• 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!

All room corrections are detrimental to sound clarity for me

Matthias McCready

Active Member
Joined
Jul 2, 2021
Messages
198
Likes
243
EQ is tool, and just as not all problems call for hammer it is not always a helpful tool or the right solution.

---
I come from a live audio background, and I know for tuning PA systems that flattening out every single little peak and dip would be a mistake....

The owner of Rational Acoustics (the predominate measurement software for live work) would call feeding this urge "measurebation."

Because if you move that mic over a single centimeter you will get a completely different set of peaks and nulls. Not to mention a foot away or 70ft away.

Most well tuned systems in a larger context have more minimal EQ with a wider Q. The best system tuners in the world often only do 3-4 bands of wide-Q filters. That is it. Anything more, and it is probably more destructive than constructive.

---

Now obviously a smaller room is different; you will not have the variance (high frequency air loss) and problems with timing inherent with a large spaces, but this still does hold true to extent. What is a prescriptive correction for one part of the room, might be quite destructive in another. Audio is game of compromises.

Thereby when applying EQ as a correction many measurements need to be taken to assess trends and make an average. This is balanced with where you know you will be listening.

Some things EQ will not account for:

1. EQ will not fix room problems (nodes, decay etc).
2. Speaker quality (you could make a $150 speaker measure the same as $5,000 but they wouldn't sound the same).
 
OP
czt

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
If EQ ( low bass only, miniimal phase only ) doesn’t sound good to you then don’t implement it.
Keith
Exactly what I do. ARC3 ("natural"/linear phase) and Sonarworks (mixed/linear phase) also doesn't sounded good.
 
OP
czt

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
EQ is tool, and just as not all problems call for hammer it is not always a helpful tool or the right solution.

---
I come from a live audio background, and I know for tuning PA systems that flattening out every single little peak and dip would be a mistake....

The owner of Rational Acoustics (the predominate measurement software for live work) would call feeding this urge "measurebation."

Because if you move that mic over a single centimeter you will get a completely different set of peaks and nulls. Not to mention a foot away or 70ft away.

Most well tuned systems in a larger context have more minimal EQ with a wider Q. The best system tuners in the world often only do 3-4 bands of wide-Q filters. That is it. Anything more, and it is probably more destructive than constructive.

---

Now obviously a smaller room is different; you will not have the variance (high frequency air loss) and problems with timing inherent with a large spaces, but this still does hold true to extent. What is a prescriptive correction for one part of the room, might be quite destructive in another. Audio is game of compromises.

Thereby when applying EQ as a correction many measurements need to be taken to assess trends and make an average. This is balanced with where you know you will be listening.

Some things EQ will not account for:

1. EQ will not fix room problems (nodes, decay etc).
2. Speaker quality (you could make a $150 speaker measure the same as $5,000 but they wouldn't sound the same).
We are on the same page here. And also in that, not so much difference in a nearfield/MLP situation. But my main concern was the side effects/artifacts of the commercial and free eq solutions.
 

Matthias McCready

Active Member
Joined
Jul 2, 2021
Messages
198
Likes
243
We are on the same page here. And also in that, not so much difference in a nearfield/MLP situation. But my main concern was the side effects/artifacts of the commercial and free eq solutions.

If there are artifacts than I would guess something is off. EQ is EQ, and unless you are EQing each of your speakers differently with a non-linear EQ there should not be artifacts or a hit to the "clarity."

Please note that this is not to knock your intelligence by any means. As someone who reguarly commits "operator error," it is a common enough phenomenon. Audio is complex!
 

Purité Audio

Master Contributor
Industry Insider
Barrowmaster
Forum Donor
Joined
Feb 29, 2016
Messages
5,587
Likes
5,543
Location
London
Exactly what I do. ARC3 ("natural"/linear phase) and Sonarworks (mixed/linear phase) also doesn't sounded good.
I wouldn’t use those automatic corrections, I would measure with REW then just correct ( with individual PEQ) any standing wave based peaks, check in REW that they are minimal phase then there are no phase issues and you are guaranteed a perfect inversion.
Keith
 

ZolaIII

Major Contributor
Joined
Jul 28, 2019
Messages
2,199
Likes
1,251
It depends on the room (shape), how speakers are ported, acoustic treatment helps. Take the sonar measurements only as reference (along with other response one's) and EQ with what ever you like. Try the basic windows peak limiter and use mic as dB meter live. Do the EBU R128 and equal loudness normalization (ISO 226) to a given dB lv for speakers (pleasurable mesured one) and definitely buy better speakers but you already know that. Don't expect miracles (97% of average people are in same situation with rooms and need to reduce bass 4~6 dB).
 
OP
czt

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
I wouldn’t use those automatic corrections, I would measure with REW then just correct ( with individual PEQ) any standing wave based peaks, check in REW that they are minimal phase then there are no phase issues and you are guaranteed a perfect inversion.
Keith
And not surprisingly, those were the best corrections in the end of my long experiments according to control measurements too.
 
OP
czt

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
As far as "measurebation", the automated corrections measure and average/calculate a lot, and are not as aggressive/careless as so many thinks according to my control measurements. Of course they are far from "perfect", but sophisticated enough for most studio people (their target audience).
 

