Share your in-room measurements?

[Albert Dagger]

You could make your own quasi-anechoic measurements:

How to make quasi-anechoic speaker measurements/spinoramas with REW and VituixCAD

Quasi-anechoic measurements are a way to capture the 'real' frequency response of a speaker without the influence of a room. They are called 'quasi-anechoic' because they have the potential to get you really close to the results one might achieve with a non-reflective environment like an...

www.audiosciencereview.com

Even if the speakers are flat on-axis, they might have a directivity error causing off-axis brightness.

You can use a calculator. It should only take a minute if you already know the dimensions of your room.

amroc - THE Room Mode Calculator

HTML5 room mode calculator. Read and hear the tune of axial, tangential and oblique modes. Scientific sources included.

amcoustics.com

A short-term solution could be to limit the correction to below the transition frequency and apply a cutting shelf filter (or a low Q peak filter) to the treble to tame the brightness.

Thank you.
So you mean a cutting shelf filter outside of Audacity. I only have a graphic eq in the receiver outside of that. Or I don’t know how to do it.

 

[flipflop]

Thank you.
So you mean a cutting shelf filter outside of Audacity. I only have a graphic eq in the receiver outside of that. Or I don’t know how to do it.

Use whatever tools you have at your disposal. It doesn't matter if you use Audyssey or a GEQ. What's most important is that you don't boost frequencies in the midrange and treble before you can get your hands on some anechoic data to guide you in making an informed decision.

 

[Kaameelis]

This is how they look with 1/48 smoothing. I've only captured 5K to 20K.
These were measured with a Umik1, 90 degrees, and associated calibration file.

View attachment 318963

I understand you wrong.
I thought you mean decay as this, and there is also after 10 kHz fast decay.

 

[ExPerfectionist]

I wouldn't mind having a tick rug, which makes a difference, but I can also live without.

A thick rug and/or a rug with a thick pad would definitely help if you currently have bare hardwood floors.

 

[Eleo]

Hi there! I'd like to share some measurements of my Lintons. The chain is as follows: Minipc fanless with Eq Apo => SMSL C200 =>Topping L 30 => ZeroZone Irs 2092 Monoblocks => Wharfedale Linton (I also have a tube preamp, but it was not used in measurements). I recently added a small Yamaha sub to fix a nasty dip between 60 and 90 Hz. The result was really amazing. Here are the measurements before and after. In the last image, I have added some EQ above 6K Hz. Notice that, although Infineon Irs 2092 amps should have a fair bit of load dependency, there is no sign of rise in the highs. I think that it depends on the fact that the Lintons have an almost flat impedance above 5K Hz.

 

[Sokel]

Hi there! I'd like to share some measurements of my Lintons. The chain is as follows: Minipc fanless with Eq Apo => SMSL C200 =>Topping L 30 => ZeroZone Irs 2092 Monoblocks => Wharfedale Linton (I also have a tube preamp, but it was not used in measurements). I recently added a small Yamaha sub to fix a nasty dip between 60 and 90 Hz. The result was really amazing. Here are the measurements before and after. In the last image, I have added some EQ above 6K Hz. Notice that, although Infineon Irs 2092 amps should have a fair bit of load dependency, there is no sign of rise in the highs. I think that it depends on the fact that the Lintons have an almost flat impedance above 5K Hz.

It seems nice down low but why correct above 6k?
Here's a guide from an absolute expert:

What is your favorite house curve

OK. Next question for you folks. As far as I can tell there is no spinorama for the Focal Aria 926's, which happen to be my main speakers. However, Amir did do a spinorama for the Aria 906 standmount speakers. They use the same tweeter and mid driver as the 926, but lack the two woofers as they...

www.audiosciencereview.com

You can dive even deeper to it but the above is a good way to see things.

 

[Albert Dagger]

You could make your own quasi-anechoic measurements:

How to make quasi-anechoic speaker measurements/spinoramas with REW and VituixCAD

Quasi-anechoic measurements are a way to capture the 'real' frequency response of a speaker without the influence of a room. They are called 'quasi-anechoic' because they have the potential to get you really close to the results one might achieve with a non-reflective environment like an...

www.audiosciencereview.com

Even if the speakers are flat on-axis, they might have a directivity error causing off-axis brightness.

I tried to perform these speaker measurements indoors. I only did one measurement on-axis and one at 30 degrees. It was a quick test.
I wasn't sure about the position of the mic, so I pointed at the tweeter, at 1m distance, but I think this has impacted the response in the mids (the crossover between mid and tweeter is 3.5KHz in the Monitor Audio Gold 200 5G). I used the correct mic calibration file.

This is a screenshot of the measurements, green is on axis and blue is at 30 degrees. I've windowed the results at 7ms, and this screenshot is captured with 1/24 smoothing. I'm pretty sure these are not accurate, but at least the two measurements show somewhat the difference between on-axis and off-axis response. The measurements from the listening position seems to be aligned with the on-axis measurement (obviously only at 500 and above).

