On a different note.
Log sine sweeps..
Yes and no.. it's all depending on too many factors to tell you
like this.
First off.. if you were to check ex 1 2 5 10 20 50 100 200 500
you're leaving too much space in your measurements.. you will
loose a lot of the details you need to take a good hard look at.
Generally i use a noise source that contains everything and took
snapshots of the screen for future reference.
Whether for a large sound system in a hall etc.. or for studios
we generally do it using a noise generator and a calibrated
high performance microphone.
Again .. acting in real time and taking action while you're
looking at the screen ( setting filters delays etc ) is the
conventional way.You're talking studio.. but then again .. depending
on what you see. you may not want at all to treat the signal .. but to
treat the room in itself.
Ex .. Using bass traps .. they can be tubes.. they can be cavities
within the walls etc etc
Big huge problems with the room have to be corrected for what it is .
The room is at fault.. fix the room :)
It's a complex mix.. there's tons of opinions out there
and they all have ins and outs.
You will have to make experimentation and find what's best
for your particular application.
This being said.. pizza calls :)
Ric
( the smiley guy ;)
On Fri, 23 Apr 2004 22:19:03 +0200
mrtn <martin(a)nadir.org> wrote:
hello,
i am trying to record impulse responses from
my studioequipment to correct it's phase and freq responses
with e.g. DRC and bruteFIR.
I heard the best idea would be to take log sine sweeps
for the measuring?
it would be great if someone more experienced than me
knows of a good method to create sweeps (or chirps) of equal
freq.response with linux software?!
or maybe some other tips and tricks about audio impulse response
measuring with linux?
peace
mart