• 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!

Room Measurement Tutorial for Dummies Part 1

JustIntonation

Senior Member
Joined
Jun 20, 2018
Messages
480
Likes
293
At some point I will write Part 3 that will talk about what to do with the measurements. For now, I highly recommend NOT reading AVS Forum threads. The few who corner the threads follow 1970s school of acoustics. Ask them to show you one blind test or recent research paper that backs what they say. You won't find any. Modern acoustics relies on psychoacoustics research. Without that, you are shooting in the dark as the people in 1970s did.
Do you mean EQ settings based on the measurements or will you also talk about room treatment? I can share some knowledge on the latter.
Edit: Sorry didn't realize I was replying to a more than 2 year old message! :) I found part 2 but couldn't find part 3, is it out there or yet to be written?
 
Last edited:

digitalfrost

Major Contributor
Joined
Jul 22, 2018
Messages
1,521
Likes
3,086
Location
Palatinate, Germany
I have a problem with time alignment. I used to measure via my onboard soundcard into a Focusrite Scarlett 2i2 (USB). With this, I could position the mic exactly in the acoustic center with help of HOLMimpulse.

I would then measure with REW after that using the acoustic time alignment.

Because of issues with unbalanced connections and amplifiers, I've bought another USB soundcard. Now my setup is:

Playback: RME Digiface USB - > 2x Topping D50 (sub/sat)
Record: Focusrite Scarlett 2i2

I am unable to get time coherent measurements between the left and right channel. The delay is all over the place and the impulse center often differs more than 100k samples at 48khz. Do I need to buy a single interface for recording and playback?

How does this work for people who get the UMIK-1 but use a USB soundcard for playback?

I am using DRC-FIR for room correction. With the old measurements it sounded awesome. With the new ones, it sounds okay, but especially in the bass I get weird phase issues. The bass is no longer in the mono middle, but everywhere.
 

Daverz

Major Contributor
Joined
Mar 17, 2019
Messages
1,294
Likes
1,451
I would have thought that the acoustic timing signal would have taken care of aligning the impulse response. I'm using a UMIK-1 for the measurements and feeding the exported impulse response from REW into DRC-FIR. The sweep is played from the music server. Here's the corrected left and right impulse responses:

drc-fir-corrected-impulse.jpg


I'm placing the mic using a laser ranger with millimeter accuracy, so the shift here should not be a surprise.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
I would have thought that the acoustic timing signal would have taken care of aligning the impulse response. I'm using a UMIK-1 for the measurements and feeding the exported impulse response from REW into DRC-FIR. The sweep is played from the music server. Here's the corrected left and right impulse responses:

View attachment 27901

I'm placing the mic using a laser ranger with millimeter accuracy, so the shift here should not be a surprise.


I'm having a hard time understanding what I'm seeing in your post.


If I match the visual scale of the waveform displayed with a full frequency range impulse here, the scale is in microseconds:

1560946232531.png
 

Daverz

Major Contributor
Joined
Mar 17, 2019
Messages
1,294
Likes
1,451
I'm having a hard time understanding what I'm seeing in your post.


If I match the visual scale of the waveform displayed with a full frequency range impulse here, the scale is in microseconds:

The offset may be confusing. With the offset adjusted:

impulse-adjusted.jpg


This gives me an idea of how I can fine tune the mic position. Once I've centered the mic with the laser ranger, I can do quick left and right REW sweeps and check the impulse overlay until the impulses are nearly on top of each other. It would take a very steady hand and a lot of patience, though.
 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
The offset may be confusing. With the offset adjusted:

View attachment 27911

This gives me an idea of how I can fine tune the mic position. Once I've centered the mic with the laser ranger, I can do quick left and right REW sweeps and check the impulse overlay until the impulses are nearly on top of each other. It would take a very steady hand and a lot of patience, though.

Are you using sweeps with acoustic timing signal played by the same speaker for both channels?
 

dc655321

Major Contributor
Joined
Mar 4, 2018
Messages
1,597
Likes
2,235
The offset may be confusing. With the offset adjusted:

View attachment 27911

This gives me an idea of how I can fine tune the mic position. Once I've centered the mic with the laser ranger, I can do quick left and right REW sweeps and check the impulse overlay until the impulses are nearly on top of each other. It would take a very steady hand and a lot of patience, though.


Tweeter only?
 

Daverz

Major Contributor
Joined
Mar 17, 2019
Messages
1,294
Likes
1,451
Are you using sweeps with acoustic timing signal played by the same speaker for both channels?

I use a left acoustic timing signal for the left channel sweep and a right acoustic timing signal for the right channel sweep.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I use a left acoustic timing signal for the left channel sweep and a right acoustic timing signal for the right channel sweep.

I suggest you use the same speaker for playing acoustic timing reference signal as that is the only way to correctly overlay two IRs. You don't need laser, just move the mike and make measurements untill they are alligned. You can also apply delay for one speaker if you are using brutefir convolution engine.

