pouët.net

Revision 2016 hall impulse response

category: parties [glöplog]
you guys are crazy good...
added on the 2016-04-07 22:00:14 by farfar farfar
Fucking ace stuff! Didnt know about it. :)
added on the 2016-04-07 22:59:42 by Salinga Salinga
srsly, this is such awesome stuff to add natural sounding reverb to tracks. :)

BB Image
added on the 2016-04-07 23:12:49 by Salinga Salinga
jco: Yes, because a) to reach the back end of the hall the arrays would have to be WAY bigger (-> WAY more expensive), and b) it's perfectly fine if there are tables where the music is a bit quieter anyway :). So the arrays were calculated to reach up to the middle.
added on the 2016-04-08 00:51:24 by kb_ kb_
kb: cool, was just wondering whether it was intended or not. I was in the back area of the hall this time, having conversations there worked well. listening to compos and concerts was great too, and the ear bleed factor was appropriate. so, good job with the PA! :)
added on the 2016-04-08 11:41:26 by jco jco
Quote:
in the empty Revision hall.


so..this is to demo how your prod would sound in the hall with no audience.
=D
added on the 2016-04-08 12:26:28 by 1in10 1in10
jco: check out this http://www.soundonsound.com/sos/Sep02/articles/sintefex2000.asp and check out also the nebula stuff from audio-acoustica, its very interesting and gets amazing resullts in the latest Core8 release.
Not that it didnt get amazing results before, now the compressor is behaving much better.
added on the 2016-04-08 13:23:04 by leGend leGend
Awesome, thanks!
added on the 2016-04-08 14:26:42 by numtek numtek
Convolution with impulse responses is great, you can do stuff that you couldn't really do in practice or couldn't afford. Like for example, if you convolve with a carefully constructed single-sample impulse response that I heard is called Identity or Delta or something in technical terminology, you can simulate having a completely linear and transparent, perfect audio system, which is not even possible to do in the real world!! Anything you put in, comes out _exactly_ the same! This is like a dream come true for high-end audio enthusiasts. Digital signal processing is so great. I'm thinking of purchasing a high-quality impulse response system from a company called Identity Digital Intelligent Original Technologies, but I'm saving up for it, it costs at least $1000. But it's a small price for perfect audio.
added on the 2016-04-08 16:52:33 by yzi yzi
Haha, what a great idea! Now I want to do one for the Solskogen venue as well :)
added on the 2016-04-09 14:58:05 by gloom gloom
gloom: And the Darklite van? :-P
added on the 2016-04-09 18:15:50 by Sesse Sesse
On second thought, what I want is an impulse response of the PA and/or the Darklite van as measured at the organizer sleeping area. And at the nearest neighbor down the road.

Plus a breathalyzer on the master fader. :-)
added on the 2016-04-09 18:26:01 by Sesse Sesse
This will prove to be super useful ! thanks for the good job!
added on the 2016-04-10 12:51:17 by alkama alkama
... but please dont master a track/demo, so it sounds good at revision but sucks on anybody elses stereo. ;)
added on the 2016-04-11 00:10:31 by Salinga Salinga
No problem with that as long as it's a party version. ;)
Interesting. Can you post the unenhanced recording, too? I am thinking some brutal mastering optimization here :-)
added on the 2016-04-11 03:19:26 by wertstahl wertstahl
jco: this is HOT <3
added on the 2016-04-12 22:00:10 by argasek argasek
@wertstahl: I didn't do anything to it but cutting and fading to make it usable. there were more recordings, i selected the one with the least noise (orgas yelling and stuff). for fun, and expecting it to not work, i tried denoising (fft noiseprint based). as expected, it sucked. i also considered eqing the response, but then left it as homework for you guys ;) let me know if you manage to improve it and how.
added on the 2016-04-12 22:37:56 by jco jco
@absence: I bumped into Volterra series in something different and read up a bit on them, but estimating them well seems to be a nontrivial amount of work; do you know if there's readily available software for doing that? From what I understand, your input signal is typically (known) Gaussian white noise and not a frequency sweep, but the estimation process appears… complicated.
added on the 2016-04-27 12:59:02 by Sesse Sesse
Has anyone been able to come up with a more accurate version of the response?
added on the 2017-02-09 17:40:00 by netpoet netpoet
or at least the 2017 update!

login