• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Best Room EQ settings for a desk setup?

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
View attachment 222106
That's the result of applying filters

Measurement looks good and filters seem to be calculated correctly. So, your impressions are that filters made the sound "hollow and lifeless"? That certainly comes as a surprise judging by the graph you posted. Can you post the exact filter values and describe how exactly are you applying the filters? What convolution engine are you using?

Btw, please limit the filter effect up to 400Hz only.
 

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
One more thing - I'm guessing the green line is showing expected frequency response with the filters applied, but what I need to see is the actual response while filters are active, so please measure the response using the same procedure as before with the filters applied.

In addition to measuring actual response with the filters applied using RTA moving mic method, please also measure single sweep response to one of your speakers while the mic is pointed to the center of the speaker. Post measurements in original REW format as a zipped file.
 
OP
juliangst

juliangst

Active Member
Joined
Dec 11, 2021
Messages
264
Likes
243
One more thing - I'm guessing the green line is showing expected frequency response with the filters applied, but what I need to see is the actual response while filters are active, so please measure the response using the same procedure as before with the filters applied.

In addition to measuring actual response with the filters applied using RTA moving mic method, please also measure single sweep response to one of your speakers while the mic is pointed to the center of the speaker. Post measurements in original REW format as a zipped file.
The green line is actually the measured result using the RTA method. I applied the filters using EQ APO.
I’ll send the filter setting and a sine sweep measurement when I’m back home
 
OP
juliangst

juliangst

Active Member
Joined
Dec 11, 2021
Messages
264
Likes
243
One more thing - I'm guessing the green line is showing expected frequency response with the filters applied, but what I need to see is the actual response while filters are active, so please measure the response using the same procedure as before with the filters applied.

In addition to measuring actual response with the filters applied using RTA moving mic method, please also measure single sweep response to one of your speakers while the mic is pointed to the center of the speaker. Post measurements in original REW format as a zipped file.
Those are the filters I use:

Filter 1: ON PK Fc 30.00 Hz Gain -5.10 dB Q 4.993
Filter 2: ON PK Fc 73.50 Hz Gain 5.70 dB Q 1.944
Filter 3: ON PK Fc 73.60 Hz Gain -13.70 dB Q 4.991
Filter 4: ON PK Fc 129.0 Hz Gain -14.90 dB Q 4.998
Filter 5: ON PK Fc 245.0 Hz Gain -9.20 dB Q 4.644
Filter 6: ON PK Fc 484.0 Hz Gain 3.10 dB Q 2.800
Filter 7: ON PK Fc 528.0 Hz Gain -4.60 dB Q 3.801

I turned off filter 7 and 6 in EQ APO but it didn't effect the sound that much.
I also did the sine sweep of a single speaker. Everything should be commented well enough in REW:
 

Attachments

  • LS50.zip
    2.5 MB · Views: 14

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
View attachment 222089
Interesting results (red is the RTA method).
I did about 80 averages and moved the mic around my listening postion. As you can see the 2-3kHz trough is gone and the bass response is pretty similar.
I did a 20Hz-20kHz pink noise RTA first but the drivers moved quite a lot so I switched to 30Hz-20kHz. I will now create filters up to ~600Hz and play around with bass shelf filters.

Let's clear this one first: that dip centered around 2.7kHz doesn't exist. It is caused by cancellation between speakers at mic point in space because of both speakers playing the sweep. That is why it doesn't show in RTA measurement. Sweep measurement with both speakers playing is valid only for LF, up to say 400Hz.
 

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
Those are the filters I use:

Filter 1: ON PK Fc 30.00 Hz Gain -5.10 dB Q 4.993
Filter 2: ON PK Fc 73.50 Hz Gain 5.70 dB Q 1.944
Filter 3: ON PK Fc 73.60 Hz Gain -13.70 dB Q 4.991
Filter 4: ON PK Fc 129.0 Hz Gain -14.90 dB Q 4.998
Filter 5: ON PK Fc 245.0 Hz Gain -9.20 dB Q 4.644
Filter 6: ON PK Fc 484.0 Hz Gain 3.10 dB Q 2.800
Filter 7: ON PK Fc 528.0 Hz Gain -4.60 dB Q 3.801

I turned off filter 7 and 6 in EQ APO but it didn't effect the sound that much.
I also did the sine sweep of a single speaker. Everything should be commented well enough in REW:

Ok, let's try it like this:

1.) RTA of both speakers playing from LP with no filters
2.) RTA of left speaker playing from LP with no filters
3.) RTA of right speaker playing from LP with no filters
4.) Sweep of left speaker from LP with no filters
5.) Sweep of right speaker from LP with no filters
6.) RTA of both speakers playing from LP with 3 filters shown below active


Capture.JPG


I would also need your subjective impressions with these 3 filters on vs off. Try switching the filters on individually and describe how they alter the sound.
Try listening to the pink noise with the filters on vs off.

These 3 filters are all with high Q and all in LF area so they should absolutely cause no "hollow" sound. Instead, they should reduces "boomines" when music hits that particular LF note at the frequency of the filter. They are not going to alter the perception of the music in any other respect.
 
Last edited:

dominikz

Addicted to Fun and Learning
Forum Donor
Joined
Oct 10, 2020
Messages
528
Likes
1,832
View attachment 222106
That's the result of applying filters
Just one comment - you are using a vertical scale of around 220dB in the graphs you post - that will hide a lot of details and make the response seem flatter than it really is.
I'd suggest to use ~50dB vertical scale (e.g. from 50 dB SPL to 100dB SPL for the above diagram) - that will give a much more realistic picture.

Perhaps you will also find the MMM measurement and REW room EQ instructions I posted here useful.

Good luck! :)
 

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
Those are the filters I use:

Filter 1: ON PK Fc 30.00 Hz Gain -5.10 dB Q 4.993
Filter 2: ON PK Fc 73.50 Hz Gain 5.70 dB Q 1.944
Filter 3: ON PK Fc 73.60 Hz Gain -13.70 dB Q 4.991
Filter 4: ON PK Fc 129.0 Hz Gain -14.90 dB Q 4.998
Filter 5: ON PK Fc 245.0 Hz Gain -9.20 dB Q 4.644
Filter 6: ON PK Fc 484.0 Hz Gain 3.10 dB Q 2.800
Filter 7: ON PK Fc 528.0 Hz Gain -4.60 dB Q 3.801

I turned off filter 7 and 6 in EQ APO but it didn't effect the sound that much.
I also did the sine sweep of a single speaker. Everything should be commented well enough in REW:

This kind of difference between a single channel sweep and both speakers RTA shouldn't exist, but let's wait for your new measurements as per my previous post.


Capture.JPG
 
OP
juliangst

juliangst

Active Member
Joined
Dec 11, 2021
Messages
264
Likes
243
Ok, let's try it like this:

1.) RTA of both speakers playing from LP with no filters
2.) RTA of left speaker playing from LP with no filters
3.) RTA of right speaker playing from LP with no filters
4.) Sweep of left speaker from LP with no filters
5.) Sweep of right speaker from LP with no filters
6.) RTA of both speakers playing from LP with 3 filters shown below active


View attachment 222272

I would also need your subjective impressions with these 3 filters on vs off. Try switching the filters on individually and describe how they alter the sound.
Try listening to the pink noise with the filters on vs off.

These 3 filters are all with high Q and all in LF area so they should absolutely cause no "hollow" sound. Instead, they should reduces "boomines" when music hits that particular LF note at the frequency of the filter. They are not going to alter the perception of the music in any other respect.

I did listening tests and got the same results as before.
I think I'm just used to the 'boominess' of those peaks in the bass. With filters applied it sounds more precise and correct but vocals also sound thinner and bass instruments are not as present as without filters.

The boominess adds 'more body' to the vocals and the bass energy just adds more 'fullness' to the sound presentation.
But on some tracks the boominess takes over and I prefer the filters (especially with complex orchestral passages).
 

Attachments

  • LS50_measurements.zip
    2.5 MB · Views: 13

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
I did listening tests and got the same results as before.
I think I'm just used to the 'boominess' of those peaks in the bass. With filters applied it sounds more precise and correct but vocals also sound thinner and bass instruments are not as present as without filters.

The boominess adds 'more body' to the vocals and the bass energy just adds more 'fullness' to the sound presentation.
But on some tracks the boominess takes over and I prefer the filters (especially with complex orchestral passages).

Personal preferences are never to be judged but uneven bass response is not something people often like. If you prefere more bass that is much better handled with high-shelf filters. Corrected response looks much better than uncorrected. If you find it to lack bass you should try experimenting with HS filters, like shown below (red line is your corrected response, lower line is with HS filter applied):

Capture2.JPG



What I find surprising is difference in the LF response up to app 124Hz of your left and right channel. It is shown on both, RTA and sweep measurements.
Can you think of any reason why this is happening? Is their position toward walls different?

Capture.JPG
 

czt

Member
Joined
Jul 14, 2021
Messages
99
Likes
61
Please send me 128k length sweeps with the microphone exactly at ear level and pointing to the center point between the speakers.
 
OP
juliangst

juliangst

Active Member
Joined
Dec 11, 2021
Messages
264
Likes
243
Can you think of any reason why this is happening? Is their position toward walls different?
The right speakers is next to this shelf while the left speaker has more free space. I can't think of anything else; I did both measurements at the same volume and didn't move the mic, so measurement mistakes shouldn't be the case here
 

Attachments

  • IMG_7007.jpeg
    IMG_7007.jpeg
    143.2 KB · Views: 39
  • IMG_7005.jpeg
    IMG_7005.jpeg
    178.3 KB · Views: 41
OP
juliangst

juliangst

Active Member
Joined
Dec 11, 2021
Messages
264
Likes
243
Please send me 128k length sweeps with the microphone exactly at ear level and pointing to the center point between the speakers.
 

Attachments

  • 128.zip
    2.6 MB · Views: 11

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
The right speakers is next to this shelf while the left speaker has more free space. I can't think of anything else; I did both measurements at the same volume and didn't move the mic, so measurement mistakes shouldn't be the case here

Measurements are looking consistent, so it's about speaker position.
 
OP
juliangst

juliangst

Active Member
Joined
Dec 11, 2021
Messages
264
Likes
243
I thought of building some rockwool acoustic panels but apparently no hardware store here has termarock available and only a few have sonorock.
I've heard basotect is a good alternative to rockwool and can be ordered online with no absurd shipping cost as with rockwool
 
OP
juliangst

juliangst

Active Member
Joined
Dec 11, 2021
Messages
264
Likes
243
What also might be interesting is the effect of port plugs. I used portplugs without the inside foam for all my measurements because that's the set up that always sounded best to me.
I did RTA of fully assembled and no port plugs for comparision
 

Attachments

  • rew measurements.zip
    35.7 KB · Views: 11

czt

Member
Joined
Jul 14, 2021
Messages
99
Likes
61
Try to convolve with the attached impulses with -6.9 dB gain. Fell free to ad HS to the EQ if you find the flat target to bright for near field.
 

Attachments

  • EQ.zip
    4.2 MB · Views: 9

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
I thought of building some rockwool acoustic panels but apparently no hardware store here has termarock available and only a few have sonorock.
I've heard basotect is a good alternative to rockwool and can be ordered online with no absurd shipping cost as with rockwool

I suggest you try to move both sepakers as close to the wall as possible and measure again to see if LF response improved.
 
Top Bottom