Let's say it again. "Its about the music"

quote:

---

I know CD’s do this.

---

CD's do it, speakers do it, vinyl does it (at an even lower frequency), and IIRC Naim amps do it too. It doesn't matter too much, though, because it plays music so well (and sounds good too )!

-=> Mike Hanson <=-

quote:

---

Originally posted by Mike Hanson:
vinyl does it

---

IIRC, the difference between the 'S' & 'K' boards is that the 'S' cut off earlier - at around 40KHz.

No doubt, someone will correct me if I'm wrong.

cheers, Martin

quote:

---

So, can someone explain what he was talking about in terms of Naim gear cutting out at 20KHz?

---

Nonsense (this is an old high-end myth, anyway)--especially if you've ever bench-tested a NAP 500 (which is able to pass test frequencies at the limit of my AP test gear).

If anything, Naim equipment's "limited" bandwidth is, like the amplifier power ratings, an assurance that everything within that region is guaranteed to perform exceptionally well, and that information outside that region may pass through, though attenuated somewhat.

I seem to remember that the "S"-type phono boards roll off about 25 kHz, the 'K'-type about 40 kHz (the top limit of compliance speed on most MC cartridges)--makes for an interesting listen with quadraphonic vinyl, I assure you...

Dave Dever, NANA

thanks Dave, Martin

Judd

It's always a nice day for it Have a good one!
Steve.
It's good to get back to normal.

quote:

---

“Hmmm. Yea, the stuff does SOUND good, but I have too much trouble with the high frequencies being filtered off at 20KHz.”

---

By his own ears presumably. Time to remove the cloth.

Ears are sensitive to time domain effects - the infamous transient - which equates to frequency response at steady state only; the great assumption of too many 'hifi' manufacturers is that 'more is better'. The width of the window says nothing about the quality of the air coming in, and I'd rather have 8khz bandwidth done well than 50Khz of spite and dross...

MC

quote:

---

CD's do it, speakers do it, vinyl does it (at an even lower frequency), and IIRC Naim amps do it too.

---

CD players definitely do it, there is a brick wall just above 22 kHz, so people who rant on about super tweeters etc with CD really are idiots and should be treated as such. Vinyl on the other hand can certainly have treble extension up there, there are many cartridges out there with output upwards of well above 50 KHz, so (despite Mike being wrong!), this is possible assuming the info is on the source material and the rest of the system can handle it. It is also worth noting that many microphones do not extend much over 20 KHz (though they will roll off smoother), plus if the material has been digitally mastered it is probably stuffed up already - most stuff is mastered to down to 2 track DAT, and that kills ultra-sonic treble extension like CD, just a couple of KHz higher.

Tony.

The thing is we just don't hear things much above 20kHz. Especially us grown ups. Dogs, bats, intruder alarms may have something else to say on the matter.

So I say the bandwidth of your hifi system is only relevant in how it affects its performance within the audible range.

Yes, the systems bandwidth can affect its peformance in the audible range. But the bandwidth outside the audible range isn't, of itself, important.

The point about phono cartridges working up to 50kHz is a good case. I believe a high bandwidth cartridge may sound better because it can track more effortlessly (less distortion and tracking error) than a slower, more massive cartridge. But pursuit of performance extremes is not the important thing.

I think the dealer has, incorrectly, concluded that performance outside the critical range is ALWAYS an indicator of better performance within the range. Naim gear rolls off its frequency response very early compared to some other brands so the dealer thinks it is inherently inferior.

One caveat. The electrical signals within your hifi always contain frequencies outside the audible band - no matter what. The electronic components themselves generate their own wide-band noise: always. So a systems performance outside the audible band can be important as some distortion mechanisms will splinter these out of band signals and induce in-band artifacts.

The other is called 'volleying', I think. This has to do with changes in air pressure. Apparently one can be aware of the start and stop of a sound even if the sound's frequency response is below your threshold of hearing. So, even though I can't hear a steady state tone at 20KHz, I can perceive some aspects of it.

This analysis may to explain some of our varying responses to music and sound.

Phil

quote:

---

despite Mike being wrong

---

It wouldn't be the first time.

-=> Mike Hanson <=-

I think not.

I suppose if there were some non-linearities in your hearing mechanism at these ultrasonic frequencies then perhaps the distortion artifacts would create some audible stuff. I'm not convinced, though.

Please explain.

You may be thinking of amplitude modulation (AM) where one signal is multiplied by the other. In this case two difference tones are produced. So if you had a 40kHz carrier and amplitude modulated it with a 25kHz signal the resulting spectrum would comprise 15kHz (40-25), 40kHz and 65kHz (40+25). In this case you could hear the 15kHz sideband.

I haven't tried the experiment myself, so this is all just theory. But I imagine if adding ultrasonic sounds did produce a difference frequency sound then we would suffer a terrible racket in the presence of two or more bats or other ultrasonic sources

Phil
ex-historian

We can all hear timing can't we? After all we do buy Naim gear for its timing properties.

If you need proof, take a look at the square wave. The square wave (as with music) is built up from the summation of many sine waves at certain frequencies. The more high frequency components in the square wave, the more accurately it is represented. This can be seen in right-hand graph in the figure below.

Note the graph on the right shows a sine wave, "square waves" built up from a few harmonics, and a perfect square wave (with harmonics stretching to an infinitely high frequency). The transient edges of the one with limited harmonics are not as sharp or pronounced as the perfect square wave - yes indeed, limited frequency extension downgrades the timing of the signal.

Apply this to a complex musical signal and the same principles apply. Sharp dynamics (drum, guitar, piano etc) and other musical events are not accurately positioned in time, when bandwidth is limited.

Also, look at the sampling frequency of 44.1kHz. This means that a musical event is sampled every 0.02ms. This means that the musical event is portrayed with a timing error of up to +/-0.01ms (!!!).

Audiable beat frequencies are picked up at the recording phase. With being a low frequency, they are also played back.

Andrew

Andrew Randle
2B || !2B;
4 ^ = ?;

[This message was edited by Andrew Randle on FRIDAY 16 November 2001 at 15:23.]

To get a reasonably volume fluctuation the frequencies of the two tones need to be within 10% or so and the amplitudes need to be close as well. Then there is the question of how fast a beat you can actually notice and I suspect its in the 10s of Hertz at best. By the way the perceived beat frequency is not the difference of the frequencies of the two tones: the maths is more complicated than that I fear. If you are skilled with Excel you can draw graphs of two tones and their sum using sin() functions and then take a look at the "envelope" of the sum signal and try to estimate its frequency.

So I agree with you in that I believe it would be possible to hear a beat from an audible tone and an ultrasonic tone of equal amplitude. However they need to be very close in frequency, say, 20000Hz and 20010Hz. I do not think any beats would be heard with 20kHz and 25kHz and certainly no 5kHz tone.

BAM