Lyngdorf TDAI 3400

[flashman]

Glad you are enjoying your 3400. I had one once, replacing a Devialet Expert 220 Pro. I found the 3400 thin in the bass department, however, and eventually sold it. My digital front-end is the same as your previous system: Lumin X1 and Lumin L2. My amplifier is a Kinki EX-M7. I love it. That’s why our hobby is so intriguing: you’ve found the 3400 better suited to your preferences and I feel the same about my system. Happy Holidays!

 

[tmallin]

The 3400 "thin in the bass department"? You did not specify your speakers. so I don't know if perhaps this is a speaker thinness issue. But if anyone finds the Lyngdorf sounding thin in the bass, you can use the Voicings or the bass/treble controls to correct this. Unlike most modern preamps or amps (including your Kinki) the Lyngdorf has both traditional tone controls and DSP room/speaker correction via its parametrically constructed stock or custom user Voicings.

And unlike most amps (including your Kinki), the Lyngdorf doubles down its power from 8 ohm to 4 ohm loads, indicating it has plenty of power supply oomph for bass. The Lyngdorf also has 40 amps of maximum output current, far more than the 18 amps specified for the Kinki, again indicating its ability to drive speaker loads fabulously throughout the entire frequency spectrum, especially the current-demanding bass regions.

 

[flashman]

My speakers are Vivid B1 Decade. I’m not debating your experience with the Lyngdorf, only my experience with the unit. I’ve followed your previous posts about the Lumin X1, so I know that you are quite exhaustive (not a pejorative) in your analysis. At the end of the day, system dependencies vary greatly so I’m not surprised at your or my evolution. Enjoy!

 

[RC22]

So great to have found this discussion. Like you, I also had a 2170 before getting the 3400 in mid 2022. I had the 2170 for five years before that.

I absolutely LOVE the 3400. I’ve had the sense that its more powerful digital engine does a better job of room correction, but I had nothing specific to point to as you have. A Lyngdorf deaeler I know thinks it does as well.

I’d like to ensure I understand how you run RP. I get that you keep the mic pointing between the speakers after doing the focus position instead of aiming it randomly around the room as Lyngdorf recommended forever, but are also saying you leave the mic right there at the focus position for subsequent room readings and just keep reading the room from the same spot as the focus? I’m happy to try it, but it’s counterintuitive to say the least. Heck, I never get 100% knowledge in less than 6 readings not counting focus. And I almost always get no add’l knowledge added even after moving the mic at least one time in the process. So, I ask rhetorically—unless you can actually answer, how does it add knowledge by reading the room from the same spot over and over? And in only three readings, post focus? Unless they have changed the software to do so recently. I just did the latest to 3.4. Lastly, since I will likely be doing this soon, why do you run add’l readings after you’ve hit 100%?

In fact, I always found it weird that the first reading always gives 55% knowledge and the second one always goes to 69%, but subsequent ones vary. No matter where the mic is plasced for one and two and in two different rooms I’ve used.

Do make sure your voltage switch in the back under the decal is set to 110 not 220. Mine was set to 220 but it played so I was none the wiser. But the PS does not operate optimally. DO NOT change the setting with it powered on. I would unplug it, in fact.

I also am utterly spoiled at the precision of the volume conrol. .1db increments? I’m in heaven.

FWIW, the LF performance is outstanding in my rig. Streaming is my only source. Qobuz and an Innuos Pulsar.

BTW, I had the 3400 modified significantly last July. Having modded the 2170 before it, I knew Lyngdorf TDAI’s could be made to sound even better for reasonable money. The modder discovered the voltage switch set to 220v.

 

[tmallin]

Yes, I didn't touch the mike at all once I set it up at the listening position. I just ran successive iterations of the test signal without moving anything. I don't remember the confidence level shown after the first or second sequence, but certainly it showed 100% after a total of three and I ran one more for good measure. I "tricked" Room Perfect since each time it ran the tone sequence the microphone obviously heard exactly the same result. As I said, even my body was absent from the room during the calibration so even different body positions were eliminated from what the microphone heard.

I agree that owners should check that the voltage is set correctly. My new unit was set correctly when I received it. The second unit I purchased was a used one and it was set incorrectly for 220 even though I strongly suspect the unit was last used in the United States--at least that is where the package originated from and where the seller was employed and had a residence.

My theory about Room Perfect is that if you always listen from the same spot and you are not setting up for multiple listeners/listening positions, all that matters is the sound at the sweet spot. The sound at the sweet spot is therefore what you want RP to optimize, not sound at other positions in the room. This is especially true for a room like mine where the walls are treated with absorbing acoustical foam in areas around where first reflections would be heard from the sole listening position and aiming the microphone in random ways will inevitable have the microphone facing either foam-covered walls or non-foam-covered areas, fooling the microphone about the room's mid and high frequency response.