P.S. don't forget to press "Estimate IR delay" and shift it if necessary
 

Daverz

Major Contributor
Joined
Mar 17, 2019
Messages
1,294
Likes
1,451
He is asking if that is IR from tweeter only or from all drivers in the speaker.

I suppose I could get away with only doing upper frequencies for the mic alignment. The full sweep uncorrected impulse response is pretty messy:

uncorrected impulse.jpg

But I feel bad now that digitalfrost has not got his problem solved.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I suppose I could get away with only doing upper frequencies for the mic alignment. The full sweep uncorrected impulse response is pretty messy:

View attachment 27914
But I feel bad now that digitalfrost has not got his problem solved.
B

These 2 IRs are practically perfectly aligned, you don't need them to be perfeclty one over the other.

And what was his problem?
 

Daverz

Major Contributor
Joined
Mar 17, 2019
Messages
1,294
Likes
1,451
These 2 IRs are practically perfectly aligned, you don't need them to be perfeclty one over the other.

I'll have to try it with a common acoustic reference (e.g. left channel only).
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I have a problem with time alignment. I used to measure via my onboard soundcard into a Focusrite Scarlett 2i2 (USB). With this, I could position the mic exactly in the acoustic center with help of HOLMimpulse.

I would then measure with REW after that using the acoustic time alignment.

Because of issues with unbalanced connections and amplifiers, I've bought another USB soundcard. Now my setup is:

Playback: RME Digiface USB - > 2x Topping D50 (sub/sat)
Record: Focusrite Scarlett 2i2

I am unable to get time coherent measurements between the left and right channel. The delay is all over the place and the impulse center often differs more than 100k samples at 48khz. Do I need to buy a single interface for recording and playback?

How does this work for people who get the UMIK-1 but use a USB soundcard for playback?

I am using DRC-FIR for room correction. With the old measurements it sounded awesome. With the new ones, it sounds okay, but especially in the bass I get weird phase issues. The bass is no longer in the mono middle, but everywhere.


So, if I understand correctly with your first setup you were playing via USB and recording with analogue and now you are doing playing and recording via USB?

Have you set it up like this?

Capture.JPG
 
Last edited:

digitalfrost

Major Contributor
Joined
Jul 22, 2018
Messages
1,521
Likes
3,086
Location
Palatinate, Germany
So, if I understand correctly with your first setup you were playing via USB and recording with analogue and now you are doing playing and recording via USB?

Have you set it up like this?

View attachment 27915

First setup was PCIe soundcard out, USB in. Now it's USB out, USB in, but two separate devices. I don't use ASIO since I use EqualizerAPO for filtering which only works with WDM, I use the normal Java driver. I pretty sure it's set up correctly, I'll sent a screenshot once I'm home.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
First setup was PCIe soundcard out, USB in. Now it's USB out, USB in, but two separate devices. I don't use ASIO since I use EqualizerAPO for filtering which only works with WDM, I use the normal Java driver. I pretty sure it's set up correctly, I'll sent a screenshot once I'm home.

I think I remember vaguely reading somewhere that this kind of setup (USB out, USB in) may be problematic when measuring with timing reference signal but I can easilly be wrong on this. Hopefully @JohnPM will chime in and clear this for you..

In the meantime I suggest you try to do the measurements with ASIO to see if things will be ok again. ASIO can coexist with WDM so no harm if you don't need it with Equalizer APO.
 

digitalfrost

Major Contributor
Joined
Jul 22, 2018
Messages
1,521
Likes
3,086
Location
Palatinate, Germany
I think I remember vaguely reading somewhere that this kind of setup (USB out, USB in) may be problematic when measuring with timing reference signal but I can easilly be wrong on this. Hopefully @JohnPM will chime in and clear this for you..

In the meantime I suggest you try to do the measurements with ASIO to see if things will be ok again. ASIO can coexist with WDM so no harm if you don't need it with Equalizer APO.
I absolutely do need EqAPO because it is my crossover. The RME Digiface is a 8 channel WDM device towards windows, and I do 5.1 downmixing, high- and lowpass filters as well as convoling in EqAPO. Since it is bypassed in ASIO mode, ASIO is of no use to me.

I had some success by lowering my soundcard buffers (which were 2048 for the RME) to the lowest possible (32). The thing is, it depends on the length of the measurement. Getting 1M sweeps which are 21 seconds long to match is really hard, 128k works most of the time.

Basically I have to try until I get measurements where the impulses match at 0. Then DRC FIR will correctly find the impulse peak at 48k samples in both files and my bass actually comes from between the speakers.

Here's my preferences page, version is 5.19 btw:
 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
The RME Digiface is a 8 channel WDM device towards windows, and I do 5.1 downmixing, high- and lowpass filters as well convoling in EqAPO. Since it is bypassed in ASIO mode, ASIO is of no use to me.

I suggested you use ASIO only to play sweep file to get correct measurement, nothing more.

If you install ASIO drivers only for that purpose I am pretty sure they won't mess with your playout setup once you get the measurements right for the DRC-FIR to generate filters properly. You can deinstall them after that if you think they will mess with your setup.
 
Top Bottom