• 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!

UMIK-2 First time Room EQ measurements.

  • Thread starter Deleted member 35357
  • Start date
OP
D

Deleted member 35357

Guest
Encountered a issue with the Generator when trying to have it output on only the Right speaker for RTA to do the MMM at higher SPL
Right channel is the one with shared sub.
It stops the output after a few seconds when I start it, just goes silent on the output. It just doesn't playback any sound even if it says it's doing it. Have to change output to get it working but the same thing just happens after a few seconds.

Had no issue on left channel. Is it the "stop if heavy input clipping occurs" happening? Or some other configuration issue?

EDIT: Changed the value from -12,00 dBFS to -20,00 and it didn't occur again, didn't stop the output with the -20,00 dBFS setting.
So a clipping issue? But the volume isn't that high though.
 
Last edited by a moderator:

Kachda

Addicted to Fun and Learning
Forum Donor
Joined
May 31, 2020
Messages
903
Likes
1,578
Location
NY
Did MMM according to the settings provided earlier with pink noise, but they are at to low levels. (Will have to redo tomorrow at higher SPL, can't do higher than this at the moment without complaints)

This is with sitting in chair holding the Mic in face area moving it around and then being behind the seat and holding the mic around the head area in generality as I moved it around.
Somehow it was averaged 500+ measurements, it did them quicker than I thought.


View attachment 250518
Are you applying any kind of smoothing?
 
OP
D

Deleted member 35357

Guest
Are you applying any kind of smoothing?
Just used the settings below the first OP post as was suggested to be used. Might have forgot to have any smoothing for the screen-cap I did.

I'm doing new measurements right now, but having clipping issues it seems with the output from the Generator.
So might be better to wait me figure out this before anyone tries anything with yesterdays measurements.

Though they seem quite similar to what are already provided with the lower SPL measurements. (though enough diffrences with the changed SPL)
I get clipping it seems with any higher output. Might have something not configured correctly as the volume isn't that high.

Managed to get a measurement without clipping for Left & Right speakers:
221217 UMIK-2 MMM Left and Right Speakers.png


EDIT: I updated the measurements download with the new data included.
 
Last edited by a moderator:
OP
D

Deleted member 35357

Guest
Looking at the new measurements I did I have a ascending frequency response versus the wanted reference suggested
The single point measurements over emphasized the bass response measurements. While the MMM shows I have too high 10KHz-16KHz+ peaks.

7-Band EQ:
(-6dB Attenuation)
63Hz -8dB (-14)
160Hz 0dB (-6)
400Hz -4dB (-10)
1Khz 0dB (-6)
2.5KHz +6dB (0)
6.3KHz -6dB (-12)
16KHz +6dB (0)

The analogue EQ can be adjusted +-10dB per channel with 0dB or -6dB attenuation.

Should probably raise 6.3KHz and lower the 16Khz channels.
 
Last edited by a moderator:

ThatSoundsGood

Active Member
Forum Donor
Joined
Aug 12, 2022
Messages
147
Likes
114
Looking at the new measurements I did I have a ascending frequency response versus the wanted reference suggested
The single point measurements over emphasized the bass response measurements. While the MMM shows I have too high 10KHz-16KHz+ peaks.

7-Band EQ:
(-6dB Attenuation)
63Hz -8dB
160Hz 0dB
400Hz -4dB
1Khz 0dB
2.5KHz +6dB
6.3KHz -6dB
16KHz +6dB

The analogue EQ can be adjusted +-10dB per channel with 0dB or -6dB attenuation.

Should probably raise 6.3KHz and lower the 16Khz channels.
Here's my perspective. First, your measurements don't look that bad compared to most rooms. I would encourage you to use your ears and ask if it actually sounds better with any of that EQ applied. 2nd, as others have encouraged, it's important to measure in different spots since room reflections will change the frequency response for different locations. If you add 6db of 2.5KHz to compensate for one location, but another location doesn't have the same dip then you will make it sound terrible in the other location. Adding 6db of any frequency is a LOT. I would encourage you to use only subtractive EQ 99% of the time and be gentle with it. Measure in some different spots before and after EQ. You might find that using little or no EQ is the best option. And always use your ears.
 
