• 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!

Share your in-room measurements?

Jon AA

Senior Member
Forum Donor
Joined
Feb 5, 2020
Messages
339
Likes
705
Location
Seattle Area
Very nice, only the bass rolloff to the mids is way too high for my tastes.
Yeah, I don't think you can say that without taking the speakers' directivity into account. These are most similar to the JBL M2, 4367, etc. I did that very much by design:

LCR60T12.jpg


While WAG at this point (until I get better quasi-anechoic measurements) that's my best guess of what the PIR would look like for these speakers with the listening window flat anechoically. With the significant directivity control kicking in from 500-700 Hz, if you're getting flat direct sound through that area at the listening position, you will measure a steady state drop due to fewer room reflections relative to the direct sound.

The additional bass boost below 100 Hz isn't related to that--it's just a typical "house curve"/Harman preferred bump added on top which I vary from time to time depending upon mood and program material.
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
3,635
Likes
2,687
If that's a response at your seat, directivity is already taken into account.

All I'm saying is I feel things lose clarity unless the bass boost ramps down by 150 Hz at the latest. Nothing wrong with preferring a little different response.

In your second screenshot, that measurement does look more in line with my preference. I just prefer more at 20-30 Hz :).

Not a critique, just comparing yours vs mine. Otherwise I'd say you're into the "excellent" category.
 

Jon AA

Senior Member
Forum Donor
Joined
Feb 5, 2020
Messages
339
Likes
705
Location
Seattle Area
If that's a response at your seat, directivity is already taken into account.
It doesn't though. That's a steady state curve that doesn't distinguish between the direct sound and the "room sound," it just adds everything together.

I can't accurately measure the direct sound at the listening position without reflections messing things up, but just for fun to give some rough idea, here is what single mic measurements look like when windowed to 5ms:

LCR60_5ms.jpg


The dips at 700 and 950 are reflections, they were there in the steady state as well at that particular mic location--move the mic a couple inches and they go away, so if you can ignore those this should give a very rough general idea of what the direct sound is like at that location.

As you can see, the high frequencies starting at 1000 Hz or so are pretty much on level with the lower midrange down to 200 Hz. If I trusted that measurement more, if anything it's saying I should drop the target from 500-700 Hz about another 1/2 dB or so. But it certainly shows if the goal is flat direct sound, raising the high frequencies another 2 dB certainly wouldn't be the right choice.

All I'm saying is I feel things lose clarity unless the bass boost ramps down by 150 Hz at the latest. Nothing wrong with preferring a little different response.

And I'm just saying with these particular speakers, I don't think you would. You really don't want the same in room response from these as you would speakers that are very wide dispersion to a much higher frequency. Anyway, that's probably a better discussion for one of the Room Curve threads, but I do think it was useful to explain why I did what I did (even if some may think it's wrong). For those intrigued, Paul Hales explains it quite well in this video (skip to 45 minutes or so for this discussion):


Actually, his entire explanation on how he comes up with a room curve for the $1,000,000 home theaters he builds is very interesting. I don't trust my ears as much as he does though....

In your second screenshot, that measurement does look more in line with my preference. I just prefer more at 20-30 Hz :).
Yeah, I often do to. I'll go for months with a couple more dB than shown above...then watch a couple of movies in a row with "overdone" bass and decide to turn it down a bit. And the "circle of confusion" goes round and round.....

Not a critique, just comparing yours vs mine. Otherwise I'd say you're into the "excellent" category.

Thanks.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
2,135
Likes
1,522
Location
Canada
It doesn't though. That's a steady state curve that doesn't distinguish between the direct sound and the "room sound," it just adds everything together.

I can't accurately measure the direct sound at the listening position without reflections messing things up, but just for fun to give some rough idea, here is what single mic measurements look like when windowed to 5ms:

View attachment 222810

The dips at 700 and 950 are reflections, they were there in the steady state as well at that particular mic location--move the mic a couple inches and they go away, so if you can ignore those this should give a very rough general idea of what the direct sound is like at that location.

As you can see, the high frequencies starting at 1000 Hz or so are pretty much on level with the lower midrange down to 200 Hz. If I trusted that measurement more, if anything it's saying I should drop the target from 500-700 Hz about another 1/2 dB or so. But it certainly shows if the goal is flat direct sound, raising the high frequencies another 2 dB certainly wouldn't be the right choice.



And I'm just saying with these particular speakers, I don't think you would. You really don't want the same in room response from these as you would speakers that are very wide dispersion to a much higher frequency. Anyway, that's probably a better discussion for one of the Room Curve threads, but I do think it was useful to explain why I did what I did (even if some may think it's wrong). For those intrigued, Paul Hales explains it quite well in this video (skip to 45 minutes or so for this discussion):


Actually, his entire explanation on how he comes up with a room curve for the $1,000,000 home theaters he builds is very interesting. I don't trust my ears as much as he does though....


