• 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

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
20,051
Likes
23,062
Location
Seattle Area
#1
This is a follow up to my review of the Lyngdorf TDAI-3400. While that review was focused on objective measurements of the DAC and amplifier, this review is focused on the performance of the room equalization (RoomPerfect). As much as I like to perform a comprehensive test and review of Lyngdorf against a few of its competitors, I just don't have the time to do that. So instead this is a focused test by itself.

I usually don't want to mess up my my main system because it takes me forever to get it back together. In this case I thought it was important to test the Lyngdorf using the same setup given the high cost of the Lyngdorf. At first I was dreading how much rewiring I had to do, only to realize that the Lyndrof has everything my discrete system has sans speaker! So this is the setup:
Lyngdorf TDAI-3400 RoomPerfect Review.jpg


I took a rather wide shot so that you can get a feel for the room. What you don't see is the rest of the house that is open to this loft area. The volume as such is massive which helps lower room modes to lower frequencies (always a good thing). I normally use Dirac to correct what is left of the bass anomalies up to 200 Hz.

My seating position is between the measurement mic you see poking out of center left and my laptop on the right. This is my own measurement Mic which I used to measure what RoomPerfect had done before and after equalization. The Lyngdorf mic seems higher quality and comes with a long balanced cable. A bit strangely, instead of an XLR input for it in the back, there is an adapter to 3.5 mm which then allows the mic to plug into the front of the unit (this part is convenient).

The Lyngdorf is the black box sitting on my (unused currently) Mark Levinson No 532 power amplifier. For those of you complaining about the cost of the Lyngdorf TDAI-3400, the 523 costs $20,000 by itself! :) Admittedly it has 400 watts using 8 ohm and probably twice as much over 4 ohm so much more powerful than the Lyngdorf. Still, it is just an amplifier.

My everyday amplifiers are the two Mark Levinson No 53 monoblocks flanking the Revel Salon 2 speakers which were used for this testing. Those beasts have 500 watts into 8 ohm and 1000 watts into 4 ohm. In listening tests, the Lyngdorf TDAI-3400 got plenty loud although I had it pretty close to 0 dB at times. Again, this is a huge space so it takes fair bit of power to fill it and shake my seat. :D

To set up the Lyngdorf, I connected it to Ethernet cable and used a browser to configure it. I was pleasantly surprised how well implemented the interface is, feeling just like a mobile app even though it is just a web interface:

Lyngdorf TDAI-3400 RoomPerfect Web interface.png


The setup could not be any easier. You just follow a handful of steps in the guide and you are done.

Initial measurement is for the main seating (sweet spot) which Lyngdorf calls "Focus." From then, you randomly position the microphone in different parts of the room. What is the idea behind this? One of the problems with room equalization is that if you only measure in one spot, you may make the rest of the room much worse. So in applications such as home theater where multiple people may want to listen at once, you this won't work too well. The random measurements attempt to tease out what is a "room behavior" and what is a "speaker response." To the extent the same problems in frequency response show up in multiple random places, then it is a "property of the speaker" and can be safely equalized. Alternatively, such data can inform how much correction should be applied to some peak if it is not shared in other locations.

Prior to calibration a level check was performed. I was told to dial the volume up to -11 dB or some such number. I could not tolerate the tones above 20 or so dB so that is what I used, thinking I would have to re-do them later. Lower volumes allows more background noise to interfere with measurements although the impact is much lower with newer systems.

The actual measurements consist of playing two seperate sweeps of frequencies, one covering low to mid frequencies and the other from mid to highs. Usually there is only one sweep covering the full audio band. I am thinking there is probably two different filtering schemes for low and high frequencies. Low frequencies require resolution than to one hertz or even lower for precise correction (room modes are very narrow for those of you who want to get technical). This can make the filtering expensive so often a different filter is used (e.g. IIR instead of FIR). I am just speculating here. If Lyngdorf has documented this scheme, maybe someone can chime in with the right answer.

