Convolution Reverb q

Archived from groups: rec.audio.pro (More info?)

In article <42fc28d3$0$32247$5a62ac22@per-qv1-newsreader-01.iinet.net.au>,
Karl Engel <karlengel@excite.com> wrote:
>Does the SPL of the impulse have to be about the same as the max SPL of the
>sound to which reverb is to be added? If not, won't the reverb be
>unrealistically great?

If the reverberation process is linear, it doesn't matter.

>Eg clapping your hands (as an impulse) in a shopping
>centre full of people might produce little discernible echo, but firing a
>gun might (before the panic!).

If the shopping mall is closed and the air conditioning is turned off,
the reverb tail from the clap and the reverb tail from the gunshot should
have the same intensity ratio between them as the clap and gunshot do.
--scott

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

Archived from groups: rec.audio.pro (More info?)

The impulse should not have an SPL, in should be normalised before you use
it.
Only during recording it will have some SPL, and a higher SPL means better
signal-to-noise ratio.

For this reason, it is better to not record an impulse at all, but to record
a long wide bandwidth signal such as a frequency sweep, and to deconvolve it
to retrieve the impulse response with high signal-to-noise ratio.


"Karl Engel" <karlengel@excite.com> schreef in bericht
news:42fc28d3$0$32247$5a62ac22@per-qv1-newsreader-01.iinet.net.au...
> Does the SPL of the impulse have to be about the same as the max SPL of
the
> sound to which reverb is to be added? If not, won't the reverb be
> unrealistically great? Eg clapping your hands (as an impulse) in a
shopping
> centre full of people might produce little discernible echo, but firing a
> gun might (before the panic!). But wouldn't using the gunshot as an
impulse
> add too much reverb to say, speech if you wanted to replicate the sound of
> that space?
>
>

Archived from groups: rec.audio.pro (More info?)

M&M wrote:
> The impulse should not have an SPL, in should be normalised before you use
> it.
> Only during recording it will have some SPL, and a higher SPL means better
> signal-to-noise ratio.
>
> For this reason, it is better to not record an impulse at all, but to record
> a long wide bandwidth signal such as a frequency sweep, and to deconvolve it
> to retrieve the impulse response with high signal-to-noise ratio.

When you listen to one done this way, it sounds remarkably
like a gunshot, as it should. :-)


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein