In article , ~misfit~
wrote:
Once upon a time on usenet Jim Lesurf wrote:
In article , ~misfit~
wrote:
Once upon a time on usenet ~misfit~ wrote:


Interesting. I read the page and it seemed pretty clear to me. In
general I tend to think Xiph's work is recommendable.

Having read and re-read the article in question now I must say that,
despite a lot of factual information I disagree with the conclusions he
presents in the section "192kHz considered harmful".

(For the record I believe that, while we may not be able to 'hear'
frequencies above 20kHz we detect sound with more than just our cochlea
/ Oragn of Conti. Also that there are detectable products caused by
intercation between high frequencies... Etc.... Too complicated to get
into here.)

Yes. The argument here is that the hearing system is inherently nonlinear
so the presence of sounds which - in isolation - would be inaudible may
alter our perception when they occur sumultaneously with lower frequencies.

The snag being that devices like loudspeakers are also nonlinear. Present
when you reproduce the audio, but not in the original sound.

He offers as evidence as to why that arbitrary cut-off line should be
there a graph showing the effects of intermodulation distortion caused
by two notes at 30kHZ and 33kHz. If we take it at face value than, sure
there is IMD from the reproduction of those notes that manifest in the
audible range. However, as can be detected from his graph they are at
approximately -65"dbFS" (FFS - why not complicate the issue?) compared
to the original tones.

Again, if you examine 'high res' recordings, the amount of power above
about 30k Hz is also generally about three quarters of SFA compared with
what is below 10 kHz. So the same argument that this is too small to matter
can probably be deployed. :-)

I suspect this is why the MQA have changed tack and argue in terms of
'timing'. This lets them dodge the above awkward fact, and the way MQA
actually may alter and discard HF details.



I'm aware that this is a contentious issue so will leave it at that.
Suffice to say that in speakers of my own design and manufacture I
often use quasi-ribbon 'super tweeters' or tweeters capable of up to at
least 40kHz.

I tend to suspect that the main effect of 'super tweeters' is down to them
adding output at lower frequencies that then vector sums with the main
tweeter output to modulate the sounds at lower frequencies. Possibly in a
way that is room dependent due to differences in dispersion. So hearing any
'change' may be due to components that are audible without the supertweeter
or assuming we can all hear stuff well above 20kHz.

I did however learn a bit from the article once I got past the (to me)
confusing phrasing. Perhaps the author isn't a native English speaker?
That is an increasing occurance lately and I think that I need to be
more flexible. However 'old dog - new tricks' and all that...

I thin Xiph is American.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics https://www.st-andrews.ac.uk/~www_pa...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html