• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Share your in-room measurements?

er|κzvio1in

Member
Joined
May 26, 2023
Messages
74
Likes
36
Location
Europe
I set up a new miniDSP configuration, more optimized, I used the same time alignment as before but did deeper cuts in the bass and also applied PEQs between at higher frequencies. The result is more linearity in upper bass and lower mids, reduced bass and improved clarity.

Target curve I used is Harman plus 2 dB/oct slope from 80 hz downward. Pink is MMM speakers and subs time aligned, purple is MMM with PEQ applied to individual output channels, green is predicted curve with PEQ applied to linked/shared input channels.
20240326 before and after.png

20240326 before and after psychoacoustic.png
 

Vladetz

Active Member
Joined
Jul 20, 2020
Messages
132
Likes
85
Location
Russia
My R+L measurements with variable smoothing: Mission 700 (2022) in 17 sq. m. room, quite close to wall (15 cm), toed in (20 deg), no EQ
Red line is umik 1 measurement with calibration file
Green line is 5$ mic from Aliexpress



1712213388051.png
 

er|κzvio1in

Member
Joined
May 26, 2023
Messages
74
Likes
36
Location
Europe
My R+L measurements with variable smoothing: Mission 700 (2022) in 17 sq. m. room, quite close to wall (15 cm), toed in (20 deg), no EQ
Red line is umik 1 measurement with calibration file
Green line is 5$ mic from Aliexpress



View attachment 361256
So for room correction below Schroeder, the cheap mic is just as good. Is it USB or do you need an audio interface?
 

Booker

Active Member
Forum Donor
Joined
May 18, 2020
Messages
109
Likes
101
240406BAEQ.jpg
 

Booker

Active Member
Forum Donor
Joined
May 18, 2020
Messages
109
Likes
101
I bought the UMIK-1 in 2022 and made the first measurements, using REW filters. So far so good, I cannot describe the audible change, but something from the "unqed" sound disappeared. The motivation for today's measurement is the recent use of IEMs and comparing the sound of IEMs and loudspeakers. I want to get closer to the sound I hear from IEM. From 2022 I have a filter with 5 bands, from today's measurement I have a filter with 20 bands. So, let's say I am still in the learning process in this craft.

I am unsure if a subwoofer is an option since I live in a flat surrounded by neighbors. And no one has claimed against my hobby, that I would rather stay as a good guy.

My room measures 3.5 x 6 x 2.6. The speakers are placed symmetrically in the room with the longer wall. When I clap my hands I hear a reverb, so the damping could be better.
 

er|κzvio1in

Member
Joined
May 26, 2023
Messages
74
Likes
36
Location
Europe
I bought the UMIK-1 in 2022 and made the first measurements, using REW filters. So far so good, I cannot describe the audible change, but something from the "unqed" sound disappeared. The motivation for today's measurement is the recent use of IEMs and comparing the sound of IEMs and loudspeakers. I want to get closer to the sound I hear from IEM. From 2022 I have a filter with 5 bands, from today's measurement I have a filter with 20 bands. So, let's say I am still in the learning process in this craft.

I am unsure if a subwoofer is an option since I live in a flat surrounded by neighbors. And no one has claimed against my hobby, that I would rather stay as a good guy.

My room measures 3.5 x 6 x 2.6. The speakers are placed symmetrically in the room with the longer wall. When I clap my hands I hear a reverb, so the damping could be better.
With so much lack of bass, it's never going to sound like an IEM IMHO.
 

Keith_W

Major Contributor
Joined
Jun 26, 2016
Messages
2,660
Likes
6,064
Location
Melbourne, Australia
I bought the UMIK-1 in 2022 and made the first measurements, using REW filters. So far so good, I cannot describe the audible change, but something from the "unqed" sound disappeared. The motivation for today's measurement is the recent use of IEMs and comparing the sound of IEMs and loudspeakers. I want to get closer to the sound I hear from IEM. From 2022 I have a filter with 5 bands, from today's measurement I have a filter with 20 bands. So, let's say I am still in the learning process in this craft.

I am unsure if a subwoofer is an option since I live in a flat surrounded by neighbors. And no one has claimed against my hobby, that I would rather stay as a good guy.

My room measures 3.5 x 6 x 2.6. The speakers are placed symmetrically in the room with the longer wall. When I clap my hands I hear a reverb, so the damping could be better.

Don't bother clapping your hands as a test for reverb. That only tests the reverb in a narrow band of frequencies. It's for those plebs who don't own microphones ;) You have a UMIK-1 and REW, you can do much better. In fact, you already have the measurements, all you need to do is look at it. Click on the RT60 tab, Energy-Time Curve, spectrogram, and waterfall. That will tell you much more than a hand clap would.

Re: subs. Yes, annoying your neighbours can be an issue. When I lived in an apartment and bought my first stereo, I spoke to my neighbours and told them that I want to test how loud my speakers will go before they can hear them, so that I can be a considerate neighbour and avoid disturbing their peace. The idea was received very well. I picked an evening on a weekend, and gave them my phone number. I asked them to ring me when they could hear my music. Then I turned up the volume every few minutes and waited for the phone call. It turned out that I could actually turn it up pretty loud.

