• 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?

Joined
May 5, 2018
Messages
88
Likes
51
To me it looks very lean, the sub should be ~10dB hotter (after the 47Hz peak has been flattened). Now you have more energy in the mids than in the bass region.
It doesn't sound lean in room at all. Here is a MMM L/R, which fills in that 60-90 Hz region a bit more. The 47.6 Hz peak is still there.
Edit: This is with pink noise at ~ 71 dB(C), which is close to my normal listening loudness.
2022_01_16 MMM 8030 1092.png


Applied 3 cuts to flatten the worst of the peaks:
2022_01_16 MMM 8030 1092 EQ Roon.png
 
Last edited:

Eetu

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2020
Messages
763
Likes
1,180
Location
Helsinki
It doesn't sound lean in room at all. Here is a MMM L/R, which fills in that 60-90 Hz region a bit more. The 47.6 Hz peak is still there.
Edit: This is with pink noise at ~ 71 dB(C), which is close to my normal listening loudness.
View attachment 179470
Yeah, looks better (partly due to the slope/scale) but I would still try shelving the bass up. OTOH if you feel the bass is not lacking, don't worry about it.
IMG_20220116_221756.png
 

Erici

Active Member
Forum Donor
Joined
Jun 17, 2020
Messages
176
Likes
258
Location
Seattle Area
It doesn't sound lean in room at all. Here is a MMM L/R, which fills in that 60-90 Hz region a bit more. The 47.6 Hz peak is still there.
Edit: This is with pink noise at ~ 71 dB(C), which is close to my normal listening loudness.
View attachment 179470

Applied 3 cuts to flatten the worst of the peaks:
View attachment 179473
The high frequencies look way too low level. Using MMM you need to change a setting and then remeasure. I had the same issue at first.
Click on the RTA icon. Click on the wheel icon in the upper right corner of the RTA screen. Click on appearance icon. Click on the Adjust RTA levels box. Then remeasure.
It would also help if you change the scale of your SPL level for the graphs that you post. The scale should be 50 dB. For most people that seems to be around 40-90.
 

Eetu

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2020
Messages
763
Likes
1,180
Location
Helsinki
Are you thinking like a low shelf from 200 down and then pull those peaks down?
I would start by adjusting the gain on the sub higher (and turning off the -2dB bass switch?) while still keeping that 47Hz filter in place. Give it a go and remeasure. Btw did you take measurements when experimenting with the different phase settings?
 
Joined
May 5, 2018
Messages
88
Likes
51
I would start by adjusting the gain on the sub higher (and turning off the -2dB bass switch?) while still keeping that 47Hz filter in place. Give it a go and remeasure. Btw did you take measurements when experimenting with the different phase settings?
I did take the measurements while adjusting the phase settings. This was the flattest I could get the crossover region. I have the MMM settings wrong. I need to redo them. I had REW set to spectrum not 1/48 FFT. Will remeasure and scale differently.
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,739
Likes
16,171
It doesn't sound lean in room at all. Here is a MMM L/R, which fills in that 60-90 Hz region a bit more. The 47.6 Hz peak is still there.
Edit: This is with pink noise at ~ 71 dB(C), which is close to my normal listening loudness.
View attachment 179470

Applied 3 cuts to flatten the worst of the peaks:
View attachment 179473
That MMM has almost 30 dB less highs than low mids (also compared to your sweep measurements which look correct), did you use in the RTA options "spectrum" instead of "RTA 1/48 octave"?
 
Joined
May 5, 2018
Messages
88
Likes
51
That MMM has almost 30 dB less highs than low mids (also compared to your sweep measurements which look correct), did you use in the RTA options "spectrum" instead of "RTA 1/48 octave"?
Yup. That's my bad. This should be correct now. RTA 1/48 octave, not spectrum. 50 dB range from 40-90 dB.
MMM correct RTA not Spectrum.png
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,739
Likes
16,171
Yup. That's my bad. This should be correct now. RTA 1/48 octave, not spectrum. 50 dB range from 40-90 dB.View attachment 179507
That looks correct now and too "cold" (= opposite of "warm" tonal tuning), if placing the loudspeakers somewhere else is not an option I would recommend using 1 or 2 positive low shelf filters to bring the response below 300 Hz to a similar level and a negative PEQ for the 45 Hz mode:
1642375579397.png


If you upload the mdat file I can also create those filters for you.
 
Joined
May 5, 2018
Messages
88
Likes
51
That looks correct now and too "cold" (= opposite of "warm" tonal tuning), if placing the loudspeakers somewhere else is not an option I would recommend using 1 or 2 positive low shelf filters to bring the response below 300 Hz to a similar level and a negative PEQ for the 45 Hz mode:
View attachment 179509

If you upload the mdat file I can also create those filters for you.
Awesome. Thanks. Attached. Fresh set of MMMs.

Moving speakers is difficult, but not impossible.
2022_01_16 MMM 8030 1092 LR.png
 

Attachments

  • 2022_01_16 MMM Genelec 8030 1092.mdat.zip
    17.4 KB · Views: 44

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,739
Likes
16,171
Awesome. Thanks. Attached. Fresh set of MMMs.

Moving speakers is difficult, but not impossible.View attachment 179516
Here is a minimal correction with maximum +4dB positive filter for your wide bass dip, not to reduce your max SPL too much:

1642379120626.png


Would consider though moving the loudspeakers too in the future as its still quite bass shy.
 

Attachments

  • 2022_01_16 MMM Genelec 8030 1092.zip
    24.7 KB · Views: 51
Last edited:

dominikz

Addicted to Fun and Learning
Forum Donor
Joined
Oct 10, 2020
Messages
803
Likes
2,626
Thanks flipflop, the question for me a this point is how, not what.
Hello, perhaps this post might also be useful.
Personally I find MMM a more user-friendly in-room measurement baseline for room correction EQ (vs the single-point sweep measurement). IMHO it is much easier to derive an appropriate target curve with MMM measurement as the resulting curve will better match anechoic PIR in HF, while still capturing the room resonances in LF.
Good luck! :)
 

MCH

Major Contributor
Joined
Apr 10, 2021
Messages
2,581
Likes
2,197
Hello, perhaps this post might also be useful.
Personally I find MMM a more user-friendly in-room measurement baseline for room correction EQ (vs the single-point sweep measurement). IMHO it is much easier to derive an appropriate target curve with MMM measurement as the resulting curve will better match anechoic PIR in HF, while still capturing the room resonances in LF.
Good luck! :)
Thank you Dominik, i will certainly do. The measurements i shared are actually an average of 3 mic positions, but will try to do it properly next time i try (btw, the three measurements were very similar)
 
Last edited:

Bill Brown

Member
Joined
Sep 25, 2021
Messages
83
Likes
87
Probably a dumb question with an easy answer, but I can't figure out. Wanting to start doing MMMs with REW on a Mac.

- I have no problem doing conventional measurements, UMIK to laptop to RME ADI DAC
- Open "RTA," copied settings I have seen people post (I think):
1642435340251.png

- Hit record button, no pink noise, mic working as I can see/measure the noise floor of the room
- Clicked on every option I can find to find the noise type options, but can't find anything that looks like the noise option window I have seen screenshots of (maybe this is the problem?)

Any ideas? Thank you,

Bill
 

dominikz

Addicted to Fun and Learning
Forum Donor
Joined
Oct 10, 2020
Messages
803
Likes
2,626
Probably a dumb question with an easy answer, but I can't figure out. Wanting to start doing MMMs with REW on a Mac.

- I have no problem doing conventional measurements, UMIK to laptop to RME ADI DAC
- Open "RTA," copied settings I have seen people post (I think):
View attachment 179680
- Hit record button, no pink noise, mic working as I can see/measure the noise floor of the room
- Clicked on every option I can find to find the noise type options, but can't find anything that looks like the noise option window I have seen screenshots of (maybe this is the problem?)

