• 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.

Review and Measurements of Lyngdorf RoomPerfect EQ

BYRTT

Active Member
Patreon Donor
Joined
Nov 2, 2018
Messages
148
Likes
175
Location
Denmark (Jutland)
#81
Thanks for posting those but you shouldn't set the bottom of the graph to 0 dB. That is the quickest way to artificially make the graph flat. :) Noise floor in your room is at least 30 dB so you want it at least that high. Comparing REW graphs is impossible if scales are not the same.

Oh, you're right, I wasn't thinking about that. And yes, noise is app at 30dB, even higher at both ends of the audible spectrum.....

Somebody had to do it then :)
1000.png
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,911
Likes
1,378
Location
Zg, Cro
#82
I think on the center vs two ear measurements it makes less difference than many think. Even over a somewhat wider area.
Here are some sweeps in REW done 1ft right and 1ft left of the center LP. 1/24th smoothing. Looks uneven especially below 1 khz.
View attachment 22120

But it is worth remembering this isn't how it will sound. Our ears do about 1/6th octave smoothing at the highest frequencies. And then 1/5th, 1/4th and so on until it is 1/2 octave smoothed in the bass. Here is the same measures 2 ft apart at 1/6th octave smoothing.
View attachment 22121

For the most part above 1 khz there is only a fraction of db difference. I'm not showing it because it clutters the graph, but the center measurement between these two points for the most part splits the difference in the two measurement points. So if a REQ measure was done at the LP it isn't going to be grossly wrong a foot or so either side of it. It would be even less difference 4 inches each side of the center measuring position.

Next are the same two points with Effective Rectangular Bandwidth based smoothing which mimics the smoothing our ears do. It is a little closer still and much better below 1 khz. This is all with the 500 msec IR window to preserve LF resolution.

View attachment 22122
Very interesting! So these are sweeps of both speakers playing taken at different points lef and right from the center of LP?

Scale is quite large so the response is more linear than it looks on the first sight, but what is more impotant is that differences are indeed small.
I'll try to make this measuerement with my speakers as soon as my son wakes up. As he has just finished his exams for this semester this can take a while.. :D
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
18,975
Likes
19,260
Location
Seattle Area
#86
Wow, thank you for the effort you took to make this! I will repeat the measurement with one speaker and kindly ask if you can join it with @amirm graph once more as you get more relevant bass response when measuring both speakers.
Sure although I will be returning the Lyngdorf tomorrow so it will be without it....
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,911
Likes
1,378
Location
Zg, Cro
#87
Sure although I will be returning the Lyngdorf tomorrow so it will be without it....
I was thinking he uses the same graph you already made and overlay it with the sine sweep measurement of my both speakers playing (the way you did your measurement), so we can have more meaningful comparison.
 

Juhazi

Active Member
Joined
Sep 15, 2018
Messages
217
Likes
156
Location
Finland
#88
I haven't done many measurements with LR playing together, but I found this from five years back. I was very careful to get the mic centered. This measurement is in front of my listening spot, there bass gets boosted because of the wall (gyproc). I have intentionally set different response curves (voicing) and also tried to eq sharp bumps in semi-nearfield and also farfield, and seems like I like it more without room-eq corrections.

I have four-way speakers with sealed woofer, others dipole. No room eq at all, except litfed lowest bass like linkwitz transform. I use Minidsp 4x10HD and I have had different upper mids and tweeters, this version had B&G Neo8 upper mid and Fountek NeoCD3.5H as tweeter. Woofer is SEAS L26ROY and mid a 12" Beyma in all versions. Amplification with B&O ICEPower modules.
ainogneo v11 l r lr spl 500ms 112.jpg
 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,911
Likes
1,378
Location
Zg, Cro
#90
I think on the center vs two ear measurements it makes less difference than many think. Even over a somewhat wider area.
Here are some sweeps in REW done 1ft right and 1ft left of the center LP. 1/24th smoothing. Looks uneven especially below 1 khz.
View attachment 22120

But it is worth remembering this isn't how it will sound. Our ears do about 1/6th octave smoothing at the highest frequencies. And then 1/5th, 1/4th and so on until it is 1/2 octave smoothed in the bass. Here is the same measures 2 ft apart at 1/6th octave smoothing.
View attachment 22121

For the most part above 1 khz there is only a fraction of db difference. I'm not showing it because it clutters the graph, but the center measurement between these two points for the most part splits the difference in the two measurement points. So if a REQ measure was done at the LP it isn't going to be grossly wrong a foot or so either side of it. It would be even less difference 4 inches each side of the center measuring position.

Next are the same two points with Effective Rectangular Bandwidth based smoothing which mimics the smoothing our ears do. It is a little closer still and much better below 1 khz. This is all with the 500 msec IR window to preserve LF resolution.

View attachment 22122

Here is how it looks with my speakers. These 2 measurements were taken 1.5 feet of the left and right from the center of LP. They are not as consistent as yours, but still pretty similar.

