Room correction without Sonarworks?

[Erik DuijsParticipant]

I have really bad acoustics in my room, so I was considering something like Sonarworks.
But I also did a test to do this using the tools I have:
- Set up a condenser microphone at my listening position
- Create a sample of white noise
- Record the white noise played back through my speakers with the microphone
- Use Voxengo Curve-EQ (which came with my DAW) to match the mic'ed noise with the sample

I suppose this is more or less what Sonarworks would do, right?

Although my microphone isn't really a reference mic so it isn't perfect, it still seems to work: The boomyness I was experiencing is pretty much gone at least, and it seems more balanced. Especially the low-end cleared up quite nicely.
Pretty crazy EQ-ing going on though, but I'm definitely going to try mixing this way.

If I were to do this with a proper reference mic, what benefits would something like Sonarworks still have?

[Arthur LabusModerator]

Hi Erik,

truly brilliant idea.
I have a calibration microphone and matching EQ.
I will give a try.

[Arthur LabusModerator]

There could be an IR response insert in chain on the output.
IR from well treated room. Best from control room 😉

[Erik DuijsParticipant]

Hi Arthur,

I'd be interested to hear your experience with this, using a proper calibration mic 🙂

Regarding your suggestion that Sonarworks might be using some IR algorithm to match with a well treated room, I suppose a 'well treated room' would still be far from 'perfect'?
I mean I'm not sure how something like that could actually be used to accurately change your own room acoustics into some other room, for example making your own room reverberate less/differently? You might be right though that it's more clever than just a matching EQ and a microphone, I don't know.

Cheers,
Erik

[Arthur LabusModerator]

I have no clue what Sonarworks do.
Just read once that someone did a measurment, reproduced the correction with EQ plugin on master bus and it was nearly the same result.
Maybe i wasn't that clear - i suggest to apply IR responce of "good room" to the output so it would be a part of measurment and matching process.
I will try to find some free IR libraries and try it.

[Arthur LabusModerator]

Did first, simple test - it works so far.
- played white noise sample and recorded it from my position with t-bone MM-1 microphone
- used EQuivocate with sidechain to match the original file
- eq "correction match curve" generated.
I must go right now, back in some time with some pics.

[Arthur LabusModerator]

So far so good ... aftermatch 😉

[Erik DuijsParticipant]

Interesting, I'm also seeing that huge top-end boost here.
I figured that might be a limitation of my (rather cheap) microphone since the ribbons of my speakers are supposed to go way beyond 20k, so I tucked that down a bit (not that I can hear anything that high :)).

[Arthur LabusModerator]

I did actually a little shotout with all my microphone types and this proudly amount of four mics 😉
I also used two variations of white noise.
Will post the results soon.

[Arthur LabusModerator]

Here are the results of matching the white noise sample with
- t.bone MM-1 measurement microphone
- t.bone SC-300 very cheap condenser
- t.bone SC-400 cheap condenser
- t.bone MB 55 very cheap dynamic
White noise sample, uniform distribution https://www.audiocheck.net/testtones_whitenoise.php

[Arthur LabusModerator]

And here are the results of matching the white noise (Gaussian distribution) sample with same mics and position
- t.bone MM-1 measurement microphone
- t.bone SC-300 very cheap condenser
- t.bone SC-400 cheap condenser
- t.bone MB 55 very cheap dynamic

[Arthur LabusModerator]

I would trust the MM-1 😀
Not really a difference between the white noise versions.
I would probably also try to match different noise signals and sweeps.

Any thoughts ?

[Author]

Posts