Yeah, I often do to. I'll go for months with a couple more dB than shown above...then watch a couple of movies in a row with "overdone" bass and decide to turn it down a bit. And the "circle of confusion" goes round and round.....



Thanks.

Could you show the same measurements with frequency dependent window of maybe 7 or 10 cycles… as well as either left or right front’s spectral decay using a 20 ms “rise time” and 1/3 or 1/6 smoothing setting? I’m just curious.
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
3,635
Likes
2,687
It doesn't though. That's a steady state curve that doesn't distinguish between the direct sound and the "room sound," it just adds everything together.
And I'm just saying with these particular speakers, I don't think you would. You really don't want the same in room response from these as you would speakers that are very wide dispersion to a much higher frequency.
It seems like we're talking about the same thing. If the measurement includes reflections, then a wider dispersion speaker would have a greater proportion of the measurement as reflected sound. You can't see how much the reflected sound is contributing, but it's in there.

Also, if your on and off-axis curves are similar, this would be fine as the sound wouldn't change much with the inclusion of the extra width, and windowing wouldn't show big differences (save for a terrible room environment).

Now, we're only talking in terms of tonality, which is what these measurements show. Obviously dispersion width also affects perception of soundstage, clarity, and more.

Anyway, we are getting way too far into the weeds here, because none of this relates to my comment on the house curve! I just said in my experience I find the bass boost should flatten out before it gets into the lower midrange. I had a very similar curve to your first one, which came with Dirac Live. The lower midrange sounded too congested, which is how I came to realize curves similar to the Harman target were not always ideal.

I'll go for months with a couple more dB than shown above...then watch a couple of movies in a row with "overdone" bass and decide to turn it down a bit. And the "circle of confusion" goes round and round.....
I know the feeling. Lately I've been adding 2-3 to my sub levels for my Blu-ray input. Then along comes Bladerunner 2049...I might lose my home theater card for saying this, but I think the bass levels are almost too high in that movie.

Edited to add a graphic:

1561980206300.png


This illustrates what I was talking about. There is variability, but I'm very close to the green (average) line.
 
Last edited:

Vuki

Active Member
Joined
Apr 8, 2018
Messages
271
Likes
280
It seems like we're talking about the same thing. If the measurement includes reflections, then a wider dispersion speaker would have a greater proportion of the measurement as reflected sound. You can't see how much the reflected sound is contributing, but it's in there.

Also, if your on and off-axis curves are similar, this would be fine as the sound wouldn't change much with the inclusion of the extra width, and windowing wouldn't show big differences (save for a terrible room environment).

Now, we're only talking in terms of tonality, which is what these measurements show. Obviously dispersion width also affects perception of soundstage, clarity, and more.

Anyway, we are getting way too far into the weeds here, because none of this relates to my comment on the house curve! I just said in my experience I find the bass boost should flatten out before it gets into the lower midrange. I had a very similar curve to your first one, which came with Dirac Live. The lower midrange sounded too congested, which is how I came to realize curves similar to the Harman target were not always ideal.


I know the feeling. Lately I've been adding 2-3 to my sub levels for my Blu-ray input. Then along comes Bladerunner 2049...I might lose my home theater card for saying this, but I think the bass levels are almost too high in that movie.

Edited to add a graphic:

View attachment 222830

This illustrates what I was talking about. There is variability, but I'm very close to the green (average) line.
That curve shows prefered response for in-ear headphones- very much dependent on ear-headphone shape compatibility...
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
3,635
Likes
2,687
That curve shows prefered response for in-ear headphones- very much dependent on ear-headphone shape compatibility...
It is basically the same with speakers in a room. The difference is how you get there.
 

Andysu

Major Contributor
Joined
Dec 7, 2019
Messages
1,098
Likes
549
Just made some changes to my system, so I figured I'd throw this out there for the peanut gallery:

View attachment 222797


Damn I hate SBIR. I really need to get that new house built.
it's close but you ran out of those, instagram likes for filters lol . use 1/48 and switch between 1/3rd and move the mic small fractions or go manual with mic and hold it and move it around and listen to the matched if got proper matched LCR behind at-screen . you're close but it's got gabs or peaks or does it ? move mic small fractions and some of those gaps dips or peaks change soother with real time pink noise . once it's taken days to do then or if can use multi mics , please throw the auto-eq in the cat litter where auto-eq belongs . manual eq will solve it only takes patience .

move the mic around or figure out the shape void dip . and please extend or place HP filter on the higher end and raise the level up , or lower the other middle range down with BP with wide Q and lower it down , so the frequencies are not over loud , otherwise the upper brilliance highs will fly past your ears unnoticed , human hearing contour . need few more PEQ get yourself few behringer FBQ 2496 can daisy-chain channel A to B to increase bands from 20 to 40 per single each channel and use anywhere from tiny fraction Q bandwidth to slice some of those frequencies or do they need slicing smoothing down ? your near , your close .

