[Arny Krueger[_4_]]

"Mxsmanic" wrote in message
...
Arny Krueger writes:

Right now, 110 dB down is a good number for the noise floor of a
good 24 bit recorder, and this can be a useful advantage over 16 bits.

And for a small, hand-held 24-bit audio recorder, what would a typical
noise
floor be?

90 to 110 dB. If memory serves, the noise floor of my Microtrack in 24 bit
mode is just under 100 dB.

If I understand correctly, and assuming I have a good idea of the noise
floor
of my recorder, I could try to set the average recording level about
midway
between that floor and 0 dB and get the best result out of the
recorder--right?

Average levels don't really matter when you are recording. Recording is all
about how you handle the peaks. When I make recordings of things that I
don't know too much about, I set the level so that the peaks are a minimum
of 10 dB below digital full scale.

Average levels do matter when you are providing someone with a recording to
listen to. But, unless you do dynamics processing, they are what they are as
a consequence of the peaks.

When I make a recording I adjust levels so that the peaks are about 1 dB
below the maximum.

During production after the recording session, I generally don't do
compression or gain riding as such, but if there are anomalous short peaks,
I attenuate them. By short I mean on the order of several milliseconds.

[Arny Krueger[_4_]]

"Scott Dorsey" wrote in message
...
Mxsmanic wrote:
Arny Krueger writes:

Right now, 110 dB down is a good number for the noise floor of a
good 24 bit recorder, and this can be a useful advantage over 16 bits.

And for a small, hand-held 24-bit audio recorder, what would a typical
noise
floor be?

One that I tested recently had the noise floor at around -62 dBFS, with
the analogue gains set about where I'd have them for orchestral work.
The limiting factor was entirely the preamp. But what do you expect when
you shoehorn precision low-noise electronics in a box with a bunch of
noisy digital stuff?

With good mixed-signal design, low level analog and digital in close
physical proximity need not be a problem. Of course, the outer box may be
subdivided.

[Arny Krueger[_4_]]

"Mxsmanic" wrote in message
...
Mike Rivers writes:

By the way, there are a couple of handheld recorders that have a good
solution for you. The Sony PCM-D50 has the best limiter for this sort of
recording that I can imagine. It records a second copy of the audio in a
buffer at 20 dB or so lower than the main recording. When it detects an
overload, it takes that section from the buffer, normalizes it, and
automatically and seamlessly splices it in to replace the overloaded
section. It's really amazing.

Cool! I've heard about the recorder but not about that feature. Although
I'm a
bit wary of automation unless there's also a way to turn it off.

The Zoom H4n lets you record four channels, two with
the internal mics and two with external mics, so you could do the same
thing with that, just setting one stereo track pair 20 dB below the
other.

Hmm, I hadn't thought of that ... but then you'd get a difference because
you
are using two different pairs of mics, no?

Good point, and this brings up a problem with the H4 - it doesn't have 4
equal concurrently-operable input channels. I seem to recall that I've seen
some mods that turn the internal mic inputs into line level inputs or some
such.

The world awaits a good product with H4-like pricing and 4 symmetrical
separate mic/line inputs.

You can probably pretty closely duplicate the sound of the internal mics
with external mics if you tried.

[Arny Krueger[_4_]]

"Mxsmanic" wrote in message
news
Ty Ford writes:

In the video world, some digital camcorders seem to have inelegantly
engineered audio sections. On those camcorders, if you try to hit even -6
dB,
things start to sound gnarly.

The drawbacks of camcorders are some of the reasons why I've been
experimenting with recording sound alone lately. The most obvious
limitation
is that built-in mics aren't ideally positioned for a lot of video work.
You
can connect external mics, but it has occurred to me that just recording
the
audio separately and then syncing it later on might be the best solution,
which is why I got a separate audio recorder. Heck, if it's good enough
for
Hollywood, it should be good enough for me. My current videos are not
highly
dependent on good audio, but I want to be prepared in case I need it.

