GUIDE FOR SETTING UP YOUR ROOM

it’s crazy how much difference it can make. Frequencies get really directional as they get higher up, so it’s really important to set the acoustic centre of the monitors where the manufacturer recommends. [...]
An excellent tip! I hadn't even thought to look in the manual of my Neumanns yet. Instead, I relied on articles on the internet, most of which stated to place the tweeter at or even slightly below ear level. Now I've looked it up and lo and behold - Neumann state that the acoustic axis of the KH 120 A speakers is midway between the tweeter and woofer. So I have now selected the height setting on my stands that most closely corresponds to this specification. The difference is really remarkable. Everything suddenly sounds somehow more direct and the bass also comes across better. I haven't taken a REW measurement yet, but I suspect there will be a clear difference.
 
Well, I just received this nice little delivery. ;)

IMG_1001.jpeg
 
So, the panels are now all installed. The first measurement results show moderate improvements. While the decay time below 150 Hz has indeed improved quite significantly, the room modes have hardly changed at all. What has improved, however, is the midrange, which now looks somewhat flatter in the measurement results.

Spectrogram before:
before.jpg


Spectrogram after:
after.jpg


Frequency spectrum before:
Bildschirmfoto 2025-01-26 um 09.23.16.png


Frequency spectrum after:
Bildschirmfoto 2025-01-26 um 09.23.05.png
 
Last edited:
So, the panels are now all installed. The first measurement results show moderate improvements. While the decay time below 150 Hz has indeed improved quite significantly, the room modes have hardly changed at all. What has improved, however, is the midrange, which now looks somewhat flatter in the measurement results.

Frequency spectrum before:
View attachment 37261

Frequency spectrum now:
View attachment 37262
Definitely a step in the right direction. That first dip could well be a boundary reflection rather than anything modal, and it’s hard for porous absorbers to do much below 80hz anyway (resonance based treatment is needed for that). Really not looking bad at all frequency response wise.

Can you post spectrogram (and maybe waterfall) plots? They’ll show what’s going on in the room a bit more clearly, the frequency response is kind of like the sum of everything else happening over a period of time.

and when you do your measurements, I’d recommend choosing the option to do L and R sequentially, it looks like you’re doing L+R here?

this video (and part 1) is helpful

 
Yep, I added spectrograms to my post above. The improvement is much more visible there.
Yeah, looking much better there, way smoother across the board and the low end is tightened up a bit too.

I think generally it’s looking pretty decent, no major issues or anything to worry about. More broadband panels would continue to help further (if there’s any more space/desire to keep going).

Typical ceiling heights of 2.3-2.4m are going to have a cancellation at just above 70hz where our ears typically are positioned.

How’s it sounding?
 
Yeah, looking much better there, way smoother across the board and the low end is tightened up a bit too.

I think generally it’s looking pretty decent, no major issues or anything to worry about. More broadband panels would continue to help further (if there’s any more space/desire to keep going).

Typical ceiling heights of 2.3-2.4m are going to have a cancellation at just above 70hz where our ears typically are positioned.

How’s it sounding?
I could hear the tidier low end immediately. The lower mids are not quite so bloated anymore and everything sounds more direct overall. Also I had the feeling that the midrange became a little bit more prominent. Definitely an audible difference!
After the first positive listening impression, I was honestly a little disappointed when I took measurements and the lower frequency range looked almost unchanged at first glance. I thought my ears had deceived me and I had invested 1000 euros for nothing. Fortunately, the spectrogram view eventually confirmed what I had heard.
My ceiling is 2.67 metres high. However, the ceiling panelling is suspended and the room is therefore actually a little higher.
Perhaps I'll buy one or two more panels soon. But then it really should be enough, after all, the whole shebang isn't cheap. However, I have planned not to buy any other equipment such as guitars, amplifiers or similar this year, but instead to invest exclusively in my listening experience and then enjoy the equipment I already have.
 
After the first positive listening impression, I was honestly a little disappointed when I took measurements and the lower frequency range looked almost unchanged at first glance. I thought my ears had deceived me and I had invested 1000 euros for nothing. Fortunately, the spectrogram view eventually confirmed what I had heard.
Yeah this can be quite a common experience. Sometimes, things can even measure worse in the frequency graphs after spending loads of money, because as one problem gets fixed, another thing reveals itself more. For instance, there could be one frequency being boosted by one part of the room and cancelled by something else. The frequency domain might look flat, but in reality it just the sum of various things going on in the room that are giving a kind of "false positive".

The main thing is to look at the time domain and work out what is happening in the room, and to follow any improvements in the time domain. The frequency domain can show us a few things but to make sense of anything you need to consider what is causing the graph to look like that (speaker positioning, treatment, time domain etc). An ideal room is never going to measure completely flat, even in the absolute best designed rooms there'll be some unavoidable dips. As you fix certain time based issues, the boundary ones that you can't do much about will still be there.

It can also be frustrating because we want a couple of panels to make a drastic improvement, and in reality, we need SO MUCH absorption to have a significant effect, combined with the fact that as you add more treatment, they'll have less and less effect.

With your tall ceilings, it should be fairly easy to hang quite a big thick cloud (or several). If you have the means to do DIY, they are VERY easy and inexpensive to build (like a fraction of GIK prices), and you have the benefit of choosing your own sizes and materials. Once you go over 20cm or so, it's best to use less dense insulation. With your dimensions, I'd probably consider something like a 40cm thickness but with low density fibreglass. Ceiling clouds also don't take up any space that you'd use for something else so you can have pretty significant gains from it without losing much.
 
Last edited:
MirrorProfiles was right again. I redid the measurement, this time the left and right channel sequentially, and then let REW calculate the average. And lo and behold - suddenly the highs also look flat. I had already wondered why the measurements always showed a drip from 5 kHz upwards, although my hearing told me that I had sufficient treble.

sequential.jpg


It is becoming more and more apparent to me that the measures taken were a definite success. Without Ed's informative and detailed posts, I would not have been able to achieve such a good result.

Big thanks to Ed. He definitely is one of the nicest and most helpful persons I have ever met on forums.

@MirrorProfiles
You Rock GIF by BLKBOK
 
I took a few more REW measurements and let its EQ function suggest a curve to smooth out some of the last few bumps. RME provides detailed instructions on how to configure the EQ settings in REW so that they match the integrated EQ of the interface driver. The suggested values could then be easily exported as a text file and applied to the Room EQ of my RME UCX II.

Bildschirmfoto 2025-01-30 um 14.34.02.png

I just listened to some songs from a few of my favorite albums and I have to say, the end result is...

Proud Of You Yes GIF
 
Back
Top