• 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 Treatment opinions and seeking Help with REW

Disruptor

Member
Joined
Mar 26, 2018
Messages
42
Likes
7
Hello everyone, I want to ask for some help with my REW skills and im quite interested in your opinions!


First of all here is my room! (ye i know its kinda scuffed up and needs some paint and all that but that is not my priority right now ;))

IMG_20190912_135456_1.jpg


IMG_20190912_135519.jpg
IMG_20190912_135658.jpg
IMG_20190912_135604.jpg


The bed and the furniture on the sidewall will be gone soon but for now im stuck with this.

Those absorbers are DIY made of Rockwool 100cm x 60cm x 8cm
Skyline difusers are also DIY made to those specs
2019-09-12 15_38_50-HiFi Loudspeaker Design.png

In total i have 15 made 6 of those are in a 3x2 array on the front wall and in a 3x3 on the back wall.

The rooms dimensions are: 370cm x 330cm x 254cm or 12ft x 11ft x 8ft
Speakers are set up at the long wall.
Speakers themselves are 135cm away from each other and are 30° tilted.

Gear:
JBL LSR 305 1. gen
JBL LSR 310 sub
Focusrite 2i2 1. gen

measuring mic: Sonarworks XREF 20

Those are my measurements 2 meters away from the front wall. (yes i know this is further than the isosceles triangle rule says but im still playing with the sitting distance and tilt) Ive measured various distances from 80cm to 200cm and found that 200cm is the best starting point.

200cm.jpg

Green line represents no EQ. 1/12 smoothing
Mic and Soundcard calibration files have been used.

After playing for quite some time with EQ´s and auto EQ´s from REW im still far away from my goals and this purple line feels like it lacks bass.
Ive just cut the peaks and didnt boost anything with the EQ that is applied here.

Since ive bought the Sonarworks mic i had a Sonarworks Reference trial and honestly i didn't fully like the results either.
Are there any tricks you guys could tell me and help me out or am i doomed with this sound curve^^
Would this freq curve be acceptable for you or are you disgusted with it?

Im also playing with the idea of buying a Genelec setup with 8351 and possibly two subs and just use their GLM system and be done but still i dont want to spend the money (around 12k €) and love to make the best of what i have now.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Those are my measurements 2 meters away from the front wall. (yes i know this is further than the isosceles triangle rule says but im still playing with the sitting distance and tilt) Ive measured various distances from 80cm to 200cm and found that 200cm is the best starting point.

View attachment 33159
Green line represents no EQ. 1/12 smoothing
Mic and Soundcard calibration files have been used.

After playing for quite some time with EQ´s and auto EQ´s from REW im still far away from my goals and this purple line feels like it lacks bass.
Ive just cut the peaks and didnt boost anything with the EQ that is applied here.
Are there any tricks you guys could tell me and help me out or am i doomed with this sound curve^^
Would this freq curve be acceptable for you or are you disgusted with it?

Try cutting red circled region so it follows blue line. You would probably need 2 filters to do that:

1. 1800Hz, Q=2, gain= -5dB
2. 4500Hz, Q=3, gain -4dB

These are of course just orientation figures with which you should play untill you get it close to blue line.

You could also consider raising that dip at 65Hz to see if it will react. Something like 70Hz, Q=8, gain=8dB. If this doesn't help don't be tempted to use higher gain as it won't help but can screw amp/speakers.

200cm.jpg
 

pozz

Слава Україні
Forum Donor
Editor
Joined
May 21, 2019
Messages
4,036
Likes
6,825
Hard to say form the pics, but it looks like you have your absorbers/diffusors mounted too high. I'm also not sure if your measurement from 2m away represents where you're actually sitting (that would be nearly 2/3 of the way into the room?).

So a few simple things would be to raise/tilt your monitors so that they fire more directly into your ears and lower your panels so that center around your ears.

When you measure, make sure you position the microphone at your sitting position, at ear height, measuring separately for R&L speakers and then averaging the results—you'll be able to see how differently the room affects each speaker. Use the variable smoothing function in REW as well to account for frequency-dependent psychoacoustic sensitivity.

The high frequency loss might be because your microphone either isn't positioned or calibrated correctly. It would be really unusual for your monitors to drop in response that fast.

Another thing you can try is to add a space between your Rockwool panels and the wall using brackets. Even a 3cm—4cm space will increase their effectiveness in the low mids/high bass. They are velocity-based absorbers, so having them directly against the wall isn't optimal.
 
OP
D

Disruptor

Member
Joined
Mar 26, 2018
Messages
42
Likes
7
ry cutting red circled region so it follows blue line. You would probably need 2 filters to do that:

1. 1800Hz, Q=2, gain= -5dB
2. 4500Hz, Q=3, gain -4dB

for now ive played just a little bit but tomorrow ill report back in since i dont have any more time for the room
full range eq.jpg

brown line is with the eq you recommended with a little tweaks
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
for now ive played just a little bit but tomorrow ill report back in since i dont have any more time for the room
View attachment 33164
brown line is with the eq you recommended with a little tweaks

It's getting better.

Try adding 3 more fitlers:

32Hz, Q=10, gain=-5dB
50Hz, Q=12, gain=-2dB
1500kHz, Q=3, gain=-3dB

Is your microphone aimed directly at speakers? You may try easing the 4500Hz filter as now it seems a bit too low.
 

pozz

Слава Україні
Forum Donor
Editor
Joined
May 21, 2019
Messages
4,036
Likes
6,825
This purple line feels like it lacks bass.