I've done a fair amount of work with a standard camcorder and my Microtrack
doing the sound. The crappy sound track from the camcorder is good enough to
make synching the Microtrack recording a cinch. For really long scenes
there's a little drift, but I just resynch the two between words.

[Mxsmanic]

Arny Krueger writes:

I've done a fair amount of work with a standard camcorder and my Microtrack
doing the sound. The crappy sound track from the camcorder is good enough to
make synching the Microtrack recording a cinch. For really long scenes
there's a little drift, but I just resynch the two between words.

Hmm, I hadn't thought of just syncing the independent audio to the camcorder
audio. I was trying to decide whether I'd need a clapper or something for
sync. But as you point out, I can just line up the external audio with the
camcorder audio, and I'm done.

[Trevor]

"PStamler" wrote in message
...
3. To Trevor who thought it was common for low-noise electronics to
coexist happily with noisy digital circuits: yes, it is common. But
not in cheap portable equipment.

Right, but since it can be done, and many do so even in some not too
expensive gear, the "what do you expect" line was clearly irrelevent.

Trevor.

[Trevor]

"Arny Krueger" wrote in message
...
"Mxsmanic" wrote in message
...
Mike Rivers writes:

By the way, there are a couple of handheld recorders that have a good
solution for you. The Sony PCM-D50 has the best limiter for this sort of
recording that I can imagine. It records a second copy of the audio in a
buffer at 20 dB or so lower than the main recording. When it detects an
overload, it takes that section from the buffer, normalizes it, and
automatically and seamlessly splices it in to replace the overloaded
section. It's really amazing.


I've been doing that manually for *many* years when I have spare tracks and
unknown levels, with the benefit that I have no transition change of levels
introduced by the device automatically splicing anything. Better done in
post IMO.


Cool! I've heard about the recorder but not about that feature. Although
I'm a
bit wary of automation unless there's also a way to turn it off.

The Zoom H4n lets you record four channels, two with
the internal mics and two with external mics, so you could do the same
thing with that, just setting one stereo track pair 20 dB below the
other.

Hmm, I hadn't thought of that ... but then you'd get a difference because
you
are using two different pairs of mics, no?

Good point, and this brings up a problem with the H4 - it doesn't have 4
equal concurrently-operable input channels. I seem to recall that I've
seen some mods that turn the internal mic inputs into line level inputs or
some such.

The world awaits a good product with H4-like pricing and 4 symmetrical
separate mic/line inputs.

You can probably pretty closely duplicate the sound of the internal mics
with external mics if you tried.

Or perhaps use two if necessary.

Trevor.

[Trevor]

"Mxsmanic" wrote in message
...
Arny Krueger writes:
I've done a fair amount of work with a standard camcorder and my
Microtrack
doing the sound. The crappy sound track from the camcorder is good enough
to
make synching the Microtrack recording a cinch. For really long scenes
there's a little drift, but I just resynch the two between words.

Hmm, I hadn't thought of just syncing the independent audio to the
camcorder
audio. I was trying to decide whether I'd need a clapper or something for
sync. But as you point out, I can just line up the external audio with the
camcorder audio, and I'm done.

The clapper simply gives you a nice sharp peak to line up with the vision.
It's not always simple to accurately line up complex audio when recorded
from different mic locations remember. But I have been doing it for a decade
or more without a clapper myself. For lip sync you don't have to be sample
accurate anyway.

Trevor.

[Mark]

If you are putting (say) 5 watts into each speaker, I don't see how the
total power into the room as a whole is other than 10 watts, unless you
put the cones face to face so there is significant coupling between them.

You'd hear the 6dB increase in the listening position you describe, but
there'd be a corresponding null in a different listening position.


yep its an array.... that's correct, 6 dB more in some places and a
reduction in other places.....

I had to smile when I read this...this is the EXACT question I got
wrong on my physics final many years ago... :-)

Mark

[Ty Ford]

On Wed, 18 Jan 2012 19:40:20 -0500, Mxsmanic wrote
(in article ):

Ty Ford writes:

In the video world, some digital camcorders seem to have inelegantly
engineered audio sections. On those camcorders, if you try to hit even -6
dB,
things start to sound gnarly.