After each pass, you are instructed to move the mic to another random location. I casually did that until the system was happy that it had enough data saying it had understood the room 93%. Objectively I think there was enough data there for 92.295% but I let it go with that. :)

Subjective Results
Nearly 20 years ago I had my first experience with room correction courtesy of a now defunct company, TacT. I remember being startled with what I heard. The transformation is incredible. I forever became convinced that room equalization is mandatory in any home system. I tell you this story because the RoomPerfect was just the same. As soon as I turned on the Focus mode, the (somewhat) boomy bass disappeared, resolution in bass became much higher (allowing individual tones to be distinguished). Soundstage opened up with sound losing its congested character. Everything you think "hifi" is, is applied yet again to your system. You want to sit there and re-listen to your entire library again.

If you have not heard a proper room equalization system, I don't know that you can ever understand the words above. Suffice it to say, nothing, nothing you can do to your system to uplift its performance as much as room equalization can. Why? Details will become obvious in the objective section next but for now, the room gets a hold of the sound out of your speakers and massively modifies the frequency response. Since we are so sensitive to frequency response changes, we hear that as massive coloration. In addition, the changes in bass frequencies translate to elongated (in time domain) notes causing the boominess and lack of resolution.

The effect dies down above a few hundred hertz ("transition frequencies"). Above that the room has subtle changes that are much harder to counter with any kind of equalization. So the transformation in smart systems there is to tonally shape the response. Research shows that we prefer room sound that has more bass than highs. Flat response sounds too bright to us. This type of "target curve" is applied to the equalization as a global touch. For my testing, I used the default target curve. Lyngdorf lets you change that but I did not bother looking for it. Some EQ systems get the target curve wrong and if there is no ability to customize it, you are stuck.

Comparing Focus mode to Global (optimized for wide area), the latter lost a lot of the improvements. It was still a clear step above no EQ but big degradation still. So for any single listening, or even two people side-by-side, I would opt for Focus mode. Indeed shifting my head a foot or two side by side showed more change in Focus mode than in Global (which was almost immune to it).

Let's crystalize these points with actual measurements.

Measurements
I fired up Room Eq Wizard (see my tutorial on REW here: https://www.audiosciencereview.com/...om-measurement-tutorial-for-dummies-part-1.4/). For microphone, I used my USB UMM-6 from minidsp with its calibration file (although I don't necessarily trust the SPL values shown). I placed the mic more or less where the focus measurement point was for RoomPerfect. I then connected to Lyngdorf TDAI-3400 using a USB cable and use it as a "DAC" to drive the speakers with and without RoomPerfect enabled. Here is a comparison of the system without equalization and with RoomPerfect set to "Focus:"

Lyngdorf TDAI-3400 RoomPerfect Focus Measurement.png


We immediately see the disease and its cure. We clearly see the wild swings in low frequencies that I mentioned (in blue). There is a whopping 32 dB variation in frequency response there!!! :eek: As notes go up and down in frequency in your music, their levels go up and down by these swings, causing massive (linear) distortion of what we are meant to hear. The peaks cause the notes to become louder and last longer, obscuring detail. The valleys are caused by cancellations of bass waves and tend to be less audible due to their narrow width. Still, if they can be dealt with, you want to do that.

For now, ignore everything above 200 Hz or so. RoomPerfect has done exactly what we want it to do: it knocked down the peak around 110 Hz. This is "easy" as you are just reducing level. Not so easy is to fill in the dip which it also managed to do. Improper correction here can result in driving the amplifier and speaker too hard and getting very little for it (cancellations get stronger the more you try to fill in the gap). So very nice to see the gap filled.

Notice the 1/12 filtering of the measurements. For bass frequencies you want to see the detail so I have applied this light filtering. For manual correction I may not apply any filtering but I thought for presentation it made things more clear.

Indeed, when frequencies go up, our hearing resolution, i.e. bandwidth, gets worse and worse. For this reason, to get a good idea of what we are "hearing" as opposed to measuring, you want to dial up the filtering. I choose to 1/3 octave although sometimes I use 1/6 if I am focused more on upper bass:
Lyngdorf TDAI-3400 RoomPerfect Focus one third octave Measurement.png


The extra filtering has smoothed the measurements allowing us to see the trend, and target curve applied (approximately). We see that as I mentioned earlier, Lyngdorf is following the classic target curve of more bass than highs. It has wisely decided to not mess much with the response of my speakers above a few hundred hertz.

