• 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.

Share your in-room measurements?

Joined
Oct 9, 2020
Messages
30
Likes
81
Just joined to say thank-you to all that have contributed to this thread, I found it useful comparing my DIY results with others..
So in fairness , here is my contribution. My room has 11 sides all up and is as far from a conventional room as you get...Im pretty happy with the results ..My legend is as follows. red is stereo at listening pos average of 4. Green is Direct at LP average of 4, Blue is Direct at Bar average of 3, Orange is overall average across both listening positions. All 1/6. Only threw the Stereo/Audessy one in there because occasionally its used..
movedsub.jpg
 
Joined
Jun 29, 2020
Messages
76
Likes
106
Not sure how good the measurements are since I only just got my UMIK this week, but : JBL308p mkII, 6 measurement average around listening position(I'll play around with the MMM in the future), 1/6 smoothing:
View attachment 83942
No EQ, speakers set to HF trim 0dB and boundary EQ 0dB. Overall mostly makes sense given they're in my pretty small home office room.

I'll probably wait with trying to EQ them until I have get my subwoofer next month.
Got the subwoofer in.

JBL 308p mk. II + SVS SB12-NSD - same methodology, 6 measurement average around listening position, 1/6 smoothing, but this time EQ to a generic harman house curve. All EQ generated by REW, don't have the time to play with it too much, and I want to decide first if I like the slope or want to go flatter.

308+SB12.png

But I'm pretty pleased with the results so far.
 

Zedly

Member
Forum Donor
Joined
Mar 9, 2020
Messages
95
Likes
102
I'm using Klipsch ProMedia 2.1 THX connected to my office computer. They're cheap and low-end, but I've been impressed by the response and sound quality I've been able to get from these.

2020-10-10 13_37_52-REW V5.20 Beta 61.png
 
Joined
Aug 13, 2020
Messages
58
Likes
68
Canton CT-120 with Audyssey EQ below 400Hz. No subs. Denon 3600H into Denon POA-2200 power amp. Green is corrected, red uncorrected. Bad positioning I’m trying to address. Pretty amazed with 40 year old speakers that are this flat from 20 to 20000!

The room is 23’x24’, carpeted.

284E2DB9-E54A-4DC6-9F71-E44647296051.jpeg
 
Last edited:

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
1,491
Likes
1,911
Looking for help from those with more acoustics knowledge about a problem that seems to have manifested recently. I'm getting some weird peaking in the 350-1,500Hz region that I know for sure didn't use to be there, and it's really bugging me. Only reason I remeasured is because I upgraded to Dirac 3 (from Dirac 1), and I wanted to see the differences in the bass response; otherwise, I wouldn't have noticed.

Below is MMM measurement. Dirac Live 3 is used to EQ everything below 300Hz. I've circled the problem area in green. Blue line is drawn in paint to show what I would consider to be a perfect response in my room. The problem didn't exist last year when I measured, and if anything there was a bit of a suckout in that same region. Honestly, I probably wouldn't care if I didn't know for sure that the problem didn't use to exist. It's also in perhaps the most critical range, and those are ~4dB peaks/dips, so I'm sure it's unfortunately very audible :(.

350HzProblem.jpg



I'm trying to think what could have possibly changed. Things I can think of:

1. There use to be a pretty heavy single seat recliner behind and to the right of the speaker. I've attached a photo. You can see there is just a small office chair there now.

2. GIK accoustic panels may have changed position slightly. I recently moved 12 of my GIK panels(ranging from 4" to 8") into the garage for an hour or so, just as a test to see if it would tame the overly bright sound of that system(it did). I forgot to take before pictures, though, so I had to rely on memory to try and put all the panels back in the exact same spot. I think it's mostly the same, but it's possible the panels are in slightly different positions. Could small differences really lead to that big of a change?

3. Acoustic curtains. You can kinda see in the attached photo(behind the office chair). I'm pretty sure those were present in my old measurements, but I can't say I'm a 100% sure.

4. Speaker placement and toe in might be ever so slightly different, but should be within an inch and a few degrees. Recently pulled it out to the middle of the room to do a mono comparison with my new Genelec speakers. I did put tape down to mark where it was before, but it's impossible to line it up "exactly". I highly doubt a tiny position difference could create such a large problem, but my acoustic knowledge is very limited.

5. Upgraded from Dirac 1 to Dirac 3. I'm only EQing up to 300Hz, though. Is it possible that the bass EQ is somehow affecting the ranges above?

Those are the only changes I can think of, and I honestly don't see how any of those could create such a drastic response error.

What can I do to try and fix this?

As mentioned, I've got a bunch of GIK panels I can work with. I don't care at all about aesthetics(as you can probably tell by the panels covering the fireplace :p), so I can move them around anywhere I want. I've also got other panels I'm using in my office, garage, and bedroom that I could move to the main room, or I could always buy more. How effective would panels be for addressing peaking in this area?

