• 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!

Wanted: Proof of multiple subs and sub EQ

Purité Audio

Master Contributor
Industry Insider
Barrowmaster
Forum Donor
Joined
Feb 29, 2016
Messages
9,238
Likes
12,580
Location
London

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,254
Likes
17,227
Location
Riverview FL

dallasjustice

Major Contributor
Joined
Feb 28, 2016
Messages
1,270
Likes
907
Location
Dallas, Texas
The only good thing about it is that it prolly doesn't eat up much decay time in the mids and highs.
 

Purité Audio

Master Contributor
Industry Insider
Barrowmaster
Forum Donor
Joined
Feb 29, 2016
Messages
9,238
Likes
12,580
Location
London
I just dont see PSI introducing a product that does not work, their monitors are highly thought of.
I will ask Roger Roschnick.
Keith.
 

dallasjustice

Major Contributor
Joined
Feb 28, 2016
Messages
1,270
Likes
907
Location
Dallas, Texas
image.png
Actual measurement at 48khz sample rate, 1/12 smoothing amplitude response with source/sink subs integrated using Acourate. The picture overlays the target over the R/L speakers. It's +\-2db from the target curve down to 20hz. Subs run in mono. The crossover is symmetrical NT 2nd order at 90hz.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,254
Likes
17,227
Location
Riverview FL

dallasjustice

Major Contributor
Joined
Feb 28, 2016
Messages
1,270
Likes
907
Location
Dallas, Texas
Now that you're done, have you lost interest yet?
Lost interest in what? I could improve the step response a little more. I'm thinking about getting another stereo amp and time aligning the JBL woofer with the horn compression driver to make the speakers fully active. It's not a high priority but if the right stereo amp comes along, I might do it.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,254
Likes
17,227
Location
Riverview FL
Oh, you're not done. Nevermind...
 

dallasjustice

Major Contributor
Joined
Feb 28, 2016
Messages
1,270
Likes
907
Location
Dallas, Texas
Oh, you're not done. Nevermind...
Have you purchased Mitch's book on DSP and Acourate yet? It's available on Amazon Kindle. It's $10 and well worth it. He goes into a lot of room acoustics science and connects it up with Acourate in a detailed and logical way. It's very helpful. The only problem with it is that all of the examples are from his fully active DIY horn speakers. It's sort of depressing looking at how well his speakers perform. :(
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,254
Likes
17,227
Location
Riverview FL
I'll take a look at the book.

I have AcourateDRC, not Acourate. It doesn't have all the complexities exposed for fiddling, and is dedicated to creating IIR and FIR filters for the miniDSP OpenDRC exclusively.

IIR - biquad filters - do the work on the lower end, and the FIR does the higher frequencies and minimum phase corrections.

Basically, create a target, take a single measurement of both/left/right, pick among two more/less aggressive settings, decide how many of the available IIR filters to use, and press Ok.

From the miniDSP, download the filter and continue.

FIR L/R filter at top, the combination of IIR filters L/R below, sum those together to pretend you can see the final curve, phase adjustment not displayed.

upload_2016-5-22_22-6-39.png


Here's the what you measured and after creating the filter, the what you are expected to get trace.

upload_2016-5-22_22-20-6.png


And the (just now) measured result in REW.

upload_2016-5-22_22-27-40.png


Oh, dipole nasties! Seating position in the worst place! Subs not scattered around the room!

You win.
 
Last edited:

andyc56

Active Member
Joined
May 14, 2016
Messages
122
Likes
170
Oh, what is it called? Inquiring minds want to know. :)

Is setting the range of the graph's y axis to 160 dB, and measuring response at only one position in the modal region with 1/12 octave smoothing covered in the book?

Modal region measurements should be unsmoothed and EQ'ed as such, and checked at multiple listening positions.
 

dallasjustice

Major Contributor
Joined
Feb 28, 2016
Messages
1,270
Likes
907
Location
Dallas, Texas
It may better to buy Acourate first before reading the book. I'm afraid it would scare you off. If I had known what I was getting myself into years ago and that years later I would still be learning how to use it, I would have said fuck it, give me another massive bass trap and a bigger amp! :p

I'll take a look at the book.

I have AcourateDRC, not Acourate. It doesn't have all the complexities exposed for fiddling, and is dedicated to creating IIR and FIR filters for the miniDSP OpenDRC exclusively.

IIR - biquad filters - do the work on the lower end, and the FIR does the higher frequencies and minimum phase corrections.

Basically, create a target, take a single measurement of both/left/right, pick among two more/less aggressive settings, decide how many of the available IIR filters to use, and press Ok.

From the miniDSP, download the filter and continue.

FIR L/R filter at top, the combination of IIR filters L/R below, sum those together to pretend you can see the final curve, phase adjustment not displayed.

View attachment 1971
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,757
Likes
242,249
Location
Seattle Area
OK, read through the first chapter and had to take a pause when I got to this:

"LEDE is state-of-the-art in acoustic control room design.

Barnett, Mitch. Accurate Sound Reproduction Using DSP (Kindle Locations 220-221). Kindle Edition. "


He is completely wrong here. LEDE is old school and nothing about it is state of the art. It came about in the age of broken studio monitor designs and assumptions about sound reproduction with no controlled testing. Decades of research since has shown the approach to be improper. Dr. D'Antonio who is the founder of RGP (largest manufacturers of acoustic product) puts all of this in great context in one of his presentations:

blackbird70.png


There it is: 1979 when LEDE (live end, dead end) was invented. Then came formal studies at NRC as highlighted below:

Blackbird80.png


RFZ became the other buzz word as was NE Room as mentioned above. Just as with LEDE, improper methodologies with no controlled testing to validate them.

Then came enlightenment:

blackbirdSummary.png


To the pinnacle of what we know about sound reproduction in rooms:
blackbird2000.png

blackbirdConclusion.png


The author is old school, having done pro work years back when concepts like LEDE were the rage. As with many people of that era and domain, they still talk about these old approaches as being valid even though it is trivial to show that they are not.

Anyway I will read through the rest of the text but so far, it is not a good thesis for home listeners to follow.
 

dallasjustice

Major Contributor
Joined
Feb 28, 2016
Messages
1,270
Likes
907
Location
Dallas, Texas
image.png
Oh lord! I post 1/12 smoothing because that's sort of standard because many folks think it fits well with what we hear. However, Acourate uses a psychoacoustic amplitude preparation prior to inversion. So the point here is to validate my target. I could post the psychoacoustic smoothing too but that wouldn't make any sense to non-Acourate users. Unsmoothed plots aren't very helpful and certainly should not be used prior to inversion.

Actually Mitch covers multiple seated positions very well in the book from his active 3 way stereo pair. He has lots of better looking graphs for you. Above is a sample. I'll prolly get in trouble with Mitch for posting it.


Is setting the range of the graph's y axis to 160 dB, and measuring response at only one position in the modal region with 1/12 octave smoothing covered in the book?

Modal region measurements should be unsmoothed and EQ'ed as such, and checked at multiple listening positions.
 
Top Bottom