Previously, with my TDAI-2170, I never had problems with RP's low frequency correction. It always seemed to introduce a peak in response centered around 4 kHz, which sounded considerably too bright. Now, with this "trick" RP calibration, RP's correction sounds wonderfully smooth.

Of course, as they say, YMMV--your mileage may vary. What works in my room to my ears may not work for you. I detailed my "trick" RP set up on the chance that others also may find that this sort of calibration scheme works better for them.

 

[tmallin]

In my opinion, there is no reason to buy a separate streamer if you own the Lyngdorf TDAI-3400. It's streaming is of equivalent quality to even the fiber-optic connected top Lumin X1/L2 set up feeding a pair of Benchmark AHB2 monoblock amps I had before in my streaming-only system. When I dropped the 3400 into the system replacing those components and adding the fiber optic GigaFOILv4 Inline ethernet filter, I heard equivalent streaming quality.

For Qobuz, while we await the still-coming Qobuz Connect, the MConnect app for $6 provides very fine sounding results from the 3400, much better than running the Qobuz app via AirPlay and marginally better than the sound of Qobuz through Roon.

Roon has made sonic improvements across the board lately. If you not using its Muse DSP filters, be sure to remove all the filters you can, don't just deactivate them. This will gain you at least further subtle sonic improvements.

 

[RC22]

Thanks for the nice replay, Tom. I am glad you shared the new approach to mic placement. I’ll let you know how it goes.

I also purchased my unit from a guy locally purchased it new. Box label shows it shipped from Denmark to US. What a coincidence we had the same experience with that.

I always wondered about the roll of my body in the room. I will try using the remote from outside my door. Otherwise, I used to either sit on the floor behind my seat or crunch down in my seat below the focus mic and them sit in my seat for the rest. I figured my body is in the seat while playing music, so….

I used to have a Muon internet filter by Network Acoustics and ended up determining it was not helpful after making some other changes in the rig. I’’ll look into the GigaFoil you use, though.

 

[RC22]

Also, I’ve wanted to hear those Watkins speakers.

 

[tmallin]

Additional Lyndorf Voicings

After you set your system's "foundation EQ" by running Room Perfect, you can use the Voicings function of the Lyngdorf TDAI-3400 to manipulate the frequency response of individual programs to make them more realistic or agreeable to your ears and sensibilities. The stock Voicings which come already available with the Lyngdorf software are indeed helpful in this endeavor. For a text and graphical description of these voicings, download and review the file available at this link.

Here are a few other Voicings you may find useful and instructions for how to construct them using the Setup>Voicing setup>Add new voicing function from the menu.

You may want to create a new Voicing patterned after what DSPeaker recommended as a presence range dip. In the instruction manual for its Dual Core equalizer, DSPeaker recommended a parametric filter of Q = 2 centered at 3kHz with a dip of 4.2 dB.

The Lyngdorf Music Voicing is this except the dip is only 3 dB. You can create a new Voicing, call it Music 2 or whatever. (The other stock Music voicing has a different dip: Q of 0.8 and center of 2900 Hz, amplitude minus 2.5 dB.) I think you'll find this new 4.2 dB dip quite useful.

Also, the stock Soft Voicing is very subtle. It is patterned after the Quad preamp's "Tilt" control (also found on the Audient ASP231 analog graphic equalizer I once owned) and hinges response around a midrange frequency, boosting the bass and lowering the treble a bit overall for a gentle smooth slope down from bottom bass to top treble.

The Soft Voicing uses the first three filters, each set for high shelf, at Q = 0.1, amplitude at minus 0.5. The first filter is set to 20 Hz, the second to 200 Hz, and the third to 2000 Hz.

To make the tilt greater, I have created Soft 2, 3, 4, and 5. Each one follows the Soft Voicing pattern but increases the amplitude by another minus 0.5 dB so that Soft 5 gets to each filter set to minus 3.0 dB.

To keep the overall system SPL about the same, you can add some overall gain to the higher order Soft Voicings. For example, I added +3 dB overall gain for Soft 3 and +4 dB overall gain for Soft 5.

You might find these additional Soft Voicings useful for overly bright recordings.