I could also try extending the range of Dirac's EQ. Right now, I've got it limited to 300Hz, but I could probably bring that peak down by extending that to 1.5kHz or so. I'm a little hesitant to do this though after I test I did last year where myself and friends all preferred the limited range EQ over full range EQ. That test was done with Dirac 1, though, so maybe Dirac 3 is better at full range EQ. Writing it out now, I'm thinking this is something that I definitely need to test again, so I'll probably be trying this no matter what.

Finally, I guess it could be the speakers themselves, but that doesn't explain why it didn't use to be there. Also, there is a spin available, and I don't see anything that could cause this. There is a directivity error just below 500Hz(where the horn stops controlling directivity), but this starts at 350Hz.

Any other ideas?
 

Attachments

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
1,491
Likes
1,911
Canton CT-120 with Audyssey EQ below 400Hz. No subs. Denon 3600H into Denon POA-2200 power amp. Green is corrected, red uncorrected. Bad positioning I’m trying to address. Pretty amazed with 40 year old speakers that are this flat from 20 to 20000!


View attachment 87114
Impressive bass for a system without subs :cool:.
 

Pio2001

Active Member
Joined
May 15, 2018
Messages
234
Likes
325
Location
Neuville-sur-Saône, France
Hi,
I can see that there are no two graphs in this topic that are displayed at the same scale. It makes them nearly impossible to read, compare or comment.
On top of that, some are smoothed, some are not,

We should decide about the ideal horizontal / vertical ratio for our graphs, and also choose a proper smoothing.

VAR smoothing in REW is interesting because it is very close to the averaging that we naturally get from an MMM around one listening position.
But it would probably be better to actually use MMM without smoothing, so that the frequency response problems of the speaker remain visible in high frequencies.

The problem about the H/V ratio is more difficult because REW doesn't provide a way to define such a ratio. For given graph axis limits, the result completely depends on the size of REW's own window on the computer's desktop.
 

Chromatischism

Addicted to Fun and Learning
Forum Donor
Joined
Jun 5, 2020
Messages
957
Likes
671
Hi,
I can see that there are no two graphs in this topic that are displayed at the same scale. It makes them nearly impossible to read, compare or comment.
On top of that, some are smoothed, some are not,

We should decide about the ideal horizontal / vertical ratio for our graphs, and also choose a proper smoothing.

VAR smoothing in REW is interesting because it is very close to the averaging that we naturally get from an MMM around one listening position.
But it would probably be better to actually use MMM without smoothing, so that the frequency response problems of the speaker remain visible in high frequencies.

The problem about the H/V ratio is more difficult because REW doesn't provide a way to define such a ratio. For given graph axis limits, the result completely depends on the size of REW's own window on the computer's desktop.
For years I've used 45-105 dB as suggested in Austin Jerry's guide.

Horizontal axis:
15-300 Hz for bass
15-20,000 for full-range

And, with John's recent updates to the graph export feature, specifying the output resolution (I set 1920 width as standard) you will always get 5 dB increments.

I agree with either Variable smoothing for full-range, or none. Bass always gets zero smoothing.
 

Pio2001

Active Member
Joined
May 15, 2018
Messages
234
Likes
325
Location
Neuville-sur-Saône, France
Thank you for your answer. It's been some time that I stopped using the export buttom and that I've been taking raw screen captures...

But I was wrong : REW does allow to specify the aspect ratio. This free software will never stop to amaze me!

Test export REW settings.png


This way, the graph limits (45-105, 15-20k) are not critical.
25 dB / decade seems to be the right ratio for me, at least for full range graphs.

But I think that 1920 pixels for the image width might be too large. The text written along the axis become small, and it gives an unusual grid that has 0.5 dB per division, and 2 dB per section:

Testexport 25.png


Maybe an image width of 800 (still 25 dB / decade) would be better for the forums. The grid then uses 1 dB per division and 5 dB per section:

Test export 25 - 800.png


And since there is a lot of wasted space in the image, we can lower the graph near the bottom and crop the top part after it is exported (so as to keep the frequency numbers at the bottom of the picture):

Test export 25 - 800 bottom.png


Now it looks fine, at least on my laptop, 32" screen, 1920x1080 desktop resolution, webpage 100% zoomed.

The above graph is an MMM (pink periodic noise 64 K, moving microphone, 1/48e oct RTA, 64 k FFT). Below is the same graph with VAR smoothing:


Test export 25 - 800 bottom VAR.png


I think that the part between 150 and 1000 Hz is easier to read on the VAR graph. I mean, I may be wrong, but I expect it to be a better representation of what I'm hearing.

On the other hand, VAR smoothing conceals the 15 kHz peak, that is a relevant information, as it is caused by the speaker itself, as we can see in its anechoic measurement (Neumann KH-120): https://www.audiosciencereview.com/...asurements-take-two.11323/page-22#post-425238
 

Pio2001

Active Member
Joined
May 15, 2018
Messages
234
Likes
325
Location
Neuville-sur-Saône, France
As plotted ?
Then the result varies according to the size of your REW window.

Here is the exact same curve, restricted to 15 - 20k / 45 - 105 limits, exported two times with an image width of 800, "as plotted".
The first time from an extremely shallow REW window, the second time from an extremely narrow REW window.

