aligning with the schroeder integral is in essence attempting to have an even, exponentially decay field.
you'll notice you still have lots of little 'hot' (in gain) specular reflections poking up.
modeling your ETC to look like some ideal response is not necessarily what you're after.
you're doing a few things.
0) you're now actually able to physically measure and SEE just how energy is impeding and decaying with respect to a listening position. for example, you said you could 'hear' that flutter between the front and rear wall with regards to the center speaker. now you have measurements showing the anomaly, and will have measurements once you fix the issue. you now have the visibility.
1) verifying that your broadband traps are truly (measured) attenuating the specular reflections of which they are designed to do. this includes VERIFYING with the ETC that the reflection is indeed attenuated to the appropriate levels (eg, if you had a design requirement to keep everything -20dB down)...and that the panels are attenuating this reflection across the entire listening position.
2) you are able to actually see how the specular energy behaves and DECAYS within your room.
3) you can see the effect that a diffuser has at breaking up a loud (measured) specular reflection in the ETC (and turning it into a diffused energy return, decaying with time)... this is why i always ask what 'tool' people use to show what their diffusers are doing within the room - the ETC is the tool to do just that!
4) on some of your graphs, it looks like there is some very early energy (like 1-3s after the original signal)
compare that with this where it is a clean drop-off after the original signal (besides that reflection around 8ms):
you would likely want to hunt this down...again, you need to zoom in further (0-50ms) to have some resolution.
now, generally, you attenuate these early reflections for a length in time (ISD-gap) - which removes the time-smearing of early reflections when the brain is unable to distinguish an early reflection as being a distinct, separate signal (echo). since these reflections are attenuated, then they are also NOT combining at the listening position with the original signal to cause constructive and destructive interference (comb-filtering).
if you were to remove all the panels within your room and take a frequency response in the specular region, you would see this comb-filtering. as you attenuate the early reflections, this will diminish. this is why you treat specular frequency response issues with regards to the time-domain with the ETC!
it's a bit backwards but once you see the reason for those freq-response anomalies (reflections + summation with the original signal at the listening position), and you remove (attenuate) these early reflections with respect to the time-domain and the ETC...then you are using the time-domain to clear up the frequency-domain. make sense?
now, when you attenuate these reflections, you're essentially creating an 'effectively anechoic' space for that length in time. it isn't anechoic, but it is effectively anechoic as when the energy is below -20dB or more, you will not be able to hear it (below detection threshold level). now, if the goal was to create a dead room then you would simply attenuate all of this energy - but the goal is not to create a dead room!
once this 'effectively anechoic' (ISD) time period is over (of which whose length gives the perception of the size of the room), you want to RE-INTRODUCE the energy back to the listening position. hence, you are terminating the ISD (the anechoic period is now OVER - and energy is now re-impeding the listening position). when energy is reintroduced either by a loud specular reflection or a loud diffused return, it will reinforce the localization cues of the direct energy (your brain will lock on
to the direct signal from the speakers). now, since this is a small acoustical space - the remaining energy bouncing around the room still has a direction component - which can confuse the brain (eg it's hearing 'loud' specular reflections/signals from directions OTHER than the speaker as that energy continues to bounce around the room). the reason why the termination and a diffused
decaying sound-field is important is to remove the localization cues from this energy.
Originally Posted by SAC
The termination of the ISD does several things as well psycho-acoustically. It provides the sense of liveliness to the listening space. Second, it aids in localization by 'removing' the localization and tonality shifting cues from later arriving signals - causing the focus and localization to lock onto the direct signal. And the later arriving exponentially decaying diffuse soundfield adds to the sense of space/size to the listening environment.
re-read some of the commentary in post #25:https://www.avsforum.com/avs-vb/showt...8#post20832678
pick a 'hot' (high gain) specular reflection on the ETC. do the string test and trace out what boundary or object that energy is incident from.
for example in this measurement:
what is that hot reflection at ~60ms??
also, why are you now seeing energy at around 6-8ms *after* you added the sheets of OC703?
another *gotcha* that you could be taking away from this exercise is the reflective component of OC703 (eg at high grazing angles). many think that OC703 will absorb regardless, but the ETC will detail you if you are getting reflections off of your rigid fiberglass! is that the source in the case of the above chart @ 6-8ms?