• 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!

tallbeardedone

Active Member
Joined
Sep 3, 2022
Messages
102
Likes
215
Hey guys, I've spent a lot of time Optimizing my Speaker Positioning using REW and after weeks of tweaking my current set-up has the following psychoacoustically smoothed frequency response at the listening position as measured by a Umik-1 microphone:

1160 fr psy.png


As you can see I've got a wonderfully flat response overall, with no extreme bass nodes at all, and only small peaks centered at 65hz, 850Hz, and 3500Hz respectively.

I quite like the 65Hz modal peak (which is a room dependant anti-node across the 5.6m width of my room, regardless of speaker positioning), as it adds slam and meat the bones and to my ears still sounds fast without too much bloom (although the waterfall does show a bit).

This leaves only the other two minor peaks at 850Hz and 3500Hz. To attenuate these I used the parametric EQ inside my Auralic Aries G1 streamer. The details of each band, including center frequency, dB attenuation, and Q value, are shown below:

auralic EQ bands.PNG
auralic EQ bands 2.jpg


Results:

I've been teaching myself to critically listen using soundgym and audiocheck and I believe that the peak at 3500Hz in the high midrange presence region may cause a bit of listening fatigue over time, so that attenuation in particular appears (at least psychosomatically) to allow me to listen longer and at higher average volume without fatigue. The attenuation at 850Hz is very difficult to hear and I'm not confident that I could pick it out in a blind ABX test with someone else switching the EQ on and off. To be honest the same can probably be said for the 3500Hz band but the mind is a funny thing and when sighted I feel I can hear the difference quite clearly as there's a slightly sharpness I feel in my ear drums over time that is attenuated with the filter in place.
Recently I tried adding Dirac to my room with a miniDSP SHD added to my digital chain between streamer and DAC, but my initial results were underwhelming. Dirac seemed to over correct problems that, to my ears, weren't there. My imaging and holographic soundstaging, which are specific strengths of my Monitor Audio gold 300's, may have sharpened slightly, but also FLATTENED, and the immersive 3-D effect I get when siting slightly nearfield, along with the WOW factor of "seeing" the performance in three-dimensional auditory space, disappeared somewhat. I'm going to do a more precise re-measure with Dirac soon and report back, but at this stage the simple EQ filters from my Auralic produce more pleasing results in my listening room.

TLDR; Two small EQ bands produces better auditory results than Dirac to my already excellent raw frequency response in my listening room.

@amirm I'm wondering what you think of the specific bands and if I should adjust either their frequency, dB, or Q-value. I'm after a flat response (outside of the 60hz bass boost) as opposed to a sloping Harmon curve, due to personal preference. Thanks!
 

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,313
Location
UK
Hey guys, I've spent a lot of time Optimizing my Speaker Positioning using REW and after weeks of tweaking my current set-up has the following psychoacoustically smoothed frequency response at the listening position as measured by a Umik-1 microphone:

View attachment 248418

As you can see I've got a wonderfully flat response overall, with no extreme bass nodes at all, and only small peaks centered at 65hz, 850Hz, and 3500Hz respectively.

I quite like the 65Hz modal peak (which is a room dependant anti-node across the 5.6m width of my room, regardless of speaker positioning), as it adds slam and meat the bones and to my ears still sounds fast without too much bloom (although the waterfall does show a bit).

This leaves only the other two minor peaks at 850Hz and 3500Hz. To attenuate these I used the parametric EQ inside my Auralic Aries G1 streamer. The details of each band, including center frequency, dB attenuation, and Q value, are shown below:

View attachment 248419View attachment 248420

Results:

I've been teaching myself to critically listen using soundgym and audiocheck and I believe that the peak at 3500Hz in the high midrange presence region may cause a bit of listening fatigue over time, so that attenuation in particular appears (at least psychosomatically) to allow me to listen longer and at higher average volume without fatigue. The attenuation at 850Hz is very difficult to hear and I'm not confident that I could pick it out in a blind ABX test with someone else switching the EQ on and off. To be honest the same can probably be said for the 3500Hz band but the mind is a funny thing and when sighted I feel I can hear the difference quite clearly as there's a slightly sharpness I feel in my ear drums over time that is attenuated with the filter in place.
Recently I tried adding Dirac to my room with a miniDSP SHD added to my digital chain between streamer and DAC, but my initial results were underwhelming. Dirac seemed to over correct problems that, to my ears, weren't there. My imaging and holographic soundstaging, which are specific strengths of my Monitor Audio gold 300's, may have sharpened slightly, but also FLATTENED, and the immersive 3-D effect I get when siting slightly nearfield, along with the WOW factor of "seeing" the performance in three-dimensional auditory space, disappeared somewhat. I'm going to do a more precise re-measure with Dirac soon and report back, but at this stage the simple EQ filters from my Auralic produce more pleasing results in my listening room.

TLDR; Two small EQ bands produces better auditory results than Dirac to my already excellent raw frequency response in my listening room.

@amirm I'm wondering what you think of the specific bands and if I should adjust either their frequency, dB, or Q-value. I'm after a flat response (outside of the 60hz bass boost) as opposed to a sloping Harmon curve, due to personal preference. Thanks!
You start with showing us objective measurements and then switch to all subjective descriptions. Why not show what happened after Dirac or your own EQ instead of describing it with words like “flattened,” “3D auditory space,” “seeing the performance?”
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,830
Likes
5,757
I wouldn't stand the first picture response,the least I can stand is tilted at least 5-6db centered at the 200 axis,sometimes even more than 3db at the lows.
 
OP
T

tallbeardedone

Active Member
Joined
Sep 3, 2022
Messages
102
Likes
215
You start with showing us objective measurements and then switch to all subjective descriptions. Why not show what happened after Dirac or your own EQ instead of describing it with words like “flattened,” “3D auditory space,” “seeing the performance?”
This is where the objective vs subjective argument comes in. With Dirac in the chain my frequency response was flatter and sloping in a variety of Harmon curves as shown below, so objectively they measured better, but subjectively sounded worse. The only way I can describe the "worse" is with the subjective verbiage used above. As an example, without Dirac the reaction I get from people who sit in the listening position and hear the illusion of a three-dimensional auditory event is "WOW, I didn't know that was even a thing!" But with Dirac in the chain, that reaction vanished. It's tonally corrected but sounds "flat" and less immersive. I'm as much of a objectivist as you can get, but I believe since Dirac is measuring and corrective for both the direct and reflected sound reaching the microphone, it's over correcting for something that's not wrong in the first place. As you can see from my original graph I have an excellent uncorrected in-room response, adding Dirac seems (subjectively, I know) to ruin it. Not sure what else to tell you. If you're in the area come have a listen and hear for yourself.

+2.5db to -1.6dB.png
+3db to -3dB.png

+3db to -3dB.png
+7dB to -3dB.png
 
Last edited:

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,313
Location
UK
This is where the objective vs subjective argument comes in. With Dirac in the chain my frequency response was flatter and sloping in a variety of Harmon curves as shown below, so objectively they measured better, but subjectively sounded worse.
I respect your opinion. I was simply asking to see the results so that we can understand what is it that you do not like. Otherwise, how can we comment?
 
OP
T

tallbeardedone

Active Member
Joined
Sep 3, 2022
Messages
102
Likes
215
I wouldn't stand the first picture response,the least I can stand is tilted at least 5-6db centered at the 200 axis,sometimes even more than 3db at the lows.
Yeah most people seem to prefer (subjectively) a Harmon curve. I prefer a flat response where I'm getting each frequency at exactly the intended volume. Personal preference. If I measure from further back in the room (near the tip of the equilateral triangle) the entire response slopes down at about 1.5db per octave, but sitting further back is less immersive and the soundstage is mostly horizontal, whereas sitting about 60cm near field from the apex of the equilateral triangle gives me a really cool wrap around 3-D effect that gives the auditory illusion of being up on stage with the performers. All that is subjective of course, but it's what I enjoy in my listening room and is super cool for everyone who hasn't heard proper three-dimensional imaging before.
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,830
Likes
5,757
That's mine at 2.8 meters just before the tip of the triangle,psychoacoustically smoothed.
(before EQ,I only use it to cut this mountain at 30Hz a little)

FR.jpg


(high level to excite the room as much as I can)
 
OP
T

tallbeardedone

Active Member
Joined
Sep 3, 2022
Messages
102
Likes
215
That's mine at 2.8 meters just before the tip of the triangle,psychoacoustically smoothed.
(before EQ,I only use it to cut this mountain at 30Hz a little)

View attachment 248436

(high level to excite the room as much as I can)
Yeah nicely sloping down, which many prefer. My response at the tip of the equilateral triangle is similar, although not sloping down quiiiite as much. What speakers and amplification do you have?
 

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,313
Location
UK
That's mine at 2.8 meters just before the tip of the triangle,psychoacoustically smoothed.
(before EQ,I only use it to cut this mountain at 30Hz a little)

View attachment 248436

(high level to excite the room as much as I can)
How little is little?
 

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,313
Location
UK
Depends,I have 2-3 profiles in Mathaudio each one according to level I'm listening.
The loud one cuts 6db with broader Q the late night one only two.
At around 30Hz you have a 10-15dB peak depending on your preferred slope. Do you have the after EQ curves?
 

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,191
Location
Northern Virginia, USA
I'll note, once upon a time we had tone controls, since one size does not fit all.
 
OP
T

tallbeardedone

Active Member
Joined
Sep 3, 2022
Messages
102
Likes
215
I more of an installation in progress,active between lows and mids and passive between mids and highs.
Amps are Ice Power edge,1200as2 for lows and 300a2 for mid-highs.
Wow sounds cool. So do the active crossovers have to run an extra ADC to DAC after your initial DAC source? I have a buddy with some Klipsch Jubilees that do such a thing.
 

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,313
Location
UK

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,830
Likes
5,757
At around 30Hz you have a 10-15dB peak depending on your preferred slope. Do you have the after EQ curves?
One of them,older,can't measure now,it's late here.
Must be the late-night one.



FR2.jpg
 
OP
T

tallbeardedone

Active Member
Joined
Sep 3, 2022
Messages
102
Likes
215
Analog xover for the time being.So...
Yeah so in your (subjective, I know) opinion do you think the extra ADC to DAC has a detrimental effect on the sound quality? I tried an ADC with my vinyl set-up to add Dirac to my analog chain and it sounded awful, but that's a bit different.
 

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,313
Location
UK
One of them,older,can't measure now,it's late here.
Must be the late-night one.



View attachment 248442
Thank you. As I expected the peak is still prominent. Even taking into consideration of the tilt it is at least 6dB. Not easy to miss.
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,830
Likes
5,757
Yeah so in your (subjective, I know) opinion do you think the extra ADC to DAC has a detrimental effect on the sound quality? I tried an ADC with my vinyl set-up to add Dirac to my analog chain and it sounded awful, but that's a bit different.
I don't worry about the sound quality of the digital ones,there are really nice ones.
I worry about their glitches,that can expose the full power of an amp in it's full glory.
 
Top Bottom