Still, I usually find that when you take a way large peaks, somethings the result can be a bit bright. I am detecting a tiny bit of this at times so if this were my system, I would dial up the bass a bit more. Since they are now so clean, having more of them is just goodness. Lyngdorf has EQ settings too that could be used for this purpose (I did not try them).

Lyngdorf RoomPerfect has another setting called "Global" which is designed to give better response in a wider area. Using our higher resolution measurement before, I compared the Focus to Global modes:

Lyngdorf TDAI-3400 RoomPerfect Focus and Global Measurement.png


As we see, Global mode is very similar to Focus. The main difference is that the amount corrected is lower. That nicely explains my subjective experience of Global not being nearly as good as focus.

Conclusions
As I had expected and hoped, the Lyngdorf RoomPerfect does a wonderful job of correcting the impact of the room, especially in low frequencies. Without correction, every system out there regardless of price of components, suffers from significant audible colorations and loss of detail and focus. No, you can't do the same with room treatment. Even the best treated room requires equalization. Over treating the room to get rid of the modes will create a dull/dead which you want to avoid.

The all-one-one aspect of the Lyngdorf TDAI-3400 showed its value in how it obsolete nearly $70,000 in gear in my room. Integration with Roon player (i.e. streaming) does it for me. Get yourself a pair of excellent speakers and pair it with the Lyngdorf and "you are done." Your tiny system will outperform tons of other systems without equalization.

Note that similar results could be had with other well implemented equalization system. While performance of RoomPerfect was excellent, it didn't do anything other good systems do. So I will be going to my Dirac system and more powerful amplification. But for someone starting fresh and aiming high, the Lyngdorf TDAI-3400 makes a ton of sense and causes you to forget the issues I found in measurements of its electronics.

Overall I am happy to give a recommendation for Lyngdorf TDAI-3400. Its room equalization is a joy to use and makes a fantastic difference to your listening experience.

------------
As always, questions, comments, recommendations, etc. are welcome.

It is snowing here for the first time this year. I need some money to buy snowshoes. Yes, it is only one inch of snow but you don't want me to fall and get hurt, do you? So please consider donating money using:
Patreon: https://www.patreon.com/audiosciencereview), or
upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
 
Last edited:

Veri

Major Contributor
Patreon Donor
Joined
Feb 6, 2018
Messages
2,064
Likes
1,867
#2
So. How about the Dirac system VS the Lyngdorf correction :):) difficult to compare I'm sure, but I'm wondering if it comes close?
 
Joined
Apr 14, 2018
Messages
31
Likes
28
Location
Washington, DC Area
#4
Thank you, Amir, for this insightful review. I couldn't agree more that room equalization (for frequencies below 200Hz) was by far the single most important improvement in my home system. It made an astonishing distance in my medium-sized room. I felt exactly as you did; I wanted to relisten to my whole collection. Which I have more or less been doing over the last couple of years. My instrument of choice is DSPeaker Anti-mode 2.0 Dual Core. Also very easy to set up. And, a more affordable $1200, about $700 - 800 used in good condition. I recommend it highly. It's cool because you can set up in various locations in the playback chain, so easy to use to improve all your signal sources, digital or analog, makes no difference to the bit-perfect Dspeaker.
 

Veri

Major Contributor
Patreon Donor
Joined
Feb 6, 2018
Messages
2,064
Likes
1,867
#5
It is very hard to AB as I have to switch speakers, amps, etc. This is what I said I wish I had the time and resources to do....
"As much as I like to perform a comprehensive test and review of Lyngdorf against a few of its competitors, I just don't have the time to do that. "

sorry about that, by the end of the review I was too excited too remember that part!
 

Eurasian

Member
Patreon Donor
Joined
Nov 2, 2018
Messages
93
Likes
52
#6
Thanks for this report, Amir. I just calibrated RoomPerfect via the discontinued KRK ERGO this very morning. In my system, I noticed a marked improvement in the bass, but more solid and tactile imaging as well. Unlike my previous drc done in a Venu360 lms, the RP correction is done to each channel individually and for frequencies up to 500 Hz. My subjective take is that the speakers now are much closer in response up to this cutoff frequency, thus improving perceived imaging. Whatever is making it happen, I like it!
 

