Home Theater Forum and Systems banner

1 - 19 of 19 Posts

·
Registered
Joined
·
99 Posts
Discussion Starter #1
I want to take 3 measurements from the LP and near the LP, after I get these measurements how do I average them to Eq?Right now my measurements have all come from the LP, so I have eq'ed from that and it sounds good from the LP. Now if you move from the LP I have too much bass I want to correct this to be more even through out the room???
 

·
Super Moderator
Joined
·
9,257 Posts

If you’re sitting near the dead center of a square or rectangular room, then other locations, which will be closer to boundaries than the LP, will exhibit “hotter” bass. It’s just physics - EQ won’t fix it. If you EQ based on an average, then you'll have reduced bass at the main LP as well as the other locations.

Regards,
Wayne

 

·
Registered
Joined
·
25 Posts
If you’re sitting near the dead center of a square or rectangular room, then other locations, which will be closer to boundaries than the LP, will exhibit “hotter” bass. It’s just physics - EQ won’t fix it. If you EQ based on an average, then you'll have reduced bass at the main LP as well as the other locations.
Regards,
Wayne
Hello Wayne,
I am also interested in this topic. I have also read you your article: "Minimal EQ, Target Levels, and a Hard-Knee House Curve".
Based on this , I wondered if all individual peaks occurring in all Listening Positions LP's should be corrected, or only peaks that are common to all LP's.

I was thinking about following procedure (for subwoofer EQ with miniDSP no Audyssey, LP's for a 2-seater couch):
Either:
Determine the filters individually for each LP, determine the filters that are common to all LP's, and make average of those filters?
Or:
Make average of the LP measurements (with or without timing reference?) and EQ the average?

Or what procedure would you propose?
Thanks for your help.
 

·
Registered
Joined
·
99 Posts
Discussion Starter #5
If you’re sitting near the dead center of a square or rectangular room, then other locations, which will be closer to boundaries than the LP, will exhibit “hotter” bass. It’s just physics - EQ won’t fix it. If you EQ based on an average, then you'll have reduced bass at the main LP as well as the other locations.

Regards,
Wayne
Hi Wayne,
My room is 13.5 x 16 I sit about 11-12 feet back from the front wall. When I took all my measurements they were from the LP only. Listening from the LP everything sounds great but, if I stand up or move to say the side of the LP and I am standing then the bass is much hotter. Can I even this out some way?
 

·
Super Moderator
Joined
·
9,257 Posts

Hello Wayne,
I am also interested in this topic. I have also read you your article: "Minimal EQ, Target Levels, and a Hard-Knee House Curve".
Based on this , I wondered if all individual peaks occurring in all Listening Positions LP's should be corrected, or only peaks that are common to all LP's.

I was thinking about following procedure (for subwoofer EQ with miniDSP no Audyssey, LP's for a 2-seater couch):
Either:
Determine the filters individually for each LP, determine the filters that are common to all LP's, and make average of those filters?
Or:
Make average of the LP measurements (with or without timing reference?) and EQ the average?

Or what procedure would you propose?
Thanks for your help.
Hey turboman,
Most people just EQ for the main listening position. That’s the one you’d be sitting in for music listening. For movies, bass is mostly about “boom,” so less-than-perfect response isn’t readily noticeable.

That said, if response in all the seats was similar, you could certainly get good results by averaging.

Regards,
Wayne

 

·
Super Moderator
Joined
·
9,257 Posts

Can I even this out some way?
There’s nothing that can be done about it. EQ certainly can’t fix it. Bass traps might be an option; you might check with some manufacturers and (even better) users to see what they think. However, my understanding is that it takes lots of traps to deal with low frequency issues, and big ones at that. So I’m sticking with “nothing that can be done about it." It’s a common issue with rectangular and square rooms, especially smaller ones like yours. I avoid those rooms like the plague for serious listening.

Regards,
Wayne

 

·
Banned
Joined
·
4,838 Posts
Wayne P's advice is solid...

Also agree: focus on the LP and getting good sound there, and forget he rest of the room. NO ONE, and I repeat NO ONE will ever care a fraction as much about the sound as you do. And if they do, you will give them your seat anyway.
 

·
Registered
Joined
·
99 Posts
Discussion Starter #10
Wayne P's advice is solid...

Also agree: focus on the LP and getting good sound there, and forget he rest of the room. NO ONE, and I repeat NO ONE will ever care a fraction as much about the sound as you do. And if they do, you will give them your seat anyway.
Agreed!!!
 

·
Banned
Joined
·
4,838 Posts
Is there a way to include phase in the averaging? It appears to only account for magnitude
It cannot be done in the form you are thinking of.

In a sense, phase is already part of the magnitude averaging, because phase determines how the magnitudes combine (180 degrees gives complete cancellation, 0 degrees gives 3 to 6 db gain). IOW, magnitude A at phase A' combined with magnitude B at phase B' gives magnitude C at phase C'.

Averaging the phase angles either before or after the fact gives a meaningless result.:wink2:
 

·
Premium Member
Joined
·
6,711 Posts
If your measurements are appropriately time aligned you can pairwise average them using the trace arithmetic features, that is a vector average (magnitude and phase rather than magnitude only).
 

·
Registered
Joined
·
5 Posts
Thanks for the replies

I was just hoping to get a waterfall plot of the room average (and average reverb time), I have been able to do them using Easera, was just curious if REW could.
 

·
Premium Member
Joined
·
6,711 Posts
The vector average produces an impulse response from which a waterfall can be generated. You need a binary power of measurements to do the pairwise averaging down to an overall averaged result though.
 

·
Registered
Joined
·
25 Posts
The vector average produces an impulse response from which a waterfall can be generated. You need a binary power of measurements to do the pairwise averaging down to an overall averaged result though.
Hallo John
What do you mean by binary power of measurements? What settings to choose to get this binary power?
Thanks for the info.
 

·
Premium Member
Joined
·
6,711 Posts
That means 2 measurements, or 4, or 8, or 16. The trace arithmetic has an (A+B)/2 task for averaging a pair of measurements. If you only have 2 that's straightforward. If you have 4 you would average A and B (let's call the result AB), then average C and D, then average the two results AB and CD to get the overall average. Same process for 8 or 16, averaging down pairs of measurements then averaging the results and so on.
 

·
Premium Member
Joined
·
6,711 Posts
That is typically what you would expect from an average, particularly the magnitude averaging produced by e Average SPL button. If there are significant phase differences between the measurements then the trace arithmetic vector average can show pronounced dips where the measurements are out of phase with each other, in that case there will be regions in the average which are below either of the measurements that were used to produce it.
 
1 - 19 of 19 Posts
Top