OP
D

Deleted member 35357

Guest
Here are some effects the Analogue 7-band EQ has on the Measurements:

221217 UMIK-2 MMM EQ effects.png
 
OP
D

Deleted member 35357

Guest
It certainly takes more time than you think to mess around with these measurements.

Make a change, check the result, do another change, check the result and it goes on.
:)

I guess I should be trying to level match the L & R speakers as they are on different output levels overall throughout the measurements.
2-4dB mismatch across the range. (though it's not obvious sitting listening to content) Though it was apparent doing the pink noise.
Would speaker placement help with that? Though they should be equal distance from the measurement area.

SUB does add lots of output for the right channel only.
 
OP
D

Deleted member 35357

Guest
I tried to EQ the Left & Right channels individually but the valleys are so big they need considerable boost for REW to even try at them.

The various guides only suggesting maximum boost of 6dB wasn't close to REW even try to EQ them.
12dB or higher needed or it leaves them untouched.

Here is a result of one test run I did with allowing 12dB for REW to try calculate EQ.
0,5dB slope/Octave
(40/60hz) cut-off between the channels. (Right channels has the SUB)

Just a straight line down from 20hz low frequency target where it stars the downturn with the cut-off for 40/60hz. 18dB/Octave using bass limited speaker settings.


Result after EQ was applied:
EQ3 v2 results.png


The left channel is running a little anaemic with 160-400hz as REW didn't touch this area for the left channel with the settings used. Would need higher 16dB allowance or something.

Just a update on how things are going. (the EQ APO profiles only needed -6dB preamp settings to go with them)

Starts to look like something.

Though I still get clipping when using higher output from REW test signals. (output stops, goes silent)
I think I might need to adjust the UMIK-2 sensitivity. (but I thought the calibration files should have been enough according to the FAQ)
 

Rednaxela

Major Contributor
Joined
Mar 30, 2022
Messages
2,052
Likes
2,675
Location
NL
Instead of an average line, why not plot a L+R measurement?
 
OP
D

Deleted member 35357

Guest
This isn’t really what you want is it?
No way to do it otherwise currently without buying stuff to add, EX5 only has 2-channels, and the speakers has no option other than a sub gain setting.
I could probably buy a separate mixer or something that takes the left & Right channel to process into one before sending it to the sub... It's old and dumb. (no volume control either on the set). You need to process the stuff before you send it over to the speakers inputs.

I'm looking to upgrade everything. Just wanted to get this out of the way done so I know how it compares and and learn the details of these software suits.
Learning this stuff first with the doing measurements and EQ before I buy what I might be needing for the new setup.
The limits of what I can do here and what needs to be done in hardware etc.

New sub, new main speakers, potential surround ones.
(New multi-channel DAC.)

I'm most likely to buy a set of Genelec. Though unsure if going for the 8010/8020 or go for the GLM kits.
The Ones are nice but out of my price range I want to spend here for the all-round desktop setup.

I could go 8030 as mains and some 8010/8020 as surrounds. Though I need to add the SUB to the calculation. Those Genelec ones aren't cheap.
2.1 setup for a start and then later add better mains and move the weaker ones to surrounds.

7050 or 7040 sub?
Go for a pair of 7040 and have them left & right for the main speakers?

Though I think they have proper controls so could manage with only one sub with the Genelec's.
I expect the 7040 to be better than what I have in my current setup.

GLM seems unnecessary looking as I have the REW and the UMIK-2. Though loading the EQ into the speakers seems like a nice feature.
Do measurements once and load the correction into the speakers and any source will get the benefit.
Though the GLM function is with added costs for the hardware to have that functionality.
 