 

[RC22]

Tom,

I just ran Room Perfect by leaving the mic in the focus position for all readings as you suggested and it worked out great. Much easier and faster than moving the mic around 5-8 times and maybe only getting to 96%. Gotta tell you though, I did the focus and then got to100% knowledge on the first reading. I did two more readings anyway as you suggested. I stood outside my room and controlled everything on the Lyngdorf app on my phone. I use the app to set up my subwoofer, but had not used it for RP before. I liked it.

The app is fantastic. Check it out if you’ve not yet.

I cannot say if there is a SQ difference between the traditional method and this method of running RP because I changed so much in my room this week that the legacy RP was not relevant anymore. I’d been evaluating changes to my set-up changes on bypass.

Rick

 

[RC22]

Tom, I wonder if you know Lyngdorf wants us to interpret the room correction percentage figure? When you press the “status” icon on the unit’s faceplate display, a room correction % is given along with the number of readings you took and the % of room knowledge.

I have guesses but that’s all they are. I will say that over 7 years of running RP on the 2170 and the 3400 I can recall getting as low as 12% and as high as mid 32% or so.

Thank you,

Rick

 

[tmallin]

Rick, I'm not sure what you are getting at in post #31. As far as I know, there is no "status" icon on the faceplate display. If there is, it is not mentioned in the manual and does not appear on my display, at least that I have noticed. In any event, the percentage of room knowledge achieved in any given Room Perfect set-up is irrelevant once that set-up is completed and being used since there is no obvious way to alter the frequency response corrections entailed in any given Room Perfect set-up other than by adding Voicings.

Yes, the Lyngdorf App is very fine and I use it frequently. If you want to make fine adjustments to volume, this is the only way to always have 1/10 dB adjustability at any volume level. Roon gives you this level of adjustment over part of the volume control range, but not at lower volume levels. The volume levels available directly from apps like Tidal, Mconnect, and AirPlay used with the Lyngdorf are often coarser, touchy, or imprecise, sometimes only giving choices in three to five dB increments.

 

[tmallin]

With my Watkins Generation 4 speaker set-up, I'm increasingly finding the modified Music Voicing I talked about in post #29 (a parametric filter of Q = 2 centered at 3kHz with a dip of 4.2 dB) very pleasant on a lot of music, especially large-scale classical. Modifying the stock Bass 1 and Bass 2 Voicings to incorporate this parametric presence-range dip also works very well to my ears in this system.

Also, you may want to try tinkering with the frequency of the high shelf filter used in the stock Lyngdorf voicings. I like warm, full bass. The stock Lyngdorf Bass 1, Bass 2, Action 1, and Action 2, and Action Movie voicings all have high shelf filters which are specified at no higher than 120 Hz. At least in my Watkins-based system, with these bass-accenting Voicings it is often helpful to move this shelf frequency a bit higher--about 160 Hz--to allow more of the upper bass range to be accented. This is especially useful for low and moderate volume listening.

With speakers which inherently have a considerably warmer bass response than the Watkins, such as the AR-303a I use in my other audio room, adding more bass via these Voicings is usually not necessary or even desirable to my ears.

 

[tmallin]

Here is an operational nicety of the Lyngdorf TDAI-3400 not well covered in the manual or any of the reviews I've read. The unit offers three levels of "off." Flipping the power switch on the back panel near the power port is truly "off." Turning the unit on again requires mechanically flipping that power switch back on. It will not turn on with either the front panel button in the lower right corner, the power button on the physical remote wand, or from the power button on the Lyngdorf app.

There are two standby modes selectable from the Power Management section of General Settings. One is "Network Standby." This mode offers the easiest and quickest mode of turning the unit on from standby. In this mode, the 3400 is still discoverable and its settings may be adjusted from the app on a computer or mobile device. With networks like mine, where I have two Lyngdorf TDAI-3400 units connected to the same network, the app will give you a choice as to which unit to control. But if I'm not careful about choosing which unit I want to control, at least some setting changes will be applied to both units at the same time.

To prevent any confusion about which unit is being controlled at any time, I choose to activate the "Deep Sleep" standby mode for both my Lyngdorf units. In this mode, when the unit is in standby, it will not appear in the Lyngdorf app as an option and any settings changes will not be applied to the Deep Sleeping unit(s). I have also noticed that the usually blinking light of the ethernet connection to any Deep Sleeping units no longer blinks. On my TP Link network extender device, for the Deep Sleeping unit's connection, the light indicating that the ethernet connection is active no longer is lit and the ethernet connection at the back of the TP Link also does not blink.

The only oddity about the Deep Sleep mode i've found is that to turn on the unit from standby usually takes two or three presses of the power button of the wand remote control or the button on the front panel. The unit cannot be turned on with the virtual button of the app since the app won't communicate with the 3400 in this Deep Sleep standby mode. If the 3400 is in Network Standby mode, the unit always turns on with a single press of either physical power button or from a single tap of the virtual power button of the app.

