• 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). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Panel vs Cones and Domes

jhaider

Senior Member
Joined
Jun 5, 2016
Messages
348
Likes
224
#22
Dirac does take multiple measurements. My experience was that it was effective, but one thing you had to do was get the initial central measurement location absolutely spot on.
I believe Dirac can generate filters off of a single measurement, though one shouldn't use it that way.

I agree that the initial position is important to get right, because Dirac cues levels and delay off of it. Otherwise, at least when you're using Dirac as room correction (i.e. below your room's statistical region) and not as EQ (full-band) exact mike placement is not critical.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
7,160
Likes
3,577
Location
Riverview, FL
#23
Experiment:

MMM (moving mic) vs SweetSpot JBL LSR 308

No smoothing on MMM, 555 averages (continuous measurement), 1/6 octave on Sweet Spot.

Dark Red - MMM ... Red - Sweet Spot ... No DRC
Gray - Sweet Spot with AcourateDRC (flat)
Levels normalized at 1kHz

upload_2017-7-6_4-34-52.png


"MMM is quick and and efficient for equalization of a loudspeakers but because it is missing the time and phase information, it is not a tool to set up crossovers, time align speakers,...(for this, you need first an MLS or sinesweep measurement at one point)." - Jean-Luc Ohl

---

Ok, another tool/method to use when appropriate.
 
Last edited:

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
7,160
Likes
3,577
Location
Riverview, FL
#24
Isn't the arithmetic mean the appropriate one since that's how we hear?
Still thinking about a good answer.

Best one so far is "I don't think so."
 

The Smokester

Active Member
Joined
May 11, 2017
Messages
136
Likes
36
Location
SF Bay
#25
Still thinking about a good answer.

Best one so far is "I don't think so."
The arithmetic mean in dB is the average sound intensity to which the ear responds (because it responds logarithmically wrt power).

The "exponential " mean is the average of the power.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
7,160
Likes
3,577
Location
Riverview, FL
#26
The arithmetic mean in dB is the average sound intensity to which the ear responds (because it responds logarithmically wrt power).

The "exponential " mean is the average of the power.
Hmm... My brain doesn't process that, so...

Maybe smoothing - as in a sweep measurement - is arithmetic? I look, and find this:

https://dsp.stackexchange.com/questions/9967/1-n-octave-smoothing

upload_2017-7-6_14-33-23.png


I'm not equipped to figure out, much less argue, if that is arithmetic or not.

May I change my "I don't think so" reply to something even less definite?
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,427
Likes
534
#27
Hmm... My brain doesn't process that, so...

Maybe smoothing - as in a sweep measurement - is arithmetic? I look, and find this:

https://dsp.stackexchange.com/questions/9967/1-n-octave-smoothing

View attachment 7726

I'm not equipped to figure out, much less argue, if that is arithmetic or not.

May I change my "I don't think so" reply to something even less definite?

I think you will find that it is true that DSP Room correction tools rely on either a single point or a multipoint mindset. They seem to get along about as well as cats and dogs on that subject. I think you will also find that those tools using multi point averaging insist that they do not use any of the simple averaging methods discussed above. Rather, they use something more sophisticated, which, alas, is proprietary and undisclosed.
 

jhaider

Senior Member
Joined
Jun 5, 2016
Messages
348
Likes
224
#28
Experiment:

MMM (moving mic) vs SweetSpot JBL LSR 308

No smoothing on MMM, 555 averages (continuous measurement), 1/6 octave on Sweet Spot.

Dark Red - MMM ... Red - Sweet Spot ... No DRC
Gray - Sweet Spot with AcourateDRC (flat)
Levels normalized at 1kHz
Thanks for the MM measurement! I am a little confused by your legends, though. Am I reading right that the MMM measurement is with no DRC? If so, did you also do an MMM to validate the DRC?

Assuming I'm right about the legend, it seems to me that the native response is much better than the equalized response.

Your DRC "flat" response reminds me of the ear-bleeding horror that Audyssey's Flat target curve was in my living room: shrieking on top and neutered on the bottom. A curve that follows the natural upper bass rise and HF absorption-in-air usually sounds more natural. Is that a target curve you arrived at to suit your taste by iterative listening, or is that just the default curve of the program?
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
7,160
Likes
3,577
Location
Riverview, FL
#29
Am I reading right that the MMM measurement is with no DRC?
Yes.

If so, did you also do an MMM to validate the DRC?
Not yet.

Assuming I'm right about the legend, it seems to me that the native response is much better than the equalized response.
Maybe, but that's where preference comes in. And references to some statistically validated tests of what folks prefer. I tend to prefer other things than what is "preferred" on average.

Your DRC "flat" response reminds me of the ear-bleeding horror that Audyssey's Flat target curve was in my living room: shrieking on top and neutered on the bottom.
I have had no complaints. AcourateDRC is not Audyssey (which I have never, to my knowledge, heard).

A curve that follows the natural upper bass rise and HF absorption-in-air usually sounds more natural.
That's the theory. I find it a bit tubby sounding after a while. It masks the lower bass. My ear gets tired, I go back to flat.

Is that a target curve you arrived at to suit your taste by iterative listening, or is that just the default curve of the program?
My approach is to launch what was recorded into the sweet spot with as little difference as I can muster in a calibrated and measurable sort of way.

Here's corrected in-room RTA vs whatever is raw/uncorrected on the CD, in this case, Bobby Previte - Music from the Moscow Circus

upload_2017-7-8_2-31-50.png

I reserve the right to change my mind or method at some future date, of course. Right now, I'm pretty happy with what I have going.

My "EQ" currently. Left, Right, and Both.

upload_2017-7-8_3-32-53.png

Yes, it's real flat.
 
Last edited:

Similar threads

Top Bottom