear those higher frequencies...especially when you can hear the differences between those pieces when listening to (or recording) 44.1 kHz digital audio...

why think that the book on hearing and kHz is closed?

Have there been any studies that have proven that we can hear above 20 kHz?

[NLP]

Im not sure about "hearing" in classical term (only with ears), maybe feeling those frequencies, harmonics, frequency interactions, interferences... (also with body and soul).
Have you read those documents?

sorry on my clumsy english

[rufus13]

Consider that your audio goes through a console many-many times before making it to a master recorder (2T tape, 2T digital, etc.). The bandwidth roll-off point affects the audio over and over on each pass.

Mackie/Soundcraft boards sound fine (especially the small ones, 8 input or fewer), as long as you only allow the audio to pass through once, and use the shortest path through.

For a small high-quality console, using good transformers is not a financial problem when comparing "high-end" products. The best US-made Jensen transformers are not over $200 each, and are wide-band (180KHz+) with low phase-shift. Using excellent input and output transformers allows a designer lots of options for the interior circuitry.

People don't hear above 16KHz much. Asthmatic Russian girls may be able to detect sine waves at 50KHz, but that ain't music. What regular people can hear is phase shift and distortion artifacts RESULTING from higher-than-audible frequencies. If you crank up an EQ and hit it with a strong signal, it might give ultrasonic oscillations, clipping the next stage, and affecting the audio. Oscillating filters are a tradition in analog synthesizers, and require attention to not inadvertently include in a mixer.

Cheers.

[Ethan Winer]

Originally Posted by kellyrmartin

I hear what you guys are saying.... But what about the point about the 12khz sine and 12khz square wave. It makes sense to me... Or is there some missing information in his explanation of why we can hear the difference?

If they sound different it's either IM distortion between the harmonics aliasing back down into the audible range, as was already mentioned, or a simple level difference. If you set up a square wave and a sine wave to measure the same level on a VU meter, they will not have the same amount of energy at 12 KHz.

--Ethan

[Ethan Winer]

Originally Posted by rufus13

If you crank up an EQ and hit it with a strong signal, it might give ultrasonic oscillations, clipping the next stage, and affecting the audio.

Exactly, and I'm certain that's what was behind the famous Rupert Never "proof" that oscillation at 50 KHz was audible.

--Ethan

[drBill]

Originally Posted by Ethan Winer

Rupert Never

Freudian slip, Ethan?