OP
D

Deleted member 35357

Guest
Instead of an average line, why not plot a L+R measurement?
EQ3 v2 results L-R.png


Left EQ:
Code:
Filter Settings file

Room EQ V5.20.13
Dated: 2022-dec-18 09:05:08

Notes:REW PEQ calc Left speaker bass limited speaker 60hz cutoff, target level 70dB 0.5dB/Octave Slope v2

Equaliser: Generic
dec 17 16:22:20 Left EQ3
Channel: L
Preamp: -6 dB
Filter  1: ON  PK       Fc   22.30 Hz  Gain -17.00 dB  Q  2.000
Filter  2: ON  PK       Fc   54.40 Hz  Gain  10.00 dB  Q  3.010
Filter  3: ON  PK       Fc   58.40 Hz  Gain -10.90 dB  Q  3.976
Filter  4: ON  PK       Fc   73.10 Hz  Gain   2.20 dB  Q  7.186
Filter  5: ON  PK       Fc   117.0 Hz  Gain   3.70 dB  Q  7.482
Filter  6: ON  PK       Fc   388.0 Hz  Gain   0.00 dB  Q  1.004
Filter  7: ON  PK       Fc   572.0 Hz  Gain   5.90 dB  Q  2.784
Filter  8: ON  PK       Fc   867.0 Hz  Gain   4.90 dB  Q  7.483
Filter  9: ON  PK       Fc   999.0 Hz  Gain   6.10 dB  Q  7.495
Filter 10: ON  PK       Fc    1040 Hz  Gain  -4.50 dB  Q  4.480
Filter 11: ON  PK       Fc    1313 Hz  Gain   3.30 dB  Q  7.475
Filter 12: ON  PK       Fc    1810 Hz  Gain  -2.60 dB  Q  4.820
Filter 13: ON  PK       Fc    2486 Hz  Gain  -2.60 dB  Q  4.998
Filter 14: ON  PK       Fc    3316 Hz  Gain   9.30 dB  Q  2.038
Filter 15: ON  PK       Fc    4833 Hz  Gain  -6.00 dB  Q  1.011
Filter 16: ON  PK       Fc    6718 Hz  Gain   7.40 dB  Q  1.928
Filter 17: ON  PK       Fc   10823 Hz  Gain  -3.90 dB  Q  4.994
Filter 18: ON  None   
Filter 19: ON  None   
Filter 20: ON  None   
Filter 21: ON  None   
Filter 25: ON  None

Right EQ:
Code:
Filter Settings file

Room EQ V5.20.13
Dated: 2022-dec-18 09:35:50

Notes:REW PEQ calc Right speaker bass limited speaker 40hz cutoff, target level 70dB 0.5dB/Octave Slope v2

Equaliser: Generic
dec 17 16:24:36 Right EQ3
Channel: R
Preamp: -6 dB
Filter  1: ON  PK       Fc   24.50 Hz  Gain  -9.30 dB  Q  6.992
Filter  2: ON  PK       Fc   28.20 Hz  Gain   6.70 dB  Q  4.445
Filter  3: ON  PK       Fc   35.75 Hz  Gain   4.30 dB  Q  5.444
Filter  4: ON  PK       Fc   57.10 Hz  Gain  11.80 dB  Q  5.018
Filter  5: ON  PK       Fc   58.70 Hz  Gain -16.80 dB  Q  8.307
Filter  6: ON  PK       Fc   74.20 Hz  Gain  11.90 dB  Q  3.163
Filter  7: ON  PK       Fc   75.60 Hz  Gain -13.00 dB  Q  2.918
Filter  8: ON  PK       Fc   104.5 Hz  Gain  -3.20 dB  Q  2.069
Filter  9: ON  PK       Fc   164.5 Hz  Gain   1.00 dB  Q  7.447
Filter 10: ON  PK       Fc   214.0 Hz  Gain  11.80 dB  Q  4.350
Filter 11: ON  PK       Fc   221.0 Hz  Gain -19.10 dB  Q  4.582
Filter 12: ON  PK       Fc   221.0 Hz  Gain   6.00 dB  Q  1.574
Filter 13: ON  PK       Fc   294.0 Hz  Gain   3.90 dB  Q  2.632
Filter 14: ON  PK       Fc   701.0 Hz  Gain  -2.50 dB  Q  4.604
Filter 15: ON  PK       Fc   928.0 Hz  Gain   4.10 dB  Q  5.987
Filter 16: ON  PK       Fc    1770 Hz  Gain   0.00 dB  Q  2.123
Filter 17: ON  PK       Fc    1794 Hz  Gain  -0.40 dB  Q  1.324
Filter 18: ON  PK       Fc    2752 Hz  Gain   3.40 dB  Q  2.197
Filter 19: ON  PK       Fc    4081 Hz  Gain   4.90 dB  Q  1.051
Filter 20: ON  PK       Fc    7427 Hz  Gain   3.90 dB  Q  2.806
Filter 21: ON  None   
Filter 25: ON  None
 

ThatSoundsGood

Active Member
Forum Donor
Joined
Aug 12, 2022
Messages
147
Likes
114
Is your measurement microphone exactly the same distance from the tweeter on the left and right speakers?
Are you measuring for more than just one listening spot?
 
OP
D

Deleted member 35357

Guest
Is your measurement microphone exactly the same distance from the tweeter on the left and right speakers?
Are you measuring for more than just one listening spot?
As was suggested I did a MMM, measurements. (with 90degree mic calibration file used)

Moving the mic in about ~1 meter cubic area of the seating position where all sound will be listened to. (desktop usage, 1 person)

*I figured out that the filter on my DAC had an effect on the clipping, changing it stopped the issue. (I reckon ultrasonic clipping happened)
Filter was set to Nr. 6 and changed it to Nr.1 for 'better attentuation'. Clipping of test signals didn't happen this time with higher SPL.
 
OP
D

Deleted member 35357

Guest
Yeah the test EQ was generally ~fine~ but felt that there were some odd quirks with the sound overall, somewhat unfamiliar feel. (spacial anomalies?)
Some bass got some kick to it while it seems to have removed volume a bit to much with the current try with the AUTO generation & targets.

Might not roll off the sub as much as I did or adjust the house curve to preserve some more base before the roll off starts.

The following song really lacked that sub being there. It needs it severely to be fun to listen to.


Vocals improved severely though for much content. These speakers really don't sound nice at all without EQ of some sort. (the straighter the line the better it gets I've seen)
 

ThatSoundsGood

Active Member
Forum Donor
Joined
Aug 12, 2022
Messages
147
Likes
114
Often times, the auto-EQ stuff goes too heavy and removes too much for my taste, especially with the subwoofer. Those EQ settings look crazy. You're removing 16db of 58.7Hz and adding 11.8db of 57Hz. That can't sound good. I would use a wider Q and some smoothing on the measurements. I see the same problem at 74 and 75Hz. It's probably reading a crazy swing in the frequency response of the room, but your ear doesn't hear it like that. Those two frequencies are too close to each other to be EQ'ing like that and there will be a ton of overlap in those two filters which is just going to add phase problems. Is there a way to get it to read your results after you add 1/12 octave smoothing?
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,843
Likes
5,777
View attachment 250793

Left EQ:
Code:
Filter Settings file

Room EQ V5.20.13
Dated: 2022-dec-18 09:05:08

Notes:REW PEQ calc Left speaker bass limited speaker 60hz cutoff, target level 70dB 0.5dB/Octave Slope v2

Equaliser: Generic
dec 17 16:22:20 Left EQ3
Channel: L
Preamp: -6 dB
Filter  1: ON  PK       Fc   22.30 Hz  Gain -17.00 dB  Q  2.000
Filter  2: ON  PK       Fc   54.40 Hz  Gain  10.00 dB  Q  3.010
Filter  3: ON  PK       Fc   58.40 Hz  Gain -10.90 dB  Q  3.976
Filter  4: ON  PK       Fc   73.10 Hz  Gain   2.20 dB  Q  7.186
Filter  5: ON  PK       Fc   117.0 Hz  Gain   3.70 dB  Q  7.482
Filter  6: ON  PK       Fc   388.0 Hz  Gain   0.00 dB  Q  1.004
Filter  7: ON  PK       Fc   572.0 Hz  Gain   5.90 dB  Q  2.784
Filter  8: ON  PK       Fc   867.0 Hz  Gain   4.90 dB  Q  7.483
Filter  9: ON  PK       Fc   999.0 Hz  Gain   6.10 dB  Q  7.495
Filter 10: ON  PK       Fc    1040 Hz  Gain  -4.50 dB  Q  4.480
Filter 11: ON  PK       Fc    1313 Hz  Gain   3.30 dB  Q  7.475
Filter 12: ON  PK       Fc    1810 Hz  Gain  -2.60 dB  Q  4.820
Filter 13: ON  PK       Fc    2486 Hz  Gain  -2.60 dB  Q  4.998
Filter 14: ON  PK       Fc    3316 Hz  Gain   9.30 dB  Q  2.038
Filter 15: ON  PK       Fc    4833 Hz  Gain  -6.00 dB  Q  1.011
Filter 16: ON  PK       Fc    6718 Hz  Gain   7.40 dB  Q  1.928
Filter 17: ON  PK       Fc   10823 Hz  Gain  -3.90 dB  Q  4.994
Filter 18: ON  None  
Filter 19: ON  None  
Filter 20: ON  None  
Filter 21: ON  None  
Filter 25: ON  None

Right EQ:
Code:
Filter Settings file

Room EQ V5.20.13
Dated: 2022-dec-18 09:35:50

Notes:REW PEQ calc Right speaker bass limited speaker 40hz cutoff, target level 70dB 0.5dB/Octave Slope v2

Equaliser: Generic
dec 17 16:24:36 Right EQ3
Channel: R
Preamp: -6 dB
Filter  1: ON  PK       Fc   24.50 Hz  Gain  -9.30 dB  Q  6.992
Filter  2: ON  PK       Fc   28.20 Hz  Gain   6.70 dB  Q  4.445
Filter  3: ON  PK       Fc   35.75 Hz  Gain   4.30 dB  Q  5.444
Filter  4: ON  PK       Fc   57.10 Hz  Gain  11.80 dB  Q  5.018
Filter  5: ON  PK       Fc   58.70 Hz  Gain -16.80 dB  Q  8.307
Filter  6: ON  PK       Fc   74.20 Hz  Gain  11.90 dB  Q  3.163
Filter  7: ON  PK       Fc   75.60 Hz  Gain -13.00 dB  Q  2.918
Filter  8: ON  PK       Fc   104.5 Hz  Gain  -3.20 dB  Q  2.069
Filter  9: ON  PK       Fc   164.5 Hz  Gain   1.00 dB  Q  7.447
Filter 10: ON  PK       Fc   214.0 Hz  Gain  11.80 dB  Q  4.350
Filter 11: ON  PK       Fc   221.0 Hz  Gain -19.10 dB  Q  4.582
Filter 12: ON  PK       Fc   221.0 Hz  Gain   6.00 dB  Q  1.574
Filter 13: ON  PK       Fc   294.0 Hz  Gain   3.90 dB  Q  2.632
Filter 14: ON  PK       Fc   701.0 Hz  Gain  -2.50 dB  Q  4.604
Filter 15: ON  PK       Fc   928.0 Hz  Gain   4.10 dB  Q  5.987
Filter 16: ON  PK       Fc    1770 Hz  Gain   0.00 dB  Q  2.123
Filter 17: ON  PK       Fc    1794 Hz  Gain  -0.40 dB  Q  1.324
Filter 18: ON  PK       Fc    2752 Hz  Gain   3.40 dB  Q  2.197
Filter 19: ON  PK       Fc    4081 Hz  Gain   4.90 dB  Q  1.051
Filter 20: ON  PK       Fc    7427 Hz  Gain   3.90 dB  Q  2.806
Filter 21: ON  None  
Filter 25: ON  None
The green is the channel without the sub,right?
 

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,537
Likes
12,004
Those EQ settings look crazy. You're removing 16db of 58.7Hz and adding 11.8db of 57Hz. That can't sound good.
that's not how PEQ works. Filters influence each other. For example, here is a 20dB boost right next to a 20dB cut:
Screenshot 2022-12-18 201212.png
as you can see, the two filters almost cancel each other out.

And here's the actual filter response of those crazy settings:
Nighthog EQ Right.png
 
Top Bottom