Share your in-room measurements?

[TimH]

This is the first time I taken a measurement with REW, I got a UMK-1 mic and gave it a shot - this is the graph I got once calibrated. I'm still learning how to read these so I'm sure there is some tweaking that could be done to improve things. Its for a pair of Meridian DSP 5000s

 

[Vuki]

View attachment 125676
This is the first time I taken a measurement with REW, I got a UMK-1 mic and gave it a shot - this is the graph I got once calibrated. I'm still learning how to read these so I'm sure there is some tweaking that could be done to improve things. Its for a pair of Meridian DSP 5000s

Nice. Just vertical scale should be corrected- usually it's 50dB not 120dB.

 

[TimH]

Thanks Vuki, now updated with the scale corrected. I need to double check these weren't using the centre speaker as well as the processor has a number of different room modes and I can't remember which was active. It will be interesting to see what kind of difference things like Meridians Tri-field setting make

 

[Ingenieur]

Here's mine. Very crude using the spectrum FFT app, pink noise and a frequency sweep.
Also used the ClapIR app. I have a Shiit Loki I used for minor trim.
Room: 34 x 14 x 8
Measured with iPhone X, from what I can find pretty accurate and low distortion up to 14 kHz range.

 

[TimH]

I re-measured the Meridians as I didn't do separate left and right the first time around. I also might have had the centre channel active.....at least this time I know I didn't! Below is the average of 12 measurements. I'm trying to figure out how bad or good these are, they seem to be quite a way off the target in some cases, unfortunately I have no space whatsoever to move the speakers but could probably add some room treatment. The other option I need to try is resetting the Meridian amp as I have some room correction coming from that and some from REW, maybe it would be better to start with a clean slate? A few more speakers will be tested in the space this month so I'm intrigued to see if I can get better results. Would a sub also help out here?

Any advice would be much appreciated!

Tim

 

[Cotina]

Hi all!
First post here.
I've noticed that Linkwitz speakers are known here in ASR, so I thought I should share some measurements of my LX521.4, to whom might be interested.
6.3 x 4.1 x 2.7m living room, with quite some treatment and Dirac Live 3:

View attachment 105271
RT60:
View attachment 105272View attachment 105273View attachment 105274

And an alternative with a little more slope between 1-10kHz:
View attachment 105275

Quite happy with the overall sound, for a 2-channel system in a living room.

I like your post, can you help me whith Dirac? Do you use a Dsp, Asp? I,a using a Nadja Dsp.

 

[AdamG]

I like your post, can you help me whith Dirac? Do you use a Dsp, Asp? I,a using a Nadja Dsp.

Welcome Aboard @Cotina.

 

[aglx]

I like your post, can you help me whith Dirac? Do you use a Dsp, Asp? I,a using a Nadja Dsp.

Hi,
I use Hypex Dlcp as a DSP.
Dirac runs on a Mac mini, as a AU plug in, in Audirvana.
You can find instructions here:

macOS: Set up Dirac Processor Plugin for Systemwide Correction with BlackHole (Free) - Dirac Live Support - Confluence

Dirac Processor Plugin - Quick Start Guide - Dirac Live Support - Confluence

for Windows here:

Dirac Processor Standalone - Quick Start Guide - Dirac Live Support - Confluence

Dirac Live® 3 - Quick Start Guide - Dirac Live Support - Confluence

The setting up procedure is a little frustrating at the beginning, for inexperience users (like me), but it works at the end with some effort.

 

[nick-v]

Pretty wild seeing what's possible with REW and Roon Convolution filters:

These are a pair of Monitor Audio SF2 (Sealed, 3-Way, On-Wall Art Speakers).

And here they are after correction (and some serious shaping). I'm using up a ton of dynamic headroom with the huge low-end boost, but these are installed on the wall in the open concept dining nook/kitchen area and they sound so much better now. They're usable without subwoofer(s) with excellent extension. This system never gets turned up that loud, and it's just staggering what you can so with DSP and EQ these days. This is my first kick at it (only a few measurements averaged pre and post-correction). I'll do a more elaborate set of measurements pre and post correction and see what I can do.

 

[YSC]

Here are mine. I am yet to get these sounding good in the room.

View attachment 64868

Woo! Channel matching seems good but those room modes looks nasty with high peaks and dips at the 100-200hz region

 

[spacevector]

Woo! Channel matching seems good but those room modes looks nasty with high peaks and dips at the 100-200hz region

Indeed. I have changed houses since but new room is similarly small. I have not made measurements yet but expect same challenges, this time also with channel match since room isn't symmetric.