It helped that I was using some strategies to avoid noise transmission. The biggie is sealing as many air gaps as you can. Close the windows and throw clothes against the door frame (a lot of sound leaks out the door from the gaps, especially at the top and bottom of the door). Consider installing self-adhesive foam gaskets on your door frame.
 

boxerfan88

Senior Member
Joined
Oct 23, 2018
Messages
401
Likes
444
Thought to share my measurements...

Over the past couple of weeks, I did a repositioning of subwoofer from the front-right corner to the middle of the front wall, plus widen the L-R speaker position so as to get closer to the equilateral triangle, and also push back the L-R speakers closer to the front wall.

This chart shows the after-EQ frequency response @ MLP.

EQ34 LR (var).png



The next chart shows the room ringing - before EQ & after EQ.

Spec Combo LR.png


The system sounds really good now.

.
 

boxerfan88

Senior Member
Joined
Oct 23, 2018
Messages
401
Likes
444
May I get some advice about the chart below:

Impulse LR.png


To figure out roughly where that 5.1ms reflection is at:
  • Speakers -- MLP = 2.6m
  • Impulse disturbance = 5.1ms = 1.75m
The Speakers--reflection_surface--MLP path is 4.35m. Is this calculation correct?

.
 

er|κzvio1in

Member
Joined
May 26, 2023
Messages
74
Likes
36
Location
Europe
Thought to share my measurements...

Over the past couple of weeks, I did a repositioning of subwoofer from the front-right corner to the middle of the front wall, plus widen the L-R speaker position so as to get closer to the equilateral triangle, and also push back the L-R speakers closer to the front wall.

This chart shows the after-EQ frequency response @ MLP.

View attachment 361964


The next chart shows the room ringing - before EQ & after EQ.

View attachment 361965

The system sounds really good now.

.
Your y axis scale is too zoomed out, general guideline is to show a range of 50 dB
 

er|κzvio1in

Member
Joined
May 26, 2023
Messages
74
Likes
36
Location
Europe
May I get some advice about the chart below:

View attachment 361966

To figure out roughly where that 5.1ms reflection is at:
  • Speakers -- MLP = 2.6m
  • Impulse disturbance = 5.1ms = 1.75m
The Speakers--reflection_surface--MLP path is 4.35m. Is this calculation correct?

.
I'm no expert but I'd say with the peak on approx. 0 (because of acoustic reference timing signal used) representing 2.6m, then the reflection is simply 1.75 meters. But the path of the reflection is triangular, so depending on the angle it will be closer than 1.75.
 

Keith_W

Major Contributor
Joined
Jun 26, 2016
Messages
2,660
Likes
6,064
Location
Melbourne, Australia
May I get some advice about the chart below:

View attachment 361966

To figure out roughly where that 5.1ms reflection is at:
  • Speakers -- MLP = 2.6m
  • Impulse disturbance = 5.1ms = 1.75m
The Speakers--reflection_surface--MLP path is 4.35m. Is this calculation correct?

.

Distance = speed of sound * time in seconds. So for 5.1ms, it's 343 * 5.1 / 1000 = 1.75m. Your calculation is correct.

A better view can be gained in the Energy Time Curve (ETC). You should be able to view same the 5.1ms reflection there AND view the sound power of that impulse. A reflection in < 20ms is bad. But it is not an issue if it is > 15dB below the impulse.
 

napfkuchen

Senior Member
Joined
Mar 9, 2022
Messages
347
Likes
477
Location
Germany
Did some measurements of my desktop system (GLM on):
At 42 Hz there is a black hole and RT-60 shows serious "bang-boom-bang" below 60 Hz. But there is only so much room treatment you can put in a small (4 x 3,6 x 2,65m) room ... :)

frequency response.png


RT-60.png


waterfall.png



energy.png


Impulse response shows a very early reflection from the desk (2m x 0,95m):

Impulse Response 1.png



After covering the desk with pillows the reflection at 600 ms is gone:

Impulse Response 2.png


In practice it does not make a difference for me though.
 
Last edited:

boxerfan88

Senior Member
Joined
Oct 23, 2018
Messages
401
Likes
444
I was reading up on SBIR and about the +6dB front wall gain.

I revisited the "before EQ" REW measurement ... and was pleasantly surprised by the uncanny match.

"Theory on the left." -|- "REW measurement on the right."

sbir-3.png



[ Edit: fixed Y-axis scale to 50dB ]

.
 
Last edited:

er|κzvio1in

Member
Joined
May 26, 2023
Messages
74
Likes
36
Location
Europe
I was reading up on SBIR and about the +6dB front wall gain.

I revisited the "before EQ" REW measurement ... and was pleasantly surprised by the uncanny match.

"Theory on the left." -|- "REW measurement on the right."

6d-B-SBIR2.png


.
Nice finding, 2 buts:

1. like I commented before: please fix the scale of your y axis. It's common practice to show a range of 50 dB. Your graph is zoomed out which makes the response look nicer than it really is and makes it more difficult to look at details.

2. Front wall gain happens at lower frequencies: like the graph you've shown on the left, it's a bump in the lowest frequencies. In your measurements the bump is in the low mids, around the Schroeder frequency. SBIR typically has a sine wave shape with regular round peaks and valleys. You can identify them by averaging multiple measurements over a wide range of positions because SBIR is independent from the listening position unlike room modes. For SBIR I find this calculator insightful: http://tripp.com.au/sbir.htm
 
Last edited:
Top Bottom