(1/6 smoothing)



Same thing shown with the measurement from center of the LP:

(L and R 1/6 smoothing, center of LP 1/12 smoothing)



This one may also be interesting, it shows sweep from center of LP vs RTA with 43 averages taken in the region +/-1.5 feet left and right from the center of LP:

 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,911
Likes
1,378
Location
Zg, Cro
#91
I haven't done many measurements with LR playing together, but I found this from five years back. I was very careful to get the mic centered. This measurement is in front of my listening spot, there bass gets boosted because of the wall (gyproc). I have intentionally set different response curves (voicing) and also tried to eq sharp bumps in semi-nearfield and also farfield, and seems like I like it more without room-eq corrections.

I have four-way speakers with sealed woofer, others dipole. No room eq at all, except litfed lowest bass like linkwitz transform. I use Minidsp 4x10HD and I have had different upper mids and tweeters, this version had B&G Neo8 upper mid and Fountek NeoCD3.5H as tweeter. Woofer is SEAS L26ROY and mid a 12" Beyma in all versions. Amplification with B&O ICEPower modules. View attachment 22143
This looks excellent, I'm sure it sounds equally well! :)
 

BYRTT

Active Member
Patreon Donor
Joined
Nov 2, 2018
Messages
148
Likes
175
Location
Denmark (Jutland)
#92
Here is a sweep of both speakers playing (same as Amir did), can you please join this graph as the comparison will be more meaningful?...

Sure here we go...
1001.png


After some small interpolation or prediction of those curve's stopband tails and omiting IIR XO filter excess phase... (in comparison someone is kind of basshead :))

1001b.png


"Destroy" seems like a more appropriate word to me! :D
Now convolute a few tracks below curve/IR to get "Destroy" response and if we lucky sound of carpet called "modern shag" is in there too :D

1001c.png
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,911
Likes
1,378
Location
Zg, Cro
#93
Wow, thanks a lot!

Btw, my speakers are in my living room and wife would never agree to change carpet she picked up recently only to have better sound! :D

Anyhow, i don't watch movies with Harlechs, only music, and not Bach or other organ stuff, so bass response from 32Hz actually works pretty well for me as it covers practically all instruments including jazz bass guitar. :)

And while it si definitely true that Amir Salons destroyed my Castles bass wyse I believe it is also true that my manual room EQ is much more linear than Lyngdorfs RoomPerfect EQ! :p
 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,911
Likes
1,378
Location
Zg, Cro
#94
Sure although I will be returning the Lyngdorf tomorrow so it will be without it....
@amirm Wouldn't it be interesting to have a comparison between RoomPerfect EQ and Dirac in terms of how they managed to correct the response of your room/speakers?

You only need to measure Dirac filters with both speakers response in the same way you did for RoomPerfect EQ and overlay them and we'll se which one of them did a better job.. :cool:
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
18,975
Likes
19,260
Location
Seattle Area
#95
@amirm Wouldn't it be interesting to have a comparison between RoomPerfect EQ and Dirac in terms of how they managed to correct the response of your room/speakers?
Hard to do a proper comparison without more trials for each system. I think the owner wants his Lyngdorf back soon. If he doesn't, I can give it a shot regardless.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,911
Likes
1,378
Location
Zg, Cro
#96
Hard to do a proper comparison without more trials for each system. I think the owner wants his Lyngdorf back soon. If he doesn't, I can give it a shot regardless.
I don't really think you need to hold Lyngdorf longer to do that. As far as I was able to understand you did setup of RoomPerfect EQ exactly according to the manufacturers procedure and the software created filters in a way it was supposed to do. If you do the same procedure with your Dirac software (those 9 points in which measurements are taken) I believe we'll have a fair comparison between the two.

What do you say? :)
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
18,975
Likes
19,260
Location
Seattle Area
#97
I don't really think you need to hold Lyngdorf longer to do that. As far as I was able to understand you did setup of RoomPerfect EQ exactly according to the manufacturers procedure and the software created filters in a way it was supposed to do. If you do the same procedure with your Dirac software (those 9 points in which measurements are taken) I believe we'll have a fair comparison between the two.

What do you say? :)
My Lyngdorf measurements were with two channels driven so I would have to re-test it with one channel at a time at least.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,911
Likes
1,378
Location
Zg, Cro
#98
My Lyngdorf measurements were with two channels driven so I would have to re-test it with one channel at a time at least.
Sure. Actually I would expect that the setup procedure is always such that you measure each speaker separately and only to control the filter effects with both speaker playing, in addition to control each speaker response as well.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
18,975
Likes
19,260
Location
Seattle Area
#99
Sure. Actually I would expect that the setup procedure is always such that you measure each speaker separately and only to control the filter effects with both speaker playing, in addition to control each speaker response as well.
I didn't mean that. Lyngdorf's calibration was done one speaker at a time. My REW measurements of the same were not.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,911
Likes
1,378
Location
Zg, Cro
I didn't mean that. Lyngdorf's calibration was done one speaker at a time. My REW measurements of the same were not.
Are you saying that you measured response of the both speakers with REW when you were doing control of the effects of Lyngdorf's calibration?

If that is the case IMO that is ok as with LF it is the response of both speakers that matters. Don't you think so?
 
Top Bottom