Those 40Hz-ish and 60Hz-ish dips are room modes to do with the length of your walls and ceiling height. Treating those will be difficult.
 
OP
D

Disruptor

Member
Joined
Mar 26, 2018
Messages
42
Likes
7
Hard to say form the pics, but it looks like you have your absorbers/diffusors mounted too high. I'm also not sure if your measurement from 2m away represents where you're actually sitting (that would be nearly 2/3 of the way into the room?).
yes the front wall is way to high but for now it is what it is and cannot be moved.

When you measure, make sure you position the microphone at your sitting position, at ear height, measuring separately for R&L speakers and then averaging the results—you'll be able to see how differently the room affects each speaker. Use the variable smoothing function in REW as well to account for frequency-dependent psychoacoustic sensitivity.

Yes mic height is of course considered and placed at my sitting position and ear level. Ive measured R&L and applied a little balance EQ since the right one plays 2,4db louder but never measured R&L for freq EQ purposes separate. And ill guess will read something about psychoacoustic sensitivity since ive seen it mentioned a few times

The high frequency loss might be because your microphone either isn't positioned or calibrated correctly. It would be really unusual for your monitors to drop in response that fast.
In the pic below the black line represents the mic calibration compensation. Could it be that my room takes up to much high freq?
mic cal.jpg


Another thing you can try is to add a space between your Rockwool panels and the wall using brackets. Even a 3cm—4cm space will increase their effectiveness in the low mids/high bass. They are velocity-based absorbers, so having them directly against the wall isn't optimal.

Well they have about 2.5cm behind them and i know they only work in the mid/high freq band since they are way to thin
 
OP
D

Disruptor

Member
Joined
Mar 26, 2018
Messages
42
Likes
7
yes the front wall is way to high but for now it is what it is and cannot be moved.

what ive forgot to mention is that the middle of the speakers aim at least at the lower third of the side and back panels
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
[QUOTE="Disruptor, post: 227076, member: 1658"
In the pic below the black line represents the mic calibration compensation. Could it be that my room takes up to much high freq?
[/QUOTE]

I think it is because mic is not pointed directly at the speakers. Does your convolution engine support separate left and right filters? It would be better to aim for those than for joint filter.
 
Last edited:

pozz

Слава Україні
Forum Donor
Editor
Joined
May 21, 2019
Messages
4,036
Likes
6,825
Could it be that my room takes up to much high freq?
No. Fabric-covered panels become reflective above 4kHz as far as I know. Calibration looks alright and in line with what I've looked up about your mic. It's definitely an issue with positioning—point the mic directly at each speaker from ear height to approximate what you're hearing most closely.

Beyond that, Room EQ isn't my thing—Krunok can give you much better advice there.
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
640
Likes
2,397
@Krunok and @pozz have good suggestions. While there is no RT60 for a small room, it would be good to see in REW an overlay of both channels and clicking on the RT60 will give you a display of Topt (for small rooms). Set the frequency response display to cover 100 Hz to 10 kHz. This way we can tell how much absorption is in the room over this frequency range.

Ideally you want the mic pointing straight ahead like you have at ear height and at the listening position.
 

pozz

Слава Україні
Forum Donor
Editor
Joined
May 21, 2019
Messages
4,036
Likes
6,825
Ideally you want the mic pointing straight ahead like you have at ear height and at the listening position.

Point the mic directly at each speaker from ear height to approximate what you're hearing most closely.
Seems like we're divergent on this point—what's the reasoning here @mitchco?
 
OP
D

Disruptor

Member
Joined
Mar 26, 2018
Messages
42
Likes
7
[QUOTE="Disruptor, post: 227076, member: 1658"
I think it is because mic is not pointed directly at the speakers. Does your convolution engine support separate left and right filters? It would be better to aim for those than for joint filter.

ive just tried pointing the mic at each speaker than averaging the two responses and its basically the same just 1-2db lower overall than having the mic dead straight pointing at the wall.

Im using Equalizer APO so i dont think it would be possible to have separate filters for each speaker
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
640
Likes
2,397
I don't know the mic, but I presume it is a calibrated omnidirectional measurement mic. If so, then there should be little to no high frequency attenuation 30 degrees of axis. My measurement mic shows about a 2 dB droop in the top octave 90 degrees of axis...

Maybe I have not had enough coffee here in PST time zone and I am missing something...?
 
OP
D

Disruptor

Member
Joined
Mar 26, 2018
Messages
42
Likes
7
@Krunok and @pozz have good suggestions. While there is no RT60 for a small room, it would be good to see in REW an overlay of both channels and clicking on the RT60 will give you a display of Topt (for small rooms). Set the frequency response display to cover 100 Hz to 10 kHz. This way we can tell how much absorption is in the room over this frequency range.

Ideally you want the mic pointing straight ahead like you have at ear height and at the listening position.
Topt.jpg

I hope this is what you meant
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I don't know the mic, but I presume it is a calibrated omnidirectional measurement mic. If so, then there should be little to no high frequency attenuation 30 degrees of axis. My measurement mic shows about a 2 dB droop in the top octave 90 degrees of axis...

Maybe I have not had enough coffee here in PST time zone and I am missing something...?

Hard to tell from the pic but mic angle toward on-axis to the speakers may be more than 30deg.

It is calibrated mic.
 
Top Bottom