Any ideas? Thank you,

Bill
You can start the periodic pink noise playback from the "Generator" button in REW main window (so you need both the Generator and RTA windows in parallel). More details in this post. Hope it helps!
 

Bill Brown

Member
Joined
Sep 25, 2021
Messages
83
Likes
87
That was the simple answer I was looking for. Lordy, not feeling terribly bright at the moment. Oy. :)

Thank you,

Bill
 

dominikz

Addicted to Fun and Learning
Forum Donor
Joined
Oct 10, 2020
Messages
803
Likes
2,626
That was the simple answer I was looking for. Lordy, not feeling terribly bright at the moment. Oy. :)

Thank you,

Bill
Glad it helped! And don't put yourself down - learning one's way around specialized SW usually takes a bit of time :)
 
Joined
May 5, 2018
Messages
88
Likes
51
If these are 8030a + 1092a it's not possible to simply increase the sub level relative to the 8030s? You should be getting good output down to below 30Hz. Am I missing something here?
I can increase the level. However, then I start to localize the sub. I’m downloaded a trial of Dirac to play around with this afternoon as well.
 

Reibekaese

New Member
Joined
Jan 17, 2022
Messages
2
Likes
5
Location
Germany
Hey everyone, I thought as my first post on the forum I'll share some in room measurements I just took.
This might be of interest since I know there is some interest in ME-Geithain speakers here, but not much data to discuss.
Anyway, the measurements are taken in the living room music/tv setup with a RME-ADI-2 DAC going into a PAS 2002PCA Amp which feeds a Pair of Geithain ME25s. So the simple 5" bookshelf, the smallest and cheapest speakers from their line-up. Main reason for the measurements was to develop a 5 band EQ for the adi.

Measurements are an average of 5 measurement points around the main listening position executed with an Earthworks M23 into my RME Babyface. Some time ago i calibrated this measurement chain with a proper calibrator (Larson Davis CAL200) at my workplace, so also the absolute level should be reasonably accurate (but definitely not class 1, since calibration was some time ago and with different temperature).

First set of data is the averaged frequency response at the MLP with psychoacoustic smoothing, comparing without and with EQ, averaged for both speakers.
220116_absH_MLP_both.png


The room has minimal acoustic treatment (one low-mid frequency trap in a corner) and normal furniture for a living room. I think out of the box the resonse is OK and most of the filtering I implemented was to optimize the low-mids. What I find curious is the small wide-Q peak centered around 4 kHz. I recall Mr. Kiesler saying, that he thinks speakers should have a wider radiation around 4 kHz, since our ears tend to rate the direct sound at those frequencies higher compared to the reflected sound and thus, speakers without a widened 4 kHz dispersion will sound too close and present: German_Interview_MEGeithain.
So that small peak seems to stem from a neutral on-axis with wide dispersion at 4kHz, but to say for sure further measurements would be necessary. Interestingly enough I found the sound to be more pleasing with the peak knocked down a bit.

Let's have a closer look at the response:
220116_absH_lr_modal.png
This time 1/12th octave smoothing and individual responses for both speakers. Biggest take-aways for me: the left speaker definetely suffers from sitting in an actual bookshelf, compared to the right speaker, which is stitting on a proper stand. The low-mid dips and peaks are more pronounced on the left speaker. And the next thing: the REW Roomsim has a reasonable correlation for the lowest frequencies, but from above ~ 100 Hz, where the absorber and the furniture start to absorb some energy, the actual measured response looks a bit better than the simulation. I didn't find a way to implement frequency dependent absorption in the REW roomsim, which would make it a really(!) powerfull tool.

One last bit of information I wan't to drop: I can't easily take free-field measurements of the speakers, but I tried to measure the distortion profile. Measurement was taken at 31,6 cm from the grille, the level was already compensated to 1 m in the attatched figures. I set the level to be around 90 dB to compare to the company specs and found a reasonable correlation. Btw. is there an easy way to have REW display distortion % with linear y-Axis? I didn't find the time to export and plot with other software.
220116_distortion_dB.png220116_distortion_percent_comparision.png

Cheers.
 
Top Bottom