I will try to do this more scientifically when I have time, and I will do all the steps described in the post you quoted. Thank you.

 

[Albert Dagger]

A short-term solution could be to limit the correction to below the transition frequency and apply a cutting shelf filter (or a low Q peak filter) to the treble to tame the brightness.

I changed my Audyssey settings to correct the response up to 350Hz, and then stop. These are the results. Most of the room modes, peaks and dips between 20 and 300Hz are corrected. Some are not corrected but the result is so much better than the peaks and dips before Audyssey.

The pink/purple speakers are the surround left and surround right. The other three speakers are the fronts and centre, which have a slight tilt up between 1 and 6KHz. Obviously, without correcting above the transition frequency, this aspect cannot be fixed. When I purcahse a MultEQ X license, I will try adding a target curve to that portion. With the pc MultiEQ X app, one can constrain room correction to low frequencies but still apply parametric eq to other frequencies, so I can try that for the fronts and centre, as you suggested above.

 

[garyrc]

This is the only nearly full range (20Hz to15KHz) sweep I have at the moment (the old ones were done through a defective mic preamp and an ill suited microphone). I about had it with some early CDs (& 1 Blu-ray) that were incredibly weak in the bass, so I cranked up the bass and futzed around, then measured with my new USB mic. So this is my special in-room curve for bass-shy recordings, one mic position at 13 feet (listening position), in a 4,200 cu.ft. room with Klipschorns (A5) and Belle Klipsch center (all the same drivers, with a K401 mid horn put on the Belle to match the Khorns), a Klipsch RSW 15 sub, with Audyssey Flat and tone control manipulation by ear as seen, response averaged, with REW psychoacoustic smoothing.

Now sounds "live" with great dynamics and bass.

 

[Eleo]

It seems nice down low but why correct above 6k?
Here's a guide from an absolute expert:

What is your favorite house curve

OK. Next question for you folks. As far as I can tell there is no spinorama for the Focal Aria 926's, which happen to be my main speakers. However, Amir did do a spinorama for the Aria 906 standmount speakers. They use the same tweeter and mid driver as the 926, but lack the two woofers as they...

www.audiosciencereview.com

You can dive even deeper to it but the above is a good way to see things.

I understand that the prevailing opinion is to apply no eq above the Schroeder frequency and in fact my Rew made filters extend only to 300 Hz. However I thought to add some manual filters that I can enable/disable with a click in order to see what is like to have a bright speaker. It can be interesting to compare the two settings although I acknowledge that the first is more natural

 

[bitdo]

I am looking for some advice with cross correlation results of my speaker measurements (5.1.2). I am using @OCA 's methods/videos to extract Audyssey measurements out of a MultEQ ady json file and imported them into REW. My height speakers are the silly kind that reflect off the ceiling and they seem to be 180 degrees out of phase in the measurements. Using vanilla Audyssey or manually generating my own filters using OCA 's methods, the Atmos channels have so far been a let down. Perhaps I just need to get a pair of proper height speakers and mount them up high on the front wall. And my space has lots of echos/reflections (needs a lot more room treatment).

The first picture show 5 speakers aligning more-or-less nicely. The 2nd picture adds the 2 height speakers that seem 180 degrees out of phase. And 3rd picture shows the sub relative to all the speakers. These are the results or using REW to auto cross correlation.

1. Should I just swap the speaker terminals on my height speakers and try again?
2. Does my sub cross correlation look correct?

Measurements done with Denon AVR-X3800H's measurement mic pointed 90 degrees (at ceiling)
Front and Height: Klipsch RP-280FA (atmos speakers built into top of the tower)
Center Klipsch RP-504C (in a BDI Corridor cabinet with doors closed)
Surround Klipsch RP-402S
RSL Speedwoofer 10S MKII

Thank you!

p.s. @OCA thank you for the guides! keep the videos coming!

 

[TurtlePaul]

Here we go. I have been playing with a MiniDSP 2x4 HD, and had time to tweak my setting today. Here is where I ended up. Note that I am using speakers without waveguides, so the on-axis sound isn't really flat to 2 khz, and a downward sloping in-room response sounds very dark. I am also targeting for 'apartment bass' with no low bass hump to appease my neighbors (I also like luscious mids without too much lows). 1.6 smoothing.

 

[OCA]

I am looking for some advice with cross correlation results of my speaker measurements (5.1.2). I am using @OCA 's methods/videos to extract Audyssey measurements out of a MultEQ ady json file and imported them into REW. My height speakers are the silly kind that reflect off the ceiling and they seem to be 180 degrees out of phase in the measurements. Using vanilla Audyssey or manually generating my own filters using OCA 's methods, the Atmos channels have so far been a let down. Perhaps I just need to get a pair of proper height speakers and mount them up high on the front wall. And my space has lots of echos/reflections (needs a lot more room treatment).