 

[tmallin]

There have been online comments suggesting that the Lyngdorf TDAI-3400 does not benefit from, and may actually react adversely to power filtration. And in fact, in my Blue Room system using the 3400 with my new Acoustic Research AR-303a speakers, the the system sounds remarkably fine with both music and video streaming sources without either power filtration or ethernet filtration. I think a major part of my subjective reaction to this system is that the AR-303a's "big hearted" low end and gradual roll off of highs above 5 kHz covers a multitude of sins in program material, most of which is balanced at least a bit or more on the too-bright side.

But I have now experimented in my more highly resolving Watkins Generation 4 based system and in fact the 3400's sonics are considerably better when using both power line filtration and ethernet filtration.As with prior electronics and speakers in this acoustically well-treated room, I usually use both ethernet filtration (GigaFOILv4 Inline ethernet filter, powered by 5V output of Farad Super3 with optional Level 2 Classic DC cable) and power line filtration equipment (P.I. Audio Group UberBUSSes, P.I. Audio Group BUSS Depot)

It is easy to experiment with the audibility of this filtration in my Watkins-based system. To bypass the ethernet filter, I merely connect the ethernet cable coming from the router directly to the Lyngdorf rather than to the input of the GigaFOIL. To bypass the power line filtration I just move the power plugs of the Lyngdorf and Farad directly to the wall outlets rather than powering them from the outlets of the P.I. Audio UberBUSSes.

My experiments revealed that, as with other electronics and speakers in this room, both power line filtration and ethernet filtration each make substantial sonic improvements. Used together these two "cleaners" make a truly remarkable improvement in the natural detail (more detail from instruments and voices without added brightness), tonality (much less spurious brightness), and spatial presentation (both much more three dimensionally open and focused). It really isn't close at all and you'd have to be near deaf not to almost instantly hear these changes as a truly substantial improvement.

 

[tmallin]

In posts #29 and #33 above I talk about additional Lyngdorf Voicings I have constructed which I find helpful for more pleasingly natural reproduction of material that is either inherently bass weak, treble heavy or both (much commercial material). Keep in mind that this is for speakers which, like my Watkins Generation 4, are either fairly neutral sounding or a bit on the thin/bright side themselves. In my AR-303a speakers, I rarely need such Voicings since the speakers are inherently balanced with a warm, rich "big hearted" bass and a gradual treble rolloff.

To get a balance which smoothly rolls off by a total of about 10 dB from 20 Hz to 20 kHz--which is more or less the native balance I found so natural with the Dutch & Dutch 8c speakers--I have again followed the pattern of the Lyngdorf "Soft" Voicing but tilted the response much more than in the stock Lyngdorf Soft Voicing which only tilts the response down by 1 dB or less over the full spectrum. The get a 10 dB downsloping response from 20 Hz 20 kHz, here are the parameters to use:

Filter 1: High Shelf 20 Hz, Q = 0.1, Amplitude = -5.2 dB
Filter 2: High Shelf 200 Hz, Q = 0.1, Amplitude = -5.2 dB
Filter 3: High Shelf 2,000 Hz, Q = 0.1, Amplitude = -5.2 dB
Overall System Gain = +10 dB

I have also found that a broader presence range dip can provide much of the relaxed magic of the above Soft Voicing, without the apparent bass emphasis and while allowing a bit more air/sparkle in the top two octaves:

Filter 1: Parametric centered at 3,000 Hz, Q = 0.7, Amplitude = -4.2 dB

I think lovers of classical music who use speakers not already having an inherently rich bass response and down-sloping treble may find these Voicings very pleasantly natural sounding indeed, especially when listening at moderate or higher SPLs.

 

[tmallin]

It was pointed out to me by Robert E. Greene that there is no real reason at all that the roll off of higher frequencies should be a steady 1 dB per octave
and indeed the 10 dB total is is not written in stone either. Moreover, it is not really the case that the roll off in a concert hall is steady. Typically the bass is up,
then the response of the hall is fairly flat from around 300 Hz up to around 2 kHz and then there is a steeper roll off from there on up with the steady state response almost gone above 8 kHz.