 

[YSC]

This is L+R for my 8030C+7040A with only dip switches tinkered at LP, seems reasonably nice to me without DSP

 

[astcal]

Neumann KH 80 + KH 750, calibrated with Neumann MA-1 kit. Then measured with the MA-1 microphone:

 

[richard12511]

Pretty wild seeing what's possible with REW and Roon Convolution filters:

These are a pair of Monitor Audio SF2 (Sealed, 3-Way, On-Wall Art Speakers).

View attachment 132335

And here they are after correction (and some serious shaping). I'm using up a ton of dynamic headroom with the huge low-end boost, but these are installed on the wall in the open concept dining nook/kitchen area and they sound so much better now. They're usable without subwoofer(s) with excellent extension. This system never gets turned up that loud, and it's just staggering what you can so with DSP and EQ these days. This is my first kick at it (only a few measurements averaged pre and post-correction). I'll do a more elaborate set of measurements pre and post correction and see what I can do.

View attachment 132330

Ha! That is some serious shaping indeed. Software based EQ like that does have great potential.

 

[YSC]

After a day of fiddling with placement and the various dip switches without using any DSP, this is my final setting overlay with REW target curve, seems a bit bass heavy but can't get any better without DSP especially the 85-200hz range where the 8030C bass roll off can only do so much

 

[Ata]

Here are my latest measurements and corrections towards a desired house curve:

I do have a couple of questions:

- What smoothing should I use when generating/tweaking the filters for the target curve in REW, Var/Psy/ERB or other?
- Are there better automated filter generators than REW, what other tools would you recommend?
- I am interested in using FIR convolution, for convenience with Audirvana/Roon and for greater accuracy and flexibility compared to a fixed number of PEQs. Is there any way to generate a convolution filter directly from the (smoothed) FR measurement?

Hope this makes sense!

 

[thewas]

Added a 10" CB sub I had in my storage room to my main KEF LS50 Meta setup today.

L and R measurements without it and only negative PEQ filter room correction below 300 Hz:

L and R measurements with it and only negative PEQ filter room correction below 300 Hz:

The audible difference in deep bass is of course huge when bass starts dropping under 30 Hz vs under 60 Hz.

 

[YSC]

Added a 10" CB sub I had in my storage room to my main KEF LS50 Meta setup today.

L and R measurements without it and only negative PEQ filter room correction below 300 Hz:

View attachment 136606

L and R measurements with it and only negative PEQ filter room correction below 300 Hz:

View attachment 136607

The audible difference in deep bass is of course huge when bass starts dropping under 30 Hz vs under 60 Hz.

yea, especially with the few songs you got that low bass content, for 20hz I didn't feel much desire though

 

[changer]

This is the room response of my DIY 2-way speakers with 12-inch Faital pro woofers, JBL waveguide on Celestion compression drivers and powered by Hypex FA123 modules (pics). Currently, I stopped advancing with the prototypes and decided to just listen and enjoy them for a while in their current state. I applied some room correction to the raw response with the in-build Hypex 8-band PEQ stage.

Measurements were conducted as moving mic measurements with pink noise and 1/48th RTA in REW, of course. The first picture shows individual channels pre-EQ and the corrected stereo sum (calculated, not recorded) of both channels post-EQ. Smoothed to 1/12 octave. PEQs where applied between 48-440 Hertz only.

The second image shows equalized response of individual channels:

Note this is a constant directivity design and the speaker has matched narrow directivity between something like 500-15000 Hertz, hence little room curve. It sounds right, though. Quasi-anaechoic response and other measurements are collected in my "I ask for help" thread over at diyaudio.com, here is the on-axis without room (gated to 4.5 ms):

I am happy to have learned about MMM, because averaging individual sweeps was kind of tricky. It has to be noted though that, due MMM not recording phase and impulse the summed response is RMS and not Vector so without phase and there might be more problems between 200-500 Hertz in my room when speaker response is summed correctly. Will dig into this another time.

 

[Ata]

I am happy to have learned about MMM, because averaging individual sweeps was kind of tricky. It has to be noted though that, due MMM not recording phase and impulse the summed response is RMS and not Vector so without phase and there might be more problems between 200-500 Hertz in my room when speaker response is summed correctly. Will dig into this another time.

I am also a fan of the MMM method as it is sufficiently accurate and much faster than using multiple single points.

Your pre-EQ measurements look great for L+R while L and R individually are much less flat. In particular, in the 40-55 Hz area there may be some phase issue since the averaged results is a lot lower than the room mode for each of the speakers? I am not an expert, merely asking...