Hi,

for outdoor recording I often use the handy Edirol R1 24bit recorder.
This recorder has 2 built-in electret mikes. As I read somewhere
electret mikes mostly have a specific amplitude raising around 12 kHz.
Actually sibilants sounds slightly excessive, respectively the high
frequencies are somewhat to strong.
I tried to reduce the 12kHz-Formants with audacity with the built-in
parametric equalizer. But there is no realtime prelistening possible
while controling the parameters, so I didn´t solve this puzzle with
Frequeny, Q and Attenuation in a way, that it sounds naturally to me.

The built in "Microphone Emulations" (menu "Int-Mic Rec" etc.) seems to
me more a gag than a helpful tool...

So my question: Has anyone been confronted with this problem and solved
it? Which are the parameters for correcting the sound of the R1 with an
parametric equalizer? Or is one of the internal programs with some
changed parameters suitable for this?

Thank you!
Thorsten

Which are the parameters for correcting the sound of the R1 with an
parametric equalizer? Or is one of the internal programs with some
changed parameters suitable for this?

I think your first problem here is that your ears should be your
guide... There is never a "magic solution" in audio, and as you have
found, presets in plugins can be decent starting points, but will never
be an end-all solution. There are just too many variables in the real
world!

Thorsten Wahn wrote:
for outdoor recording I often use the handy Edirol R1 24bit recorder.
This recorder has 2 built-in electret mikes. As I read somewhere
electret mikes mostly have a specific amplitude raising around 12 kHz.

No, this isn't true. YOUR mikes may have such a peak, but that isn't
anything generally characteristic of electrets.

Actually sibilants sounds slightly excessive, respectively the high
frequencies are somewhat to strong.

This is more likely to be a distortion problem than a frequency response
problem. But it's easy to tell... if you can fix the problem with EQ,
it's a response problem. If you cut and cut and it doesn't get any
better, it's not.

I tried to reduce the 12kHz-Formants with audacity with the built-in
parametric equalizer. But there is no realtime prelistening possible
while controling the parameters, so I didn=B4t solve this puzzle with
Frequeny, Q and Attenuation in a way, that it sounds naturally to me.

I suggest borrowing an actual parametric equalizer from someone and
trying. Set the Q tight, set it to boost, then sweep the frequency
until you find it gets really bad. Then widen the Q until it stops
getting worse, and knock it back in a little. Now, change from boost
to cut and take it out.

Unless you can do the adjustments in realtime while listening, you will
never be able to get them right. I do strongly suspect, though, that it
is a distortion (linearity) issue that EQ won't fix. But you need to
find out for sure.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

electret mikes mostly have a specific amplitude raising around 12 kHz.
No, this isn't true. YOUR mikes may have such a peak, but that isn't
anything generally characteristic of electrets.
Scott, maybe I'm wrong with my generalization.
A posting describing the "Panasonic WM61A electret element" which is
described as a "representative electret omni", led me to the
conclusion, that apparently this characteristic applies too to the
mikes of the R1.

This is more likely to be a distortion problem than a frequency
response problem. But it's easy to tell... if you can fix the problem
with EQ, it's a response problem.
As I found, the sound will be better, if I apply some lowpass
characteristic respectively lowering the frequencies around 12kHz. But
it sounds still unnatural, as long as I don´t know the frequency
response.

Set the Q tight, set it to boost (...)
Okay, this is the usual procedure - but with my insufficient
equipment...
I thought also about a procedure, where I record white noise and make a
fft analysis - but can I trust my speakers? =8-)

Unless you can do the adjustments in realtime while listening, you
will never be able to get them right.
That´s surely right. So I have to search for a wave editor with
realtime EQ -- if not somebody did this already with the Edirol R1 and
would disclose the parameters... ;-)

....maybe I should get the right tools, instead of fumble with
inadequate equipment.

Thorsten

"Thorsten Wahn" wrote ...
A posting describing the "Panasonic WM61A electret
element" which is described as a "representative electret
omni", led me to the conclusion, that apparently this
characteristic applies too to the mikes of the R1.

But note that those elements are remarkably flat for
something so small and so cheap. I'd bet that if there
is some kind of rising response, the mechanics of the
case (like a small resonant cavity between the diaphragm
and the outside) may be more likely responsible.

But it sounds still unnatural, as long as I don´t know
the frequency response.

So there are many ways of characterizing the response
of your unit. They range from amateur/cheap/time-consuming
to professional/expensive/quick. Your choice.

Thorsten Wahn wrote:
electret mikes mostly have a specific amplitude raising around 12 kHz.
No, this isn't true. YOUR mikes may have such a peak, but that isn't
anything generally characteristic of electrets.

Scott, maybe I'm wrong with my generalization.
A posting describing the "Panasonic WM61A electret element" which is
described as a "representative electret omni", led me to the
conclusion, that apparently this characteristic applies too to the
mikes of the R1.

Actually, the mikes on the R1 aren't omnis. But the WM61A is actually very
flat. They do some trickery where they build a resonant chamber in front
of the diaphragm that causes the response to rise on top just as the response
of the rest of the capsule drops off. The end result is a frequency response
that is actually quite flat out to 20 KHz and a phase response that is very
wacky.

The WM61A is pretty standard as electret omnis go. And it doesn't have a
rising top octave, although it does sound gritty as hell on top.

This is more likely to be a distortion problem than a frequency
response problem. But it's easy to tell... if you can fix the problem
with EQ, it's a response problem.

As I found, the sound will be better, if I apply some lowpass
characteristic respectively lowering the frequencies around 12kHz. But
it sounds still unnatural, as long as I don=B4t know the frequency
response.

If the problem is due to distortion, all the cutting in the world won't
make it much better, because what is happening is that there are distortion
products in that region which are being added to the actual signal. You
cannot remove one without removing the other.

If the problem is an actual response problem, cutting will fix it, and you
can do it by ear.

Set the Q tight, set it to boost (...)
Okay, this is the usual procedure - but with my insufficient
equipment...
I thought also about a procedure, where I record white noise and make a
fft analysis - but can I trust my speakers? =3D8-)

No, you can't trust your speakers or your room. And if you DID have access
to a chamber and a calibrated noise source, I bet you'd find the mikes were
actually pretty flat.

.=2E.maybe I should get the right tools, instead of fumble with
inadequate equipment.

If it's a distortion issue, there IS no right tool. But without an actual
equalizer, you won't know. I'll suggest you call up a local sound rental
place and ask if they have an Orban 622 parametric in the junk box, rather
than try and do it on the workstation. The Orbans are always handy to have
around and not expensive to buy or rent.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

On Sat, 15 Apr 2006 12:09:46 -0400, Thorsten Wahn wrote
(in article .com):

Hi,

for outdoor recording I often use the handy Edirol R1 24bit recorder.
This recorder has 2 built-in electret mikes. As I read somewhere
electret mikes mostly have a specific amplitude raising around 12 kHz.
Actually sibilants sounds slightly excessive, respectively the high
frequencies are somewhat to strong.
I tried to reduce the 12kHz-Formants with audacity with the built-in
parametric equalizer. But there is no realtime prelistening possible
while controling the parameters, so I didn´t solve this puzzle with
Frequeny, Q and Attenuation in a way, that it sounds naturally to me.

The built in "Microphone Emulations" (menu "Int-Mic Rec" etc.) seems to
me more a gag than a helpful tool...

So my question: Has anyone been confronted with this problem and solved
it? Which are the parameters for correcting the sound of the R1 with an
parametric equalizer? Or is one of the internal programs with some
changed parameters suitable for this?

Thank you!
Thorsten


Try the Yamaha NS-10 approach. Over the mic grille with sheets of toilet
paper.

Regards,

Ty Ford

-- Ty Ford's equipment reviews, audio samples, rates and other audiocentric
stuff are at www.tyford.com

Try the Yamaha NS-10 approach. Over the mic grille
with sheets of toilet paper.

It's unlikely such filtering would concentrate around a particular
frequency. It would more likely cause a gradual rolloff, or shelving.

"William Sommerwerck" wrote in message
...
Try the Yamaha NS-10 approach. Over the mic grille
with sheets of toilet paper.

It's unlikely such filtering would concentrate around a particular
frequency. It would more likely cause a gradual rolloff, or shelving.

No, it depends on the brand of toilet paper. Charmin is most useful for
lower midrange honkiness. Scotts single ply is good for when there is too
much air. Costco's store brand, when used at triple thickness, simply
knocks out sibilance. You have to know your equipment.

Steve King

Costco's store brand, when used at triple thickness, simply
knocks out sibilance.
Used or unused?

Steve King wrote:

"William Sommerwerck" wrote ...
Try the Yamaha NS-10 approach. Over the mic grille
with sheets of toilet paper.

It's unlikely such filtering would concentrate around a particular
frequency. It would more likely cause a gradual rolloff, or shelving.

No, it depends on the brand of toilet paper. Charmin is most useful for
lower midrange honkiness. Scotts single ply is good for when there is too
much air. Costco's store brand, when used at triple thickness, simply
knocks out sibilance. You have to know your equipment.

Man, you really know your fecal matter! I'm impressed, Steve. LOL!

--
ha

"hank alrich" wrote in message
.. .
Steve King wrote:

"William Sommerwerck" wrote ...
Try the Yamaha NS-10 approach. Over the mic grille
with sheets of toilet paper.

It's unlikely such filtering would concentrate around a particular
frequency. It would more likely cause a gradual rolloff, or shelving.

No, it depends on the brand of toilet paper. Charmin is most useful for
lower midrange honkiness. Scotts single ply is good for when there is
too
much air. Costco's store brand, when used at triple thickness, simply
knocks out sibilance. You have to know your equipment.

Man, you really know your fecal matter! I'm impressed, Steve. LOL!


I only know about what we have to wade through on this newsgroup. But,
about toilet PAPER, I've got this little rod on the wall. Sort of like the
dispenser on the shop wall with the red, green, blue, and pink wire on
little bitty spools. It holds five rolls, each with a different brand of
paper; each with a sound color of its own. I currently use SM57's for
everything. Want a little darker sound, like a U-67, bingo that's three,
maybe four layers of Costco. Nasel voice, one or two sheets of Charmin and
that SM57 sounds like a Royer ribbon. I've sold off my mic closet and my
Pultecs. Don't need 'em any more.

Steve King