At 1:53 PM +1100 11/20/07, Paul Jacobson wrote:
>Hi Rob,
>
>I'm using the "reverse wedge" design, at the bottom of the Curt's
>stereo rigs page. I've tried to replicate the dimensions Curt
>describes in his notes as far as possible.
The imaging performance of the reverse wedge design is quite
surprising isn't it? Here's a photo of your rig compared to one I
recently took of Curt's:
http://ad2004.hku.nl/naturesound/RobD/Olson-Jacobsen-SetBacks.jpg
Could the set-back distance on your rig be a tad greater than Curt's?
When a group of students and I studied this variable, we found that
localization stood up quite well with parallel barriers even when the
capsules were only set-back 1/2" (12.5mm) from the leading edges.
Some of us felt that localization was slightly better and the depth a
bit more uniform with the shorter set-back. I'd be tempted to try a
shorter set-back and the barriers a little less angled-out. This
might produce a tad more brilliance from front-center and perhaps a
little less front-back confusion. Its hard for me to judge without
the knowledge of where things were in the setting but I think you can
get a little better localization around center with some adjusting.
>The length of the barriers is pretty close - within a couple of
>millimetres, but the boards 10mm narrower and 6mm thinner than the
>100mmx25mm boards that Curt used. The external faces of the barriers
>are 12mm closer together than on Curt's rig, as I've matched the
>internal spacing of the barriers at the front of the array to
>maintain a similar angle between the barriers.
>
>Thinking about it, I have most likely got the the LF-cut switched in
>on the AT-3032's which could also account for some of the missing bass?
Very useful to know. Also, now that I'm listening at home (rather
than in an airport corridor saturated with low Hz), the left-right
contrast might stem from the right mic having a bit more response in
the lower mid range. I can hear this when I momentarily click on the
timeline at various points as slightly more "low growl" on the right
channel. Sonograms of a couple of seconds with minimum presence
(~6:45 in) seem to bear this out:
Frequencies under 1600Hz. Right channel on right side; left channel on left.
http://ad2004.hku.nl/naturesound/RobD/Left_Right_Channels%3C1600K.jpg
Right Channel with 100-500Hz highlighted below:
http://ad2004.hku.nl/naturesound/RobD/Right_Louder_100-500Hz.jpg
The upper harmonics from the lowest octaves might lessen this
difference if you don't use the roll-off filter. Sounds from the
greatest distances often create the lowest octaves as they interact
with the local surfaces. Leaving them in might add spatiality. One
can always "notch" out over-emphasized low frequencies that seem to
create a sustained "drone(s)" and this can leave more of the spectrum
and natural tonal balance intact. Could the left mic could be a tad
hotter under 100Hz? If so, this might also even-out with the low Hz
cut filters off. Rob D.
>
>cheers
>Paul
>
--
|