Juhazi

Senior Member
Joined
Sep 15, 2018
Messages
332
Likes
257
Location
Finland
#12
Does the measurement show two speakers playing unisono, it looks like that? If yes, the spl response shown is pretty typical (mic not in ablsolute center). It the Lyngdorf software does that only, I wouldn't give it a chance...

I have found it more "productive" to measure each driver separately. Optimal response curve is different then, because of no comb filtering (from two speakers). It is very informative to do several measurement by changing mic location. It would be informative to do several measurements by moving the speaker too. This way the user starts to learn what each wiggle in the response stands for, and could do specific adjustments to positioning and response eq. Some changes are easily smoothed by moving the speaker, others by moving the mic. Some changes are better left alone untouched.

We must remember that we perceive sound differently from a single microphone and FFT analysis. EQ softwares give different results because they use different masurements and weigh parameters differently.
 

MZKM

Senior Member
Joined
Dec 1, 2018
Messages
319
Likes
231
Location
Land O’ Lakes, Florida
#13
Does the measurement show two speakers playing unisono, it looks like that? If yes, the spl response shown is pretty typical (mic not in ablsolute center). It the Lyngdorf software does that only, I wouldn't give it a chance...

I have found it more "productive" to measure each driver separately. Optimal response curve is different then, because of no comb filtering (from two speakers). It is very informative to do several measurement by changing mic location. It would be informative to do several measurements by moving the speaker too. This way the user starts to learn what each wiggle in the response stands for, and could do specific adjustments to positioning and response eq. Some changes are easily smoothed by moving the speaker, others by moving the mic. Some changes are better left alone untouched.

We must remember that we perceive sound differently from a single microphone and FFT analysis. EQ softwares give different results because they use different masurements and weigh parameters differently.
Very true. For speaker placement, using an RTA would be a huge time savor, but it won’t show decay, so a measurement would be needed to finalize. I usually do a 3x3x3 cubic grid for mic placements, maybe with a 1ft radius.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
20,051
Likes
23,062
Location
Seattle Area
#14
Does the measurement show two speakers playing unisono, it looks like that?
My measurements do. Its own measurements use one channel at a time and so do I in manual optimization.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
20,051
Likes
23,062
Location
Seattle Area
#15
We must remember that we perceive sound differently from a single microphone and FFT analysis. EQ softwares give different results because they use different masurements and weigh parameters differently.
That's very much true as wavelength of sound approaches the distance between our ears. In bass frequencies, the wavelength is much, much larger so singe mic measurements are reflective of what we hear.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
20,051
Likes
23,062
Location
Seattle Area
#17
Amirm , off topic - just curious, have you ever measured your Mark Levinson amps ?
I have not. Stereophile has test the No 53 (they found issues with it). I don't think anyone has measured the 523.
 

rajapruk

Senior Member
Patreon Donor
Joined
Sep 22, 2018
Messages
360
Likes
344
Location
Gothenburg, Sweden
#18
Very nice looking home system Amir. We want to see all that Levinson-stuff measured, you know that! But, maybe you do not want to take the risk of killing the resale values? ;)

A comparison of the impulse response / energy time curve could be interesting to see as well. If something is changed in the time domain by the processor. I think for example Dirac does that.
And also a comparison of waterfall-diagrams (decay of the sound in the room) could be interesting to see, what difference is achieved.

It's hard to say when you do not see the full room, but maybe better sound could be achieved if your speakers were placed even farther away from boundaries (walls). If possible, I would try that to get reflections later in time and lower in magnitude. For example try to place the speakers where the mattress ends on the floor, i.e a bit more out in the room.
Just ignore this above if you are not interested in comments of your own system.

Also, I do not agree about your statements about room treatment. But that is a whole other story :)
 
Last edited:

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
20,051
Likes
23,062
Location
Seattle Area
#20
It's hard to say when you do not see the full room, but maybe better sound could be achieved if your speakers were placed even farther away from boundaries (walls).
Not possible. This is a loft and there is only 2-3 feet to the right of the laptop.

I have a dedicate room where this was supposed to go. But it in the basement with no windows. So I decided to convert the loft to listening room for music (basement is a theater).
 
Top Bottom