Originally Posted by markus767
Back to parametric EQ. These filters introduce their own ringing. If it matches exactly the ringing of the room response, modal decay will indeed be shorter. It has to be shown that this is how it works in most cases. I've yet to see proof for that. In my experience the opposite is true. While peaks are flattened, the ringing is still the same but at a lower absolute level. These two aspects improve the perceived sound quality considerably. Nevertheless, modal decay time isn't reduced. Only absorption (passive or active) can do that.
You have to measure this right. Otherwise it is easy to arrive at all kinds of conclusions from it not doing anything to what you say. There is a precise set of steps you need to follow to get there from measuring the noise floor first to setting the graph parameters in REW. If you do that, you get results like this when a single EQ is applied at 53 Hz:
The brown is without that single EQ, the green with. See how the frequency response was flattened at that point and the time domain improvement came with the ride. While the post-EQ ringing dies into the noise floor, the brown, non-EQ keeps going.
So that you don't think that was a unique example, here is a second filter applied to the same set up at 140 Hz:
Once more the filter improved the frequency response and shortened the effective ringing in time domain. Let's add even more filters:
See how it consistently does that? And this is all with casual selection of filter parameters by eye for a quick analysis.
As I show in that article, here is what happens with that single 53 Hz filter when graph settings are wrong:
Now it appears that we have only fixed the frequency response and not "time." Obviously we have done both per other graphs I showed above. The key difference is proper configuration and knowing what is being displayed.
You can read the full detail and how to set up REW properly by clicking on the first link in this google search: https://www.google.com/search?q=site%3Awhatsbestforum.com+time+frequency+amirm&rlz=1C1SNNT_enUS374US375&oq=site%3Awhatsbestforum.com+time+frequency+amirm&aqs=chrome.0.57j58&sourceid=chrome&ie=UTF-8