get many mics make horizontal beam some wood extend the width of the room and place the mics so you can move them along , sort like a radio frequency dial . if ever listened to radio with manual dial ? same principle . if move or tint mic the mic will be in different 3d space . keep them all vertical and move slide the mic or use servo motors to move the mic up and down the beam of the whole width of the room . you be surprised at what you'll see . and also way to lower beam down or up higher and another beam so you can move down the length of the room in a controlled way . remember when present in the room your body is interfering with the frequency . just as long as you bare that ib mind . otherwise your frequency is getting close to spot on . or is it ? get yourself a classic THX 3417 so you can safely check the amp trade-off of frequency and the actual frequency of the amplifier vs the speaker/s it does make considerable differences , after all this is , ASR and it's all measurable science . :)
 

Jon AA

Senior Member
Forum Donor
Joined
Feb 5, 2020
Messages
339
Likes
705
Location
Seattle Area
Could you show the same measurements with frequency dependent window of maybe 7 or 10 cycles…
Sure:

LCR60_7FWD.jpg



I added a very rough trend line from the HF on down. It shows you don't get an average solidly above the trend until bass frequencies. If this is a reasonable indication of direct sound through the mid-high frequencies for that mic location, it again shows I should make the step down lowering the high frequencies even bigger. Going the other way--elimiating the step down by raising the HF up 2 dB would make a big hole in the lower midrange.

as well as either left or right front’s spectral decay using a 20 ms “rise time” and 1/3 or 1/6 smoothing setting? I’m just curious.
You'll have to bear with me a while, I haven't used that function much in REW so it'll take me a while to figure out what I'm doing. :)
 

Jon AA

Senior Member
Forum Donor
Joined
Feb 5, 2020
Messages
339
Likes
705
Location
Seattle Area
It seems like we're talking about the same thing. If the measurement includes reflections, then a wider dispersion speaker would have a greater proportion of the measurement as reflected sound. You can't see how much the reflected sound is contributing, but it's in there.
Yes, it's in there, but it doesn't sound the same. The direct sound has a disproportionate impact on perceived tonality. That's one of the fundamental findings of all of Toole's research. Did you listen to Paul Hale's explanation posted above? I didn't just make all this stuff up.

I just said in my experience I find the bass boost should flatten out before it gets into the lower midrange. I had a very similar curve to your first one, which came with Dirac Live. The lower midrange sounded too congested, which is how I came to realize curves similar to the Harman target were not always ideal.
And I'm saying, unless your speakers have horns 1.5-2' wide, that experience just isn't relevant to speakers that do. They are very different speakers. And it's not a "bass boost." It's a drop from mid to high frequencies. My "bass boost" is exactly how you have it drawn in the chart--but it's only 2 dB for this calibration (10 dB for speakers in a room is pretty hardcore IMHO unless you listen at very low levels).

A real world example:

Erin's measurement of the JBL M2:

CEA2034 -- JBL M2 (Crown iTech 5000 Amp; M2 Base Configuration).png


As you can see, on the particular M2 he measured, the HF section looks to be about 1-2 db hot. In his subjective listening (I trust his ears more than mine as well) he said it would sound better if the HF was dropped 1-2 dB. Looking at that chart that makes all the sense in the world.

If you look at the steady state measurements one may expect when measuring in a typical room, you get this:

Estimated In-Room Response.png


As you can see, even though the HF section measured 1-2 dB too high anechoically, and Erin felt in should be reduced 1-2 dB subjectively, in a room you'd expect it to measure 1-2 dB down relative to the lower midrange. If lowered 1-2 dB from there (as Erin's ears and the anechoic measurements suggest), you'd have a more significant step down in the in-room measurements than I do.

But overall, if you mentally eliminate the vertical directivity dip (my speakers don't have that) and replace it with a straight line, you end up with an in-room curve almost the exact shape I came up with through that region.

While I've said I don't trust my ears as much as some other peoples', that's not to say I don't trust them at all. The whole reason I went down this rabbit hole was because of what I heard. When I first built these speakers, I figured the center channel was too big to fit in my current room so I only built two. I was using a smaller center channel--also a vertical horn/compression driver speaker, but much smaller. It doesn't control directivity to nearly as low a frequency, and the transition is more gradual.

For a long time I had them EQ'd to the same, very conventional looking target curve below 1000 Hz. The tonality did not match. It bothered me--particularly because the little speaker sounded better than the big speakers through the lower midrange. That's when I spent some time thinking about what the target curves should look like for each speaker. That led be to the concept shown above and once I did that the tonality match was dramatically improved.
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
3,635
Likes
2,687
Yes, it's in there, but it doesn't sound the same. The direct sound has a disproportionate impact on perceived tonality. That's one of the fundamental findings of all of Toole's research.
The other finding is that the sound we perceive in-room is on average 44% early reflections.

