[sdiy] IR Reverb

[Roman Sowa]

The impulse itself, in theory, contains all frequencies in the world, 
that is dense flat spectrum. Obviously it's not possible to generate 
such impulse in real life as it would have to be of infinite amplitude 
in zero time, let alone pass it through a speaker.
I think big spark discharge is used to make pretty good estimated pulse 
directly in air pressure. Also you can do step response instead of 
pulse, which is much easier to play, and then get impulse response out 
of it via simple math (I think it's 1st order differentiation).

Anyway, that's the impulse response and with really long FIR filter you 
can convolute it with any incoming signal. Convolution in time is the 
same as multiplication of both spectras together - of the impulse 
response and input signal. So whatever filtering was happening in the 
room and was recorded in the impulse response, it's copied to the 
spectrum of input audio just like you'd multiply their spetra.
Of course real-life audio convoluters (is that a proper name?) don't 
calculate couple of seconds convolution for every sampl as it takes 
enormous computing power. They split it somehow to subregions to make it 
1000 times easier but I can't remember details, it's been some time 
since I read about them. There are a few folks on this list that could 
explain in details how it's achieved.

So FIR filtering gives indeed time and color. If you'd set only one FIR 
coefficient to 1, and all others to 0, it would be simple delay. Make a 
few of them set apart - there you have early reflections. If you all 
coefficients are set to the same value, it will be (really crappy) 
lowpass filter.

I used to have fun with FIR filters in last century as it was part of my 
thesis.

Roman

W dniu 2018-02-15 o 08:42, Tim Ressel pisze:
> (note: laughter is understandable and probably mandatory)
> 
> So I just did a cannonball into the murky waters of impulse response 
> reverb. My tenuous grasp of DSP coupled with my sketchy math skills is 
> making this, well, interesting. So far I have acquainted myself with 
> linear convolution, which looks suspiciously like an FIR. Since it says 
> "impulse response" on the box, that seemed to make sense. But then I 
> stepped back and tried to imagine a reverb system as I understand it and 
> got confused.
> 
> Reverbs have two things going on: time delay and filtering. The time 
> component gives the reverb time and overall thickness of the reverb, 
> while the filtering can make the effect warmer or colder (yes, 
> oversimplified). So I am guessing the impulse response is a room 
> characterization used to color a reverb. However that seems incomplete. 
> Unless the impulse response is really long or is sets of impulse 
> responses over time.
> 
> I suspect that gurgling sound is me in over my head.
>