The drawbacks of camcorders are some of the reasons why I've been
experimenting with recording sound alone lately. The most obvious limitation
is that built-in mics aren't ideally positioned for a lot of video work. You
can connect external mics, but it has occurred to me that just recording the
audio separately and then syncing it later on might be the best solution,
which is why I got a separate audio recorder. Heck, if it's good enough for
Hollywood, it should be good enough for me. My current videos are not highly
dependent on good audio, but I want to be prepared in case I need it.

Then buy an Alexa camera. It has 24-bit, 48 kHz audio.

Regards,

Ty Ford

--Audio Equipment Reviews Audio Production Services
Acting and Voiceover Demos http://www.tyford.com
Guitar player?:http://www.youtube.com/watch?v=yWaPRHMGhGA

[Mxsmanic]

Ty Ford writes:

Then buy an Alexa camera. It has 24-bit, 48 kHz audio.

The Alexa is too expensive. And my impression is that Hollywood is still
recording audio separately, albeit perhaps with some sort of electronic
synchronization these days instead of a simple clapper.

[Trevor]

"Mark" wrote in message
...
If you are putting (say) 5 watts into each speaker, I don't see how the
total power into the room as a whole is other than 10 watts, unless you
put the cones face to face so there is significant coupling between them.

You'd hear the 6dB increase in the listening position you describe, but
there'd be a corresponding null in a different listening position.


yep its an array.... that's correct, 6 dB more in some places and a
reduction in other places.....

I had to smile when I read this...this is the EXACT question I got
wrong on my physics final many years ago... :-)


What you confused radiated power with SPL then too? :-)

Trevor.

[Peter Larsen[_3_]]

Mxsmanic wrote:


So the meters should hover around -10 dB?

IF live recording THEN during soundcheck/rehearsal -12 dB "at a glance", it
is a good "typical value" to aim for. Beware of wimmen singers.

Of course once it is recorded, you can bring the level
back up digitally so you don't have too quiet a recording.

What should I bring it up to?

This depends on end target and on peak density. With classical I aim
for -0.61 dB and often use hard limiting to keep the single worst peak
there. If it is pop brick mix then you may need to go lower. Beware of
wimmen singers.

There are two issues to keep in mind: not all DA converters have headroom
above zero and mp3 encode-decode overshoot. With one of my mixes on
http://raw-tracks.com I had to lower peak level to -2 dB to keep the
overshoot of encode-decode unclipped in playback. Beware of wimmen singers.

Oh, in case I didn't make it clear: Beware of wimmen singers!

It really is a balance, what also matters is whether the actual equipment
you record with sounds better when driven harder, as was the case with my
SV3800 - not using the upper 5 dB on it was not a good idea. Until I read
what the AD converter in it was incapable of - namely 16 bit linear - I
thought it was about opamp behavior.

Which leads to the real answer: do what sounds best, including do NOT add
digital gain just to get things you deliver to a mastering guy to "near
zero". Every single step of processing, be it analog or digital, comes with
a cost in terms of loss of quality and clarity, so keep the number of steps
down.


Kind regards

Peter Larsen

[Peter Larsen[_3_]]

Mxsmanic wrote:

Is it possible to "repair" clipped peaks so that they don't sound so
distorted, if by accident a recording contains them? Maybe by
smoothing out the waveform or something?

Yes, use the unclip function of your preferred audio software. With avant
garde jazz I have gotten away with 8 dB of unclipping. If you're just
"running close" some 2 dB clipping is more like what you should expect to
encounter, I prefer to unclip to smooth the edges. Generally I only unclip,
ie. do not lower the volume because there is life above 0 dB as long as you
are in 32 bit file format.

I made some tests with the learned audience of the members of an amateur
recording society to determine how much clipping would remain reasonably
inaudble on classical music, up to a couple of dB's for a couple of
milliseconds is just a blemish and not usually noticed, beyond that
unclipping gets required and not just beautification.

The single audience member that claps louder is some of the time best dealt
with by another process: remove single click, whatever it is called in your
preferred audio software.

Software that does not offer unclip and remove single click should not in my
opinion be preferred.

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

William Sommerwerck wrote:

When I recorded live, I'd ask the orchestra to hit the loudest they'd
play for that piece, then set the peak level 1 or 2dB below that.
Because when they always played a bit louder in concert.

Beware of singing wimmen!

Recording analog with dbx II was a snap. I set the pre-concert
audience noise to -10dB or -20dB (I forget which) and never had a
problem.

This does not at a glance compute, -30 to -35 seems to work well with
chamber music in a large reverberant hall, are you referring to meter
reading prior to dbx-encoding or about a vu-meter with proper lead?

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

Mxsmanic wrote:

Thanks! Isn't it possible to make digital peak meters behave like VU
meters by just averaging the peaks over a certain period of time,
though? Then you wouldn't need the analog meters.

You want Robert Orbans universal digital meter, it is a software emulation
of all the main metering standards and the price is right!

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

Trevor wrote:

"Mxsmanic" wrote in message
...
Is it possible to "repair" clipped peaks so that they don't sound so
distorted, if by accident a recording contains them? Maybe by
smoothing out
the waveform or something?

Yep, both by predictive interpolation and smoothing. FAR better to
avoid needing such drastic measures as much as possible.

Yes yes yes, but with avant garde and with wimmen you never really know and
while it is correct that the original waveform can not be recreated it still
is possible to get away with minimizing the audiblity of the imperfection.

Of course most modern music is deliberately clipped to hell and back
after recording/mixing, so it doesn't matter much if it's clipped a
bit in the recording process anyway! :-(

I tested Magix's software home studio a couple of years ago. They have a
quite nice triband compressor in it. Amazing how much cleaner it sounded
when used on a 32 bit file that was then normalized down afterwards compared
to how it sounded on a 16 bit file. Of course, using it on a 16 bit file
resulted in commercially perfect looking loud bricks.

Trevor

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

Frank Stearns wrote:

Is it possible to "repair" clipped peaks so that they don't sound so
distorted, if by accident a recording contains them? Maybe by
smoothing out the waveform or something?

Yes, but it can be damned tedious.

Cool Edit of old and Auditions 1 to 3 do it nicely, methinks the algorith
improved between 1.5 and 2.0 and later. Dunno about 4, which got labed 5 to
fit the Creative Suite version numbering.

Now, there are some automatic clip repair tools in some of the
editing platforms, but I've not played with the newer ones. The older
ones didn't work all that well (some were just plain brain dead).

Download some demos to a sandbox machine and test them. It is nót a good
idea to install demo versions to a production machine that you may later
want to install the full version on.

Frank
Mobile Audio

Kind regards

Peter Larsen

[Arkansan Raider]

Peter Larsen wrote:
Mxsmanic wrote:

Thanks! Isn't it possible to make digital peak meters behave like VU
meters by just averaging the peaks over a certain period of time,
though? Then you wouldn't need the analog meters.

You want Robert Orbans universal digital meter, it is a software emulation
of all the main metering standards and the price is right!

Kind regards

Peter Larsen


Peter, I just did a fairly quick search on the interwebbies and found
Robert Orban, but no links for the meter. Do you have that bookmarked,
by any chance?


---Jeff

[Scott Dorsey]

Peter Larsen wrote:
Mxsmanic wrote:

Thanks! Isn't it possible to make digital peak meters behave like VU
meters by just averaging the peaks over a certain period of time,
though? Then you wouldn't need the analog meters.

You want Robert Orbans universal digital meter, it is a software emulation
of all the main metering standards and the price is right!

But there is no Mac version!
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Peter Larsen[_3_]]

Arkansan Raider wrote:

Peter, I just did a fairly quick search on the interwebbies and found
Robert Orban, but no links for the meter. Do you have that bookmarked,
by any chance?

Search term: robert orban meter software =

yet another site that violates copyright and droit morale and relays usenet
while making it appear that we all are users of their not even existing bbs:

http://www.hydrogenaudio.org/forums/...howtopic=62570

actual link to meter software in post the http://www.orban.com/meter/

Bonus search result:

http://windowsvumeter.sourceforge.net/

note: not tested, but sure looks interesting.

note2: some of the time funny things that aren't funny will happen when you
patch a software meter into the audio stream in your computer, so do NOT
have your monitors turned up way loud on first test.

---Jeff

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

Scott Dorsey wrote:

Peter Larsen wrote:
Mxsmanic wrote:

Thanks! Isn't it possible to make digital peak meters behave like
VU meters by just averaging the peaks over a certain period of time,
though? Then you wouldn't need the analog meters.

You want Robert Orbans universal digital meter, it is a software
emulation of all the main metering standards and the price is right!

But there is no Mac version!

I would have thought it was a command-line version with output in roman
numerals
you would lament as non-existing.

O;-)

--scott

Kind regards

Peter Larsen

[Arkansan Raider]

Peter Larsen wrote:
Arkansan Raider wrote:

Peter, I just did a fairly quick search on the interwebbies and found
Robert Orban, but no links for the meter. Do you have that bookmarked,
by any chance?

Search term: robert orban meter software =

yet another site that violates copyright and droit morale and relays usenet
while making it appear that we all are users of their not even existing bbs:

http://www.hydrogenaudio.org/forums/...howtopic=62570

actual link to meter software in post the http://www.orban.com/meter/

Bonus search result:

http://windowsvumeter.sourceforge.net/

note: not tested, but sure looks interesting.

note2: some of the time funny things that aren't funny will happen when you
patch a software meter into the audio stream in your computer, so do NOT
have your monitors turned up way loud on first test.

---Jeff

Kind regards

Peter Larsen




Roger that. Thanks much, Peter!

---Jeff

[Peter Larsen[_3_]]

Peter Larsen wrote:

Arkansan Raider wrote:

Peter, I just did a fairly quick search on the interwebbies and found
Robert Orban, but no links for the meter. Do you have that
bookmarked, by any chance?

Search term: robert orban meter software =

yet another site that violates copyright and droit morale and relays
usenet while making it appear that we all are users of their not even
existing bbs:

Sorry, I was wrong, it appears to actually be a BBS.

Kind regards

Peter Larsen

http://www.hydrogenaudio.org/forums/...howtopic=62570

actual link to meter software in post the
http://www.orban.com/meter/
Bonus search result:

http://windowsvumeter.sourceforge.net/

note: not tested, but sure looks interesting.

note2: some of the time funny things that aren't funny will happen
when you patch a software meter into the audio stream in your
computer, so do NOT have your monitors turned up way loud on first
test.
---Jeff

Kind regards

Peter Larsen

[Mxsmanic]

Peter Larsen writes:

yet another site that violates copyright and droit morale and relays usenet
while making it appear that we all are users of their not even existing bbs:

http://www.hydrogenaudio.org/forums/...howtopic=62570

While this is unethical, it's not clear that it infringes on copyrights any
more than USENET itself (the copyright status of which has never been
clarified). And it's hard to see how moral rights would be a factor at all,
even in jurisdictions that recognize moral rights (not all jurisdictions
do--the U.S. doesn't).

[Peter Larsen[_3_]]

Mxsmanic wrote:

Peter Larsen writes:

yet another site that violates copyright and droit morale and relays
usenet while making it appear that we all are users of their not
even existing bbs:

http://www.hydrogenaudio.org/forums/...howtopic=62570

While this is unethical, it's not clear that it infringes on
copyrights any more than USENET itself (the copyright status of which
has never been clarified). And it's hard to see how moral rights
would be a factor at all, even in jurisdictions that recognize moral
rights (not all jurisdictions do--the U.S. doesn't).

Again, I was wrong, it appears to actually BE a BBS.

As for the dismal record in terms of even understanding what copyright is
the US of A has I'll us my time differently and not comment on it. The USA
did nowever sign the ... Berne Convention, hoping I remember it name
correctly.

DO look Droit Morale up and read up on what it is, do NOT use any other
term, it is not just "moral rights", it is more like an _ideal_ _obligation_
for the user of rights given, be it for free or paid for.

Kind regards

Peter Larsen

[Mxsmanic]

Peter Larsen writes:

DO look Droit Morale up and read up on what it is, do NOT use any other
term, it is not just "moral rights", it is more like an _ideal_ _obligation_
for the user of rights given, be it for free or paid for.

Yes, it _is_ "moral rights," which is simply the translation of "droits
moraux." Moral rights allow a copyright holder to allow or object to the way
in which a work is used. Sometimes they can be sold or transferred, sometimes
not, it depends on the jurisdiction.

For example, in France, which has copyright laws that veer into absurdity in
some cases (even as the French routinely infringe on the more mundane of these
laws), an architect can object to any change in the use of a building designed
by him, that is, if he designed a building as a live theater, he can object to
it being converted to a cinema. That's part of his "moral rights." It's kind
of a dorky concept.

[Mark]

The single audience member that claps louder is some of the time best dealt
with by another process: remove single click, whatever it is called in your
preferred audio software.


that's one example where IMHO clipping is acceptable and even
preferable to the alternatives.

Mark

[Peter Larsen[_3_]]

Mark wrote:

The single audience member that claps louder is some of the time
best dealt with by another process: remove single click, whatever it
is called in your preferred audio software.

that's one example where IMHO clipping is acceptable and even
preferable to the alternatives.

Please make the point, it is not instantly obvious. Have you tried the
suggested solution? - the reason for suggesting it is that the applause
sounds plain better when homogenous!

Mark

Kind regards

Peter Larsen

[Scott Dorsey]

Peter Larsen wrote:
Mark wrote:

The single audience member that claps louder is some of the time
best dealt with by another process: remove single click, whatever it
is called in your preferred audio software.

that's one example where IMHO clipping is acceptable and even
preferable to the alternatives.

Please make the point, it is not instantly obvious. Have you tried the
suggested solution? - the reason for suggesting it is that the applause
sounds plain better when homogenous!

I think his point is that you _can_ cut the clipped portion out with no
loss to the original signal, and even an improvement. The alternative
would be to ride the gain on applause, which in this case would be worse
than the clipping. So this is one of the very few cases where clipping is
acceptable and not a problem.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Peter Larsen[_3_]]

Scott Dorsey wrote:

Peter Larsen wrote:

The single audience member that claps louder is some of the time
best dealt with by another process: remove single click, whatever
it is called in your preferred audio software.

that's one example where IMHO clipping is acceptable and even
preferable to the alternatives.

Please make the point, it is not instantly obvious. Have you tried
the suggested solution? - the reason for suggesting it is that the
applause sounds plain better when homogenous!

I think his point is that you _can_ cut the clipped portion out with
no loss to the original signal, and even an improvement.

Ah, yes, I do not loose sleep over clipping those single claps, as may
happen if an omni-par is used close to the audience. Mostly that IS the
situation that will cause the issue.

The
alternative would be to ride the gain on applause, which in this case
would be worse than the clipping.

Agreed. Some of the time I'll end up using a special compressor-preset on
applause to get avoid having it louder than the music, even if it was so in
the room. But that does not really solve the loud clapper problem, just as
gain riding does not, it still sounds wrong and I'd rather have it smooth
and do less to it all. Doing less is almost always great.

So this is one of the very few
cases where clipping is acceptable and not a problem.

Yes, but clipping does not solve the problem of the applause not being
smooth. Yes, you could cut those single claps out, but it is faster, simpler
and less audible to "fix single click" them, in which case the vanish
inadibly.

Fix single click is a great tool, I have even gotten away with getting rid
of a book that feel from a table during a performance as well as stopping
the conductor from kicking the mic stand with the main pair when putting a
cable reel between him and the manfrotto-stand was not enough, it would have
been if he hadn't kicked it backwards tho' ... if only it had been a big
name, they are a lot easier to work with than the upstarts!

--scott

Kind regards

Peter Larsen

[Scott Dorsey]

Peter Larsen wrote:
Fix single click is a great tool, I have even gotten away with getting rid
of a book that feel from a table during a performance as well as stopping
the conductor from kicking the mic stand with the main pair when putting a
cable reel between him and the manfrotto-stand was not enough, it would have
been if he hadn't kicked it backwards tho' ... if only it had been a big
name, they are a lot easier to work with than the upstarts!

In the analogue world we just use a typewriter eraser on the master tape.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Frank Stearns]

(Scott Dorsey) writes:

Peter Larsen wrote:
Mark wrote:

The single audience member that claps louder is some of the time
best dealt with by another process: remove single click, whatever it
is called in your preferred audio software.

that's one example where IMHO clipping is acceptable and even
preferable to the alternatives.

Please make the point, it is not instantly obvious. Have you tried the
suggested solution? - the reason for suggesting it is that the applause
sounds plain better when homogenous!

I think his point is that you _can_ cut the clipped portion out with no
loss to the original signal, and even an improvement. The alternative
would be to ride the gain on applause, which in this case would be worse
than the clipping. So this is one of the very few cases where clipping is
acceptable and not a problem.

The enthusiastic crowds that show up for some of the classical events I record will
occasionally clip the AB main pair (KM183s) -- this is even with 15-20 dB of
headroom in relation to the program material.

This could rationally be expected, because even though fairly close to the stage
that main pair gives a sense of "being there" in the audience. They are omnis
afterall, with lots of folks clapping "behind" then. So the recorded applause will
be pretty loud, as it would be if you were sitting in the first few rows with all
that sound coming from behind.

But for "music program playback" applause like that is stupidly loud, so in post the
gain on the stereo pair gets ramped down 3-6 dB that split second before the
applause starts -- keeps the arc of attention on the music, and not the applause
that otherwise would hit you in the face (I also shorten applause time quite a bit
to keep the feel more suited to home playback, or even take it out completely except
at the end of the program).

The level change gets a little tricky if music and applause overlap but there are a
number of ways to do it so that the main pair level reduction is not noticed.

I'm not sure why you say this is worse, unless you mean dropping the applause level
after it starts -- that would sound pretty tacky, I agree.

In terms of applause clipping (assuming the occasional Herculian clap that clips),
these clips are truly inaudible, especially if the mix level has been dropped and
the clip(s) do(es) not propogate further. So, no worries, IMO.

Now, if you're clipping applause to the point of it sounding like tortured white
noise, that's a different story.

Frank
Mobile Audio
--

[Ty Ford]

On Fri, 20 Jan 2012 14:13:18 -0500, Mxsmanic wrote
(in article ):

Ty Ford writes:

Then buy an Alexa camera. It has 24-bit, 48 kHz audio.

The Alexa is too expensive. And my impression is that Hollywood is still
recording audio separately, albeit perhaps with some sort of electronic
synchronization these days instead of a simple clapper.

Too expensive for you, maybe.

Hollywood has been double recording for years with Nagras and whatever.

These days, depending on workflow and complexity, you can sync with or
without SMPTE.

I use a Sound Devices 744T for that.

Regards,

Ty Ford



--Audio Equipment Reviews Audio Production Services
Acting and Voiceover Demos http://www.tyford.com
Guitar player?:http://www.youtube.com/watch?v=yWaPRHMGhGA

[Mxsmanic]

Am I mistaken, or would the worst case of clipping be shaped essentially like
a square wave? That is, a nearly vertical component to the waveform, with an
instant transition to a perfectly horizontal component, followed by an instant
transition to a nearly vertical component again. That would seem to be the
most extreme case of clipping imaginable. Which in turn implies that the worst
clipping in digital audio would sound like a square wave.

[Mxsmanic]

Ty Ford writes:

Too expensive for you, maybe.

To whom else would I be referring?

[Peter Larsen[_3_]]

Mxsmanic wrote:

Am I mistaken, or would the worst case of clipping be shaped
essentially like a square wave? That is, a nearly vertical component
to the waveform, with an instant transition to a perfectly horizontal
component, followed by an instant transition to a nearly vertical
component again. That would seem to be the most extreme case of
clipping imaginable. Which in turn implies that the worst clipping in
digital audio would sound like a square wave.

Clean diode action clips up to a couple of milliseconds are not at all or
only marginally obvious but not all clipping is clean. My understanding is
that complications can include polarity reversals and burst of noise.

I have a poweramp (Audire dual 120 watts, great sound & wonderful deep and
powerfull bass) - now in repair queue with a silent right channel - that for
years added a burst of noise whenever it clipped that now silent channel,
conceivably the warranty refill of brown smoke wasn't done properly or
perhaps the designers aim for megahz bandwidth was plain silly. I had asked
for a known bandwidth reduction fix to be implemnted on it when the right
channel got new output transistors ... it either wasn't done or wasn't
enough or something else that was marginal ex works was overlooked in the
repair. Never a problem with left channel, including clipping as gently as
clipping should be.

Digital clipping is to the best of my knowledge just that, it is with analog
clipping that bets are off.


Kind regards

Peter Larsen

[Les Cargill[_4_]]

Mxsmanic wrote:
Am I mistaken, or would the worst case of clipping be shaped essentially like
a square wave? That is, a nearly vertical component to the waveform, with an
instant transition to a perfectly horizontal component, followed by an instant
transition to a nearly vertical component again. That would seem to be the
most extreme case of clipping imaginable. Which in turn implies that the worst
clipping in digital audio would sound like a square wave.


Right.

--
Les Cargill

[[email protected]]

PEter writes:
snip

Ah, yes, I do not loose sleep over clipping those single claps, as
may happen if an omni-par is used close to the audience. Mostly
that IS the situation that will cause the issue.
The alternative would be to ride the gain on applause, which in this
case would be worse than the clipping.
Agreed. Some of the time I'll end up using a special
compressor-preset on applause to get avoid having it louder than
the music, even if it was so in the room. But that does not really
solve the loud clapper problem, just as gain riding does not, it
still sounds wrong and I'd rather have it smooth and do less to it
all. Doing less is almost always great.

Agreed. If I'm using multiple omnis to capture audience
response from varying perspectives one thing I've found
works for me is to send them to a bus, maybe compress that
bus, or lower the mic with the over loud clapper a bit.
Quite application dependent, every gig is different it seems
g. I'll endeavor to anticipate and ride gain even if I
must, going for consistency of course. Liked Frank's
comments in his spin off thread.

snip again
Fix single click is a great tool, I have even gotten away with
getting rid of a book that feel from a table during a performance
as well as stopping the conductor from kicking the mic stand with
the main pair when putting a cable reel between him and the
manfrotto-stand was not enough, it would have been if he hadn't
kicked it backwards tho' ... if only it had been a big name, they
are a lot easier to work with than the upstarts!

rotfl YEah know the feeling. Singers playing with their
microphones can be frustrating too. I still recall working
with a guy in the studio whose version of fidget was playing
with his pocket change.




Richard webb,

replace anything before at with elspider

[Scott Dorsey]

Mxsmanic wrote:
Am I mistaken, or would the worst case of clipping be shaped essentially like
a square wave? That is, a nearly vertical component to the waveform, with an
instant transition to a perfectly horizontal component, followed by an instant
transition to a nearly vertical component again. That would seem to be the
most extreme case of clipping imaginable. Which in turn implies that the worst
clipping in digital audio would sound like a square wave.

_all_ clipping is to some extent shaped like a square wave. As soon as you
get even a little bit of clipping, you start getting huge amounts of odd
harmonics popping up, just like a square wave.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."