There are other Lyngdorf Voicings which take the approach described by REG. One is called Action Movie. It boosts the bass, is flat up through the mids and lower highs, and then rolls off. It, too, is helpful with much material, I think. The shapes of the various Lyngdorf Voicings are shown in the document linked to at:

https://lyngdorf.steinwaylyngdorf.com/downloads/tdai-voicings/

I don't believe Lyngdorf has published the parametric filter specs of these curves. However, if you own a Lyngdorf unit, you can, via the Edit Voicing function, inspect the filters used by a given Voicing and tinker with then to your heart's content if you don't believe any of the stock Lyngdorf Voicings do exactly what you want in terms of electronic equalization.

 

[tmallin]

Robert E. Greene also pointed out to me that exactly what happens in the midrange really illustrates the importance of the 0.1 dB threshold of audibility in that range. A shift in the midrange of a small and even quite narrow band nature
can have a dramatic effect on what things sound like , The problem is that it gets
really hard to fix such errors without quite exact parametric equalization, even for someone with great experience with such EQ and sensitivity to the tonal problems involved.

I don't doubt what REG says about tonal balance problems in the midrange. However, there really is no practical way to apply EQ so precisely on a recording-by-recording basis. It would take too long and be too tedious. About all you can do is attempt to fix speaker and room problems once with that level of precision to then form a foundation EQ upon which you can then overlay recording-by-recording EQ corrections. Perhaps that is what REG meant.

This is what Lyngdorf's Room Perfect and Voicings attempt to do more or less automatically, but of course the precision necessary for the type of narrow band and small corrections REG is talking about is not there--at least not without careful construction of a Voicing using numerous filters. Such precision requires manual adjustment of the tedious variety.

I think it is a mistake to criticize good and practical EQ solutions in favor of impractical near-perfection solutions. The perfect is the enemy of the good and given the imprecision of the recording and playback processes, I think it's a fool's errand to pursue what you hear as near perfection in terms of electronic EQ. Most people, if told that's the way to do it, will simply throw up their hands in despair and abandon electronic EQ in frustration, listening to much less balanced playback than could be achieved simply and easily with little frustration.

If you want to apply electronic EQ on a recording-by-recording basis, you need a device which can be adjusted in a minute or two, at most. Devices with more-or-less graphic EQ bands such as the Schiit Loki series or the old Cello Palette Preamp come to mind, but I'm sure there are others.

Or you need a parametric EQ device which can hold numerous presets which the device electronically remembers and which quickly can be applied on the fly as you listen. For disc playback you can label the disc or its container with the number of the preset you have found to yield the best-sounding result.

For example, the Lyngdorf TDAI-3400 can remember 32 different Voicings. These Voicings can be viewed and selected very quickly with its physical wand remote control by depressing the Left and Right buttons on either side of the Select button. Pressing the Left and Right buttons displays the frequency response graph and name of the chosen Voicing on the 3400's built-in screen. This information on the screen is plenty large and bright enough to be clearly viewed from 10+ feet away in any room lighting. You can quickly apply the displayed Voicing by touching the Select button. In this way you can quickly compare the effect of various Voicing presets on what you are hearing from your listening seat.

On the Lyngdorf, each stock Voicing easily can be deleted or edited if desired. I think most users can delete all the stock Voicings which decrease the bass in any way; they seem useless for music playback. This frees up additional slots for useful Voicings.

New Voicings can be constructed and added to the stock Voicings. Each voicing can use up to eight parametric filters, but none of the stock Voicings use more than three filters.

It is very easy to tinker with the Voicings. Just click on Edit on your mobile device controller (e.g., iPad) and you see a graph showing both the overall effect of all the filters used by that Voicing in one color and the effect of any chosen individual parametric filter in another color. Then you just type to change the parametric values of each chosen filter and instantly see the effect on the graph and hear the effect in real time. Once you click Save, that new Voicing is available for selection in the future.

 

[RC22]

Tom,

A late follow up to your response to my #31 post. There is a data point called Satus, expressed as Roomcorrection : _ _ %. Lyngdorf displays it like I wrote it—no spaces between words. If your unit’s menu does not have a Status , then Lyngdorf decided to eliminate it form the product, I suppose.

On your remote: Click Menu. Click right arrow on the circle in the middle of the remote till you see “set up”. Click the “select” button at the center of the circle on the remote. Then click the right arrow (on mine it’s two clicks) till you see “RoomPerfect Setup” and then choose it using the route’s select button. Finally, click the right arrow till you see the “Status”, then choose it.

See the following three data points.

# of room measures
Room Knowledge (%)
Roomcorrection (%)

Since Lyngdorf chose to report to us the percentage of room correction, I’d like to know what they wished to convey to us in doing so.

One guess is that the higher the number the worse your room was to begin with. Acoustically, I mean. I am probably off here, but if one takes it to a logical extreme, a perfect room would need 0% correction. