• 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?

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
That's what most other people told be as well. Your 3 filters also sound clearer to me than the seperate filters with boost and all that
That missing low range and asymmetry is too much, and will always be audible. Unfortunately lots of cognitive EQ dogmas here, despite the name. Clearer... maybe, but have you done level matched ABX compare (and control measurements)?
 

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
That missing low range and asymmetry is too much, and will always be audible. Unfortunately lots of cognitive EQ dogmas here, despite the name. Clearer... maybe, but have you done level matched ABX compare (and control measurements)?

There is no missing low range - response curve is supposed to be horizontal at 130cm listening distance, so in-room response of his speakers is actually pretty good.

Asymmetry at LF range doesn't matter as up to 120Hz (or even higher) we hear summed speaker response.

IME listeners report "clearer" overall response quite often once LF peaks that cause boomines are removed. When you think of it such effect is actually expected.
 

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
There is no missing low range - response curve is supposed to be horizontal at 130cm listening distance, so in-room response of his speakers is actually pretty good.

Asymmetry at LF range doesn't matter as up to 120Hz (or even higher) we hear summed speaker response.

IME listeners report "clearer" overall response quite often once LF peaks that cause boomines are removed. When you think of it such effect is actually expected.
Sorry, but without ABX test (and/or control measurements) all this is just dogma (here at "audio science review" and) for me.
 
OP
juliangst

juliangst

Senior Member
Joined
Dec 11, 2021
Messages
380
Likes
309
That missing low range and asymmetry is too much, and will always be audible. Unfortunately lots of cognitive EQ dogmas here, despite the name. Clearer... maybe, but have you done level matched ABX compare (and control measurements)?
The asymmetry is audible and the image is slightly shifted to the left side. I didn’t do ABX comparisons yet so this could be just my hearing.
 

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
The asymmetry is audible and the image is slightly shifted to the left side. I didn’t do ABX comparisons yet so this could be just my hearing.
Precisely. No, not just your hearing, which is as good as your measurements. What you hear is what I see in your measurements.
 

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
The asymmetry is audible and the image is slightly shifted to the left side. I didn’t do ABX comparisons yet so this could be just my hearing.

Your hearing is ok. Let's see what is happening. Here is left channel sweep vs right channel sweep with Psychoacoustic smoothing, so that difference in levels are more visible:


Psy.JPG


So, blue rectangle shows difference in LF area. There is no localization there so you cannot hear the direction where those sounds are coming from, they are just there, "hanging" in the room. What you are hearing in this region is the sum of those 2 signals. Beyond 100Hz those 2 curves looks pretty similar so why are you hearing the image shifted slightly to the left side?

Let's see how things are looking with 1:1 smoothing:

Sm1.JPG


Yep, there it is.. Those 2 red rectangles are showing the differences in level which are causing slight image shift toward the left speaker. Luckilly, both of them can easilly be corrected with a single very wide (low Q) filter, although I would first try correcting 100-600Hz region as the other one may just be the measurement artifact, as room doesn't play at all in the 1400-15000Hz region.
 
Last edited:

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
So we now EQ-ing (with an adequately scientific method?) in and above transition, it is progression...
 

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
So we now EQ-ing (with an adequately scientific method?) in and above transition, it is progression...

And that comes from a person who suggested that room correction should be done based on a single point sweep? :facepalm:

Ok, here's my final free lesson for you: transition frequency is not some descrete limit made of hard rock beyond which room magically stops affecting speaker's response. Beyond transition frequency room still affects speaker response all the way up to app 900-1000Hz, but on a smaller scale than below transition frequency. That region can be corrected based on spatially averaged measurement taken from LP, with filters of wider Q and smaller magnitude.

As for the region above 900-1000Hz, that is not room correction anymore, but a speaker correction. That region can also be corrected, either based on Klippel speaker measurements or, if one is not available for your speaker, you can do it yourself in your room. Speaker should be placed on a stand in the middle of the room, measurement is taken from the distance of 50-100cm and should be gated by app 6-7ms. That should give reasonably reliable response above 900Hz based on which speaker can be corrected.
 

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
And that comes from a person who suggested that room correction should be done based on a single point sweep? :facepalm:
No, I never suggested that. But please tell this to GLM, Acourate, Audiolense, DRC-FIR, etc. people too.
Ok, here's my final free lesson for you: transition frequency is not some descrete limit made of hard rock beyond which room magically stops affecting speaker's response. Beyond transition frequency room still affects speaker response all the way up to app 900-1000Hz, but on a smaller scale than below transition frequency. That region can be corrected based on spatially averaged measurement taken from LP, with filters of wider Q and smaller magnitude.

As for the region above 900-1000Hz, that is not room correction anymore, but a speaker correction. That region can also be corrected, either based on Klippel speaker measurements or, if one is not available for your speaker, you can do it yourself in your room. Speaker should be placed on a stand in the middle of the room, measurement is taken from the distance of 50-100cm and should be gated by app 6-7ms. That should give reasonably reliable response above 900Hz based on which speaker can be corrected.
Lesson... :) I don't said anyone of this: "discrete" or "room correction". Surprise... I already know all of this. (But thanks to you, beside others, I now "beginning" to understand "forums" are for what type of people (as sandbox). No offense, if a shirt not for you don't wear it. Unwatch thread, my time is too valuable for this.
 
Last edited:
OP
juliangst

juliangst

Senior Member
Joined
Dec 11, 2021
Messages
380
Likes
309
I did some new measurements and also applied the 100 degree FIR filter for the right speaker's phase correction.
Filters for the left speaker:

Filter 1: ON PK Fc 74.00 Hz Gain -13.90 dB Q 8.730
Filter 2: ON PK Fc 131.5 Hz Gain -16.90 dB Q 6.759
Filter 3: ON PK Fc 189.5 Hz Gain -6.60 dB Q 10.070
Filter 4: ON PK Fc 227.0 Hz Gain 6.00 dB Q 3.841
Filter 5: ON PK Fc 249.0 Hz Gain -10.80 dB Q 4.973

Filters for the right speaker:

Filter 1: ON PK Fc 73.50 Hz Gain -12.50 dB Q 18.990
Filter 2: ON PK Fc 128.5 Hz Gain -12.80 dB Q 9.659
Filter 3: ON PK Fc 141.0 Hz Gain -11.40 dB Q 6.486
Filter 4: ON PK Fc 150.0 Hz Gain 6.00 dB Q 4.280
Filter 5: ON PK Fc 283.0 Hz Gain -5.50 dB Q 4.906
 

Attachments

  • NewFilters_LS50.zip
    53.3 KB · Views: 16

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
I did some new measurements and also applied the 100 degree FIR filter for the right speaker's phase correction.
Filters for the left speaker:

Filter 1: ON PK Fc 74.00 Hz Gain -13.90 dB Q 8.730
Filter 2: ON PK Fc 131.5 Hz Gain -16.90 dB Q 6.759
Filter 3: ON PK Fc 189.5 Hz Gain -6.60 dB Q 10.070
Filter 4: ON PK Fc 227.0 Hz Gain 6.00 dB Q 3.841
Filter 5: ON PK Fc 249.0 Hz Gain -10.80 dB Q 4.973

Filters for the right speaker:

Filter 1: ON PK Fc 73.50 Hz Gain -12.50 dB Q 18.990
Filter 2: ON PK Fc 128.5 Hz Gain -12.80 dB Q 9.659
Filter 3: ON PK Fc 141.0 Hz Gain -11.40 dB Q 6.486
Filter 4: ON PK Fc 150.0 Hz Gain 6.00 dB Q 4.280
Filter 5: ON PK Fc 283.0 Hz Gain -5.50 dB Q 4.906

Thie result looks good. Did you loose any headroom or these are 0dB gain filters?
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
2,368
Likes
1,711
Location
Canada
I did some new measurements and also applied the 100 degree FIR filter for the right speaker's phase correction.
Filters for the left speaker:

Filter 1: ON PK Fc 74.00 Hz Gain -13.90 dB Q 8.730
Filter 2: ON PK Fc 131.5 Hz Gain -16.90 dB Q 6.759
Filter 3: ON PK Fc 189.5 Hz Gain -6.60 dB Q 10.070
Filter 4: ON PK Fc 227.0 Hz Gain 6.00 dB Q 3.841
Filter 5: ON PK Fc 249.0 Hz Gain -10.80 dB Q 4.973

Filters for the right speaker:

Filter 1: ON PK Fc 73.50 Hz Gain -12.50 dB Q 18.990
Filter 2: ON PK Fc 128.5 Hz Gain -12.80 dB Q 9.659
Filter 3: ON PK Fc 141.0 Hz Gain -11.40 dB Q 6.486
Filter 4: ON PK Fc 150.0 Hz Gain 6.00 dB Q 4.280
Filter 5: ON PK Fc 283.0 Hz Gain -5.50 dB Q 4.906

Not at a computer right now to check… But how does it sound to you?
 
OP
juliangst

juliangst

Senior Member
Joined
Dec 11, 2021
Messages
380
Likes
309
Not that different compared to the other more simple filter but a tiny bit better. I couldn't hear the effects of that FIR filter yet (except for the delay it causes)
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
2,368
Likes
1,711
Location
Canada
Not that different compared to the other more simple filter but a tiny bit better. I couldn't hear the effects of that FIR filter yet (except for the delay it causes)

Well, the delay is unfortunately unavoidable. But the dip “filling-in” should be unmistakably audible if you toggle the filter off-on while using a burst tone over the center frequency of the dip or phase cancellation from the left and right’s summed output.
 

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
Not that different compared to the other more simple filter but a tiny bit better. I couldn't hear the effects of that FIR filter yet (except for the delay it causes)

You can see the effect of that 100Hz phase filter on the graph below. It may not seem as much, but it is audible.

Capture.JPG
 

thorvat

Senior Member
Joined
Aug 9, 2021
Messages
323
Likes
378
I definietly hear a difference but it's not as prominent as it looks on the graph. But I'm really happy with the overall results now, thanks for your help!

Don't worry, things always look more drastic on the graph comparing to what you can hear. No problem, I'm glad this worked for you.
 
Last edited:
Top Bottom