The time writing spectulative posts without gaining any knowledge is much better invested reading stuff like Witteks thesis or Toole's http://www.harman.com/EN-US/OurComp...p/Documents/Scientific Publications/13686.pdf. If you really want to learn something, it doesn't get any easier than this.
At lower frequencies the sum of direct and indirect sound is perceived AS ONE. That allows for room correction to work.
Then there's a transition to higher frequencies where direct sound and room response are perceived as two different entities (although only ONE sound sensation is perceived). So we end up measuring and correcting both at the same time. This is obviously wrong because our brain distinguishes between the direct and indirect sound field, the room correction does not.
All you can do is filter out the indirect sound field by gating the measured impulse response in order to optimize the loudspeaker frequency response (but then you could have used decent loudspeakers in the first place). The problem is that placing the microphone at the listening position renders the measurement useless. Either the gating time is too short and the frequency resolution becomes too coarse or the measured frequency response is contaminated with room reflections.
Then there's a transition to higher frequencies where direct sound and room response are perceived as two different entities (although only ONE sound sensation is perceived). So we end up measuring and correcting both at the same time. This is obviously wrong because our brain distinguishes between the direct and indirect sound field, the room correction does not.
All you can do is filter out the indirect sound field by gating the measured impulse response in order to optimize the loudspeaker frequency response (but then you could have used decent loudspeakers in the first place). The problem is that placing the microphone at the listening position renders the measurement useless. Either the gating time is too short and the frequency resolution becomes too coarse or the measured frequency response is contaminated with room reflections.
If that is Key's view then he is under a misconception of what stereo is.
Talking about Wittek, here's what he concludes: "In general, the physical properties of the reproduced space do not necessarily have to be realistic or existent in any real situation." (Perceptual differences between wavefield synthesis and stereophony by Helmut Wittek, p. 4)
Talking about Wittek, here's what he concludes: "In general, the physical properties of the reproduced space do not necessarily have to be realistic or existent in any real situation." (Perceptual differences between wavefield synthesis and stereophony by Helmut Wittek, p. 4)
When looking at the curves from Olive's presentation that
show "spectral balance" judgements of the listeners i come
to a simple conclusion:
What mainly has been "balanced" or "equalized" above 500Hz
- where the room behaves statistical and not modal anymore -
is the narrowing coverage of the midrange driver,
especially around 1Khz up to the crossover frequency
of the tweeter.
When comparing the best RC (RC1) with RC4 (no correction) this
is rather obvious.
When crossing two transducers with very different coverage at
the crossover frequency, there will always be a compromise between
flatness of on-axis response (in a large and dry room) and the total
radiated power which is essential in more 'wet' and small rooms.
The optimum crossover alignment is then strongly dependent
from listening room, listening position (distance!) and the
loudspeaker's position.
Such a problem could more easily be corrected in the speakers
crossover itself, by providing different characteristics for different
listening positions and rooms, if the designer is aware of the problem
and willing to reduce it (outside of the room where the speaker has
been aligned).
It points me to the assumption, that a speaker having constant
coverage or at least coverage "only smoothly varying with frequency"
does not need "room correction", especially if the coverage is restricted
to a narrow angle.
So in the end there is no "room correction", but there is modal balancing
for the modal frequency range of the room or "room/speaker system"
.
Last edited:
on the contrary - it does - trying something Yourself is easier
BTW Toole arguments in this article indirectly support the flooder concept:
http://www.diyaudio.com/forums/multi-way/121385-loudspeakers-room-system-81.html#post2124502
best,
graaf
I would say the opposite as well
we never ever hear the frequency content of the direct sound as such
it is completely irrelevant as such
in terms of frequency content we hear a sum of direct sound and all reflections
therefore indirect sound can in a sense correct not-flat [direct] FR
and therefore actually it is easier to get a good frequency response with a CFS
not boosting as such but high volumes
anyway - in the near-corner placement of the flooder just 6 dB boost @50 Hz and below would make me happy
that is not heavy boosting, or is it?
lucky You!
BTW no Key's "errors" in case of the flooder
I don't know
only two ocurrences of the word "transient" in the whole text
and music and speech = transients
I like this one more:
"Spatial Hearing with Simultaneous Sound Sources: A Psychophysical Investigation"
http://ses.library.usyd.edu.au/bitstream/2123/576/1/adt-NU20041221.13524302whole.pdf
best,
graaf
Last edited:
I am not sure what You mean by a meaningful way but I have a CFS here in my home and so has el'Ol and we have tried some options and some placements in room
I have also tried SLS (stereolith-like-setup) and I built three different versions of that stereolith thing
And You cannot read about it in Wittek, Toole or anywhere else
The only technical article I know that mentions SLS is Manger diskus patent description. I don't know any technical articles that mention CFS.
but I just have built it and it works
best,
graaf