The first picture show 5 speakers aligning more-or-less nicely. The 2nd picture adds the 2 height speakers that seem 180 degrees out of phase. And 3rd picture shows the sub relative to all the speakers. These are the results or using REW to auto cross correlation.

1. Should I just swap the speaker terminals on my height speakers and try again?
2. Does my sub cross correlation look correct?

Measurements done with Denon AVR-X3800H's measurement mic pointed 90 degrees (at ceiling)
Front and Height: Klipsch RP-280FA (atmos speakers built into top of the tower)
Center Klipsch RP-504C (in a BDI Corridor cabinet with doors closed)
Surround Klipsch RP-402S
RSL Speedwoofer 10S MKII

Thank you!

p.s. @OCA thank you for the guides! keep the videos coming!

View attachment 319713

View attachment 319714View attachment 319715

It's quite common to measure inverted impulse responses with surround speakers and a vertical mic. Although it's possible that they need physical polarity inversion, usually if all speakers are of the same brand and you're certain all connections are correct, you should leave them alone. Best way to check for inverted polarity is to compare their phase responses relative to the other speakers in Overlays window switching polarity virtually with REW's "invert ploarity" tool and pick the one with the best match.

 

[bitdo]

It's quite common to measure inverted impulse responses with surround speakers and a vertical mic. Although it's possible that they need physical polarity inversion, usually if all speakers are of the same brand and you're certain all connections are correct, you should leave them alone. Best way to check for inverted polarity is to compare their phase responses relative to the other speakers in Overlays window switching polarity virtually with REW's "invert ploarity" tool and pick the one with the best match.

Zooming on the Overlay Phase view at 1kHz, it clearly looks like the 2 height speakers are out of phase with the rest of the speakers. I double checked the wiring and everything looks correct. Everything above 700 Hz and up looks 180 degrees out of phase. The lower frequencies 200-700 its hard to tell.

 

[OCA]

Zooming on the Overlay Phase view at 1kHz, it clearly looks like the 2 height speakers are out of phase with the rest of the speakers. I double checked the wiring and everything looks correct. Everything above 700 Hz and up looks 180 degrees out of phase. The lower frequencies 200-700 its hard to tell.
View attachment 320167

You've to check lower frequencies ie 20-200Hz, higher frequency phase response is mostly related to speaker delay (distance)

 

[snackiac]

This is my result with KH80s and a KH750 using MA1. I've tinkered with positioning a lot to try to fix the 80hz dip, and this is the narrowest I've gotten it. It shows on var smoothing but not with psychoacoustic.

 

[ozzy9832001]

Here we go. I have been playing with a MiniDSP 2x4 HD, and had time to tweak my setting today. Here is where I ended up. Note that I am using speakers without waveguides, so the on-axis sound isn't really flat to 2 khz, and a downward sloping in-room response sounds very dark. I am also targeting for 'apartment bass' with no low bass hump to appease my neighbors (I also like luscious mids without too much lows). 1.6 smoothing.

View attachment 319733

Looks like the tweeters aren't quite aligned that's why the huge drop off from 4k and above...unless that's by design, but seem drastic.

 

[TurtlePaul]

Looks like the tweeters aren't quite aligned that's why the huge drop off from 4k and above...unless that's by design, but seem drastic.

Do you mean between the left and right speakers or between the tweeters and woofers?

Edit:

You are right, there is something phasic between my left and right speakers. Perhaps because the room isn't symmetrical. I wonder if a IIR filter can solve this? There is a cancellation centered on 2.1 khz.

Edit 2: I was measuring to EQ the low end primarily and don’t have a long enough USB-C cord, so I wouldn’t discount that this measurement may have been 3-5” too far left from the main listening position.

 

[ozzy9832001]

Do you mean between the left and right speakers or between the tweeters and woofers?

Edit:

You are right, there is something phasic between my left and right speakers. Perhaps because the room isn't symmetrical. I wonder if a IIR filter can solve this? There is a cancellation centered on 2.1 khz.

Edit 2: I was measuring to EQ the low end primarily and don’t have a long enough USB-C cord, so I wouldn’t discount that this measurement may have been 3-5” too far left from the main listening position.

View attachment 320436

Sorry for the late reply, this weekend has been crazy.

This can happen when the tweeters aren't equidistant from the microphone. Maybe it was too close compared to the angle of toe in (if any at all). The tweeters are fairly directional and quite finicky. Slight variations can have dramatic results. That's not to say that's what you hear mind you. As our ears are not focused on a central point.

 

[thewas]

After some changes here are my desktop system consisting of two old KEF LS50 plus one 10" subwoofer with full bandwidth equalisation

and my couch listening system consisting of two KEF LS50 Meta plus two subwoofers with 2x 8" woofers each and equalisation only up to 500 Hz