Weeb Labs

Senior Member
Forum Donor
Joined
Jun 24, 2020
Messages
468
Likes
1,115
Location
Ireland
Could you please post a screenshot of your ARC window with the measured response, target and predicted response visible?

In the process of reviewing this monitor, I performed full range corrections using ARC, Dirac and Audiolense, with excellent results from all three. Perceived clarity was greatly improved in comparison to the uncorrected response. It may well be the case that your preference is vastly different from that of the typical listener or there may be a problem with your measurements.
 

pablolie

Addicted to Fun and Learning
Forum Donor
Joined
Jul 8, 2021
Messages
612
Likes
951
Location
bay area, ca
To me, our wired-in "sound imprint preference" sometimes stands in the way of accepting changes to the better. There is *zero* wrong with having a preference. Especially because recordings are all over the place in many ways.

I occasionally rotate equipment to avoid getting too attached to my favorite.

Also, room correction was such a radical change that indeed I had to allow myself to grow into it. We do proclaim the audiophile truth is the most possible accurate response, but sometimes that is hard to accept - it will reveal the flaws and imbalances and sometimes pathetic engineering in some of our favorite music... but it doesn't change the music.
 
OP
czt

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
Could you please post a screenshot of your ARC window with the measured response, target and predicted response visible?

In the process of reviewing this monitor, I performed full range corrections using ARC, Dirac and Audiolense, with excellent results from all three. Perceived clarity was greatly improved in comparison to the uncorrected response. It may well be the case that your preference is vastly different from that of the typical listener or there may be a problem with your measurements.
ARC3 correction looked and sounded good with the monitors alone, but mainly due to the "low end" correction and extension. (The attached example not the final correction as used then.)
 

Attachments

  • arc3-1.png
    arc3-1.png
    204.3 KB · Views: 133
OP
czt

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
It seems that the "error" sources were the usual preferred target curves. And/or a lot of albums that I listened nowadays was mastered with a (very) flat target. A/B testing again ARC3's quality correction which provides the best phase for me.
 

mightycicadalord

Addicted to Fun and Learning
Joined
Nov 10, 2021
Messages
542
Likes
521
ARC3 correction looked and sounded good with the monitors alone, but mainly due to the "low end" correction and extension. (The attached example not the final correction as used then.)

You've got no slope there really, that correction filter set is going to sound bad. Your monitors also seem to fall like a cliff after 100hz and correction is adding a lot back in, so distortion is probably rising a lot on the woofer and amp headroom is being eaten up. Those things really need a sub (or just better speakers, that's just bad low end extension). That lack of low end is probably going to greatly affect one's perception of EQ corrections and well sound in general.

This is my slope. It sounds great and I can't imagine not using EQ correction.

slopes.png
 
OP
czt

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
You've got no slope there really, that correction filter set is going to sound bad. Your monitors also seem to fall like a cliff after 100hz and correction is adding a lot back in, so distortion is probably rising a lot on the woofer and amp headroom is being eaten up. Those things really need a sub (or just better speakers, that's just bad low end extension). That lack of low end is probably going to greatly affect one's perception of EQ corrections and well sound in general.

This is my slope. It sounds great and I can't imagine not using EQ correction.

View attachment 186164
That was an older screenshot before the sub, but ARC3 low end extension was flawless and still a lot of headroom remained then. Unlike the crowd the flat ("too bright") target returned, and remains for the mentioned reason of sound clarity.
 

Attachments

  • arc3-3.png
    arc3-3.png
    209.1 KB · Views: 91

anotherhobby

Senior Member
Forum Donor
Joined
Dec 17, 2021
Messages
363
Likes
691
That was an older screenshot before the sub, but ARC3 low end extension was flawless and still a lot of headroom remained then. Unlike the crowd the flat ("too bright") target returned, and remains for the mentioned reason of sound clarity.
I'm not familiar with ARC3 at all so pardon me if this is an obvious question, but does that ARC3 graph window show actual room response after correction? Have you measured with REW to compare?

EDIT: Also, if that's an actual measured response, I'm super jealous by your lack of nulls.
 
OP
czt

czt

Active Member
Joined
Jul 14, 2021
Messages
103
Likes
63
I'm not familiar with ARC3 at all so pardon me if this is an obvious question, but does that ARC3 graph window show actual room response after correction? Have you measured with REW to compare?

EDIT: Also, if that's an actual measured response, I'm super jealous by your lack of nulls.
Orange shows ARC3 predicted/simulated "after" correction, verified by REW sweep (with acoustic timing reference) and RTA. The (almost) lack of nulls is the result of the optimization with REW RTA (first post). Eg if the sub would facing front instead of left in the same close to front wall position would have a huge ~9 dB SBIR null below 50 Hz. It is in the only best place in my current acoustic environment (treatments are NOT my world). Obviously my goal was to minimize the need of boost in ARC3, to avoid distortion as much as possible.
 

Attachments

  • ARC3.png
    ARC3.png
    129 KB · Views: 62
Top Bottom