The result can be anything. We need to constrain the aspect ratio, so that all curves are displayed the same way, without looking too flat, or too deep.

TestExport as plotted.png

TestExport as plotted 2.png
 

Chromatischism

Addicted to Fun and Learning
Forum Donor
Joined
Jun 5, 2020
Messages
957
Likes
671
Then the result varies according to the size of your REW window.
That's why I always specify 1920 width :)

The result is my screenshot above. I've never had a result like you're showing.

I'll play with it some later to see if I can recreate your issue.
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
1,491
Likes
1,911
For years I've used 45-105 dB as suggested in Austin Jerry's guide.

Horizontal axis:
15-300 Hz for bass
15-20,000 for full-range

And, with John's recent updates to the graph export feature, specifying the output resolution (I set 1920 width as standard) you will always get 5 dB increments.

I agree with either Variable smoothing for full-range, or none. Bass always gets zero smoothing.
Ok, I re ran measurements this morning and I formatted the graph to your specifications, here. Var smoothing, 15-20,000, 5 degree increments and 50dB scale, 1920 width.

RightSpeakerNew.jpg


The problem still shows(peaking from 330-1,500Hz) :(, though I guess it's a little less obvious than it was at 1/12 smoothing. I suppose I could keep my EQ only under 300Hz strategy and still fix the problem by bumping up everything below 300Hz by 2-3dB? I'm not overly concerned with headroom.
 
Joined
Sep 14, 2018
Messages
32
Likes
23
These are some uncorrected measurements I had taken of my desktop Dynaudio BM6a mkii active monitors. I'll have to measure again as the configuration of the system has changed a bit since these measurements.

BM6A mkII (Desktop).jpg
 

Chromatischism

Addicted to Fun and Learning
Forum Donor
Joined
Jun 5, 2020
Messages
957
Likes
671
Ok, I re ran measurements this morning and I formatted the graph to your specifications, here. Var smoothing, 15-20,000, 5 degree increments and 50dB scale, 1920 width.

View attachment 87342

The problem still shows(peaking from 330-1,500Hz) :(, though I guess it's a little less obvious than it was at 1/12 smoothing. I suppose I could keep my EQ only under 300Hz strategy and still fix the problem by bumping up everything below 300Hz by 2-3dB? I'm not overly concerned with headroom.
Hmm. The first thing I notice is you didn't set your vertical scale to 45-105 dB. But I'm not seeing any peaking - I'm seeing destructive interference in the 125-300 Hz area, common with floors/walls/ceilings in our rooms. We can't fix those with subwoofers so we have to use placement and bass traps. Overall though your response looks really good but is a little jagged in the bass.

Question for you and @Pio2001:

Are you running REW in a maximized window? I always do. My laptop screen resolution is 1920x1080, or 16x9. I think nearly all monitors have converged on 16x9 at this point, assuming you aren't using an ultra-wide on a desktop. Therefore I don't see a time when you'd have an awkwardly-sized window.

The reason I chose 1920 is for future proofing. Ever looked back at old images in dismay at how tiny they are on your screen because you used a lower resolution back then? So I'm attempting to avoid that. I'll get the laptop out and take a screen shot of my settings in a minute.
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
1,491
Likes
1,911
Hmm. The first thing I notice is you didn't set your vertical scale to 45-105 dB. But I'm not seeing any peaking - I'm seeing destructive interference in the 125-300 Hz area, common with floors/walls/ceilings in our rooms. We can't fix those with subwoofers so we have to use placement and bass traps. Overall though your response looks really good but is a little jagged in the bass.

Question for you and @Pio2001:

Are you running REW in a maximized window? I always do. My laptop screen resolution is 1920x1080, or 16x9. I think nearly all monitors have converged on 16x9 at this point, assuming you aren't using an ultra-wide on a desktop. Therefore I don't see a time when you'd have an awkwardly-sized window.

The reason I chose 1920 is for future proofing. Ever looked back at old images in dismay at how tiny they are on your screen because you used a lower resolution back then? So I'm attempting to avoid that. I'll get the laptop out and take a screen shot of my settings in a minute.
Thanks for the reply!

I couldn't use a 45-105dB scale as part of the graph would be off the screen :p. I guess I could measure quieter, but I have a fairly high noise floor with AC and refrigerators, so I figured louder would be more accurate.

I can see why you think it's more of a 125-300Hz problem, and there definitely are some interference issues going on there. You're probably right that I can fix those with careful placement of absorption. Still, though, I see it more as a peaking issue between 330-1,500Hz. I'm going for a Harman curve, which is not a straight downward slope from 20-20,000Hz. The 125-300Hz region, although jagged, is mostly following the correct curve.

A lot of that fall off you see between 100 and 300Hz is there on purpose to adhere to the Harman curve, which gets most of it's 8-10dB slope in the bass and top octaves. Below is the same photo with a blue line drawn in to show the Harman curve I'm going for. It highlights the excess energy in that region I mentioned. Again, not a huge problem, but it's just frustrating because I know it didn't use to be there, and I don't know what's changed.

RightSpeakerNew copy.jpg
 
Top Bottom