I got a feed back from David Brinicombe.
I think it is a very well done evaluation. I do not agree to all of
it, and a few minor matters are due to circumstances that David could
not know about. For example, there were no trees in "the middle" on
the track "justesomewind.mp3". No wonder that he didn't hear any...
Also, using my headphones, I find quite a big difference between the
two omnis and the Telinga SSM on the track "rig_omnis.mp". I find
the SSM stereo spacier, though not overdone.
It makes me happy that David doesn't seem to hear that the omnis are
a pair of totally different mic's, namely AT3032, which I have used
as reference mic's during the development of the SSM.
As a whole, I think David's way of thinking, reasoning and evaluating
is great and something to learn from.
Here it is, authorized by David himself. Thank you David!
/ Klas
----------------------------
Klas,
I've had a bit of fun with your tracks.
My ears were trained to mix stereo for loudspeakers using headphones as
well as judging binaural.
Now I want to see your rig. :-)
justsomewind.mp3
Clear leaf noise with wind in trees.
Separate wind noise on mic in low wind peaking about 50Hz,
medium 100 HZ and sometimes higher in brisk wind. There is a
local resonance at these frequencies which is being
stimulated by direct wind.
The bass wind noise in the trees is distinct from this close
mic wind noise and the tree noise is non-reverberant.
A lack of centre image. There is good separation between
left and right, but little common centre image. Track
suitable for adding to another wildlife track to give wind
atmosphere. Is this because the sound is coming from above?
(I commented: There were trees to the left and right, but not in the centre=
.)
walkinginacircle.mp3
Excellent stereo object/image location test, snow gives
quite good "white noise" source but lacking in higher
KHz. (In summer use shaking beans or peanuts)
The start sounds like 90 deg left, slightly unclear image
location, then at about 4 to 5 secs becomes a clear left
image at around "conventional" 60 deg left. Possible slight
fall off in centre, but nothing to detract from centre
image, across to the corresponding 60 deg right image.
From then on, the image spreads as on the left with a out
of phase component which remains at 180 deg back object
position, and back to 90 left.
The forward image is excellent at these frequencies and
appears not to spread with lower frequency at least in the
kilohertz band. The object/image distortion is small and I
would rate it as a 9 sonel (sound pixel) resolution which is
beyond many domestic reproduction systems.
The back image remains less distinct with a very audible
loss at higher KHz, and listening on close speakers does not
give an accurate rear image. The rear image remains fairly
static until the centre back when it speeds up to become a
right image. Resolution is barely 5 sonels (left, centre,
right, and in between those) due to image stretch.
On headphones, there is an out of phase rear component which
when combined with the loss of HF gives a rear image effect
to that part of the walk only using headphones. The rear
image is only about 3dB lower than the front image, mainly
due to the HF loss, giving an overall pickup response
similar to an omni. Thus there is very little rear noise rejection.
With this proviso, the forward stereo image placement is
excellent.
Applying my HPF Audacity HPF800-1-2-3 made the footsteps
clearer and improved the 360 degree effect without removing
the footsteps and the more distant tree noise.
rig_omnis.mp3
The tree noise was prominent and giving a 360 deg image on
headphones. There are some out of phase components but as
these are lower than the forward image, they give a slight
surround sound effect on headphones.
The birds are weak and appear to be placed fairly centrally
in the forward image. However they are only a few dBs above
background even after the HPF and therefore not recoverable
as specimen recordings with further filtering or noise
reduction.
Using the HPF 1000-3-6-9 twice, the HF stereo image was
clearer, and leaves blowing about and other birds were
audible with good forward placement.
Now for the actual test rig against omnis. With the bass cut
in, it would be difficult to distinguish the difference.
With the bass enhanced slightly, this showed up the small
difference in the mid range stereo image, which was
clearer between 5 and 10 secs in with the omnis.
With a mid range lift, the general mid to low noise was
slightly less after 5 secs and there was a slight stereo
spread in the mid frequency noise, but this would be
difficult to distinguish without a direct cross-cutting
check.
The overall verdict is that the omnis were not much better
by a small amount. What may be more important is the wind
and handling noise.
Summary
In all tracks, there appears to be very little "splash" or
stereo crosstalk where a left image spills over to the right
and vice versa. An exception is in the rear image where
there is some out of phase crosstalk, and this is what
rendered the rear image less clear, but this did not amount
to splash.
My guess as to this rig is that it uses a baffle of some
sort to give stereo separation. The imaging is similar to my
Brinibox, which does however give quite a lot of forward
gain and better rear rejection. The Brinibox is a different kind of
rig and produces level difference stereo.
The overall sound seems clean and the power spectrum shows
no peaking above 100Hz where there appears to be some
vehicle noise.
This rig shows little "fetch" aka "reach" but gives clean
sound with a good forward image, with a pickup area of 360
deg.
Bird feeder.mp3
This is where the rig works best. There is excellent
separation with the close birds against the background. I
used a mild HPF800-1-2-3 which reduced the background
without affecting the subject matter. Wing noises were
excellent in stereo and gave vivid images of the birds'
movements.
There are a few resonances in the background noise
especially around 500Hz which also show up on the audiogram
display. This could be within the rig or more likely caused
by reflections from the ground or close objects. This is one
of the reasons I like to put mics on the ground, but the
effect is slight.
Listening to the car pass at 13:45, there is no stereo image
at the low frequencies, double checked by using a bass lift,
but the pass registers in the upper 100's with quite a good
placement, so that is confirmation that stereo is being
produced from, say, 300Hz upwards.
Best Wishes, David
David Brinicombe
North Devon, UK
Cogito cogito ergo cogito sum - Ambrose Bierce
Telinga Microphones, Botarbo,
S-748 96 Tobo, Sweden.
Phone & fax int + 295 310 01
email:
website: www.telinga.com
|