I added a very rough trend line from the HF on down. It shows you don't get an average solidly above the trend until bass frequencies. If this is a reasonable indication of direct sound through the mid-high frequencies for that mic location, it again shows I should make the step down lowering the high frequencies even bigger. Going the other way--elimiating the step down by raising the HF up 2 dB would make a big hole in the lower midrange.
I've seen you mention this twice now – I do not think raising the high frequencies would be wise. I am very treble-averse. Did someone suggest that? Maybe it was buried in one of Andy's posts...hehe. I would have cut them just like you did.

And I'm saying, unless your speakers have horns 1.5-2' wide, that experience just isn't relevant to speakers that do. They are very different speakers. And it's not a "bass boost." It's a drop from mid to high frequencies.
Mid to high? I'm talking about 100-200 Hz. The point where the bass comes down to meet with the upper bass/lower mids in an in-room measurement.
 

thewas

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
4,453
Likes
9,977
The other finding is that the sound we perceive in-room is on average 44% early reflections.
Those 44% are not what we perceive (which we don't really know) but the way to approximate the measured PIR. As Toole says our hearing apparatus and processing is not the same as a typical omnidirectional mic and a computed linear transfer function.
 

Jon AA

Senior Member
Forum Donor
Joined
Feb 5, 2020
Messages
339
Likes
705
Location
Seattle Area
it's close but you ran out of those, instagram likes for filters lol . use 1/48 and switch between 1/3rd and move the mic small fractions or go manual with mic and hold it and move it around and listen to the matched if got proper matched LCR behind at-screen . you're close but it's got gabs or peaks or does it ?
Uhm, I'm sort of way ahead of you. I was merely posting results in a thread for results (where many were talking about using psychoacoustic smoothing), not analyzing my system. I figured 1/12 was more than good enough. 1/48 doesn't add anything meaningful:

LCR60T.jpg


Of course now everybody wants to analyze my system. :) And that's fine.

move mic small fractions and some of those gaps dips or peaks change soother with real time pink noise . once it's taken days to do then or if can use multi mics ,

MMM stands for Moving Microphone Method. The graphs originally posted contained the rough equivalent of ~120 mic positions around the MLP. Is that enough?

please throw the auto-eq in the cat litter where auto-eq belongs . manual eq will solve it only takes patience .

I'm not using "auto-eq." It's not your fault, I didn't explain in detail what I was doing as this was just a thread for posting results. But I'd be happy to explain.

Yes, I'm currently using MultEQ-X as the EQ device for the main channels (removed the MiniDSP HD's for the LCR's as I don't need them anymore). But there's nothing "auto" about it. While the 30+ mic positions MultEQ-X allows is a dramatic improvement over prior versions at getting a repeatable spacial average, IMHO it still falls short (especially in the bass and transition regions) of the consistency attained with MMM.

So, I'm using the MMM with a Umik-1 and calculating corrections based on those measurements. I then use those corrections to tell MultEQ-X how to shape the FIR filter for each speaker. The MQX target curves for the above calibration look like this:

4cLRTargets.png



MQX actually does a pretty decent job over 500 Hz. The largest deviations at high frequency are due to differences in Mic calibrations (yes, I measured them). But in the bass/transition region, even 30 mic positions doesn't hold a candle to MMM, so much larger corrections are needed. That file has 102 PEQ's in it to shape the FIR filters currently. There is no limit to how many can be added.
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
3,635
Likes
2,687
Those 44% are not what we perceive (which we don't really know) but the way to approximate the measured PIR. As Toole says our hearing apparatus and processing is not the same as a typical omnidirectional mic and a computed linear transfer function.
Then why was that used in the CTA-2034 standard? Is it just the best they can do?
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
3,635
Likes
2,687
MQX actually does a pretty decent job over 500 Hz. The largest deviations at high frequency are due to differences in Mic calibrations (yes, I measured them). But in the bass/transition region, even 30 mic positions doesn't hold a candle to MMM, so much larger corrections are needed. That file has 102 PEQ's in it to shape the FIR filters currently. There is no limit to how many can be added.
Interesting. You must be measuring a wider area. I get nearly perfect bass for just me, which is what I care about most, with the 8 positions Audyssey provides.

If I needed to expand that to get a better average of my other seats, I would need quite a few more.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
6,151
Likes
6,122
Location
Brussels, Belgium
@thewas

I was wondering if you're aware of any studies that calculated the delay required for sound to be registered as echo? Obviously this would be frequency specific and I was thinking that this would probably be the best approach for room EQ above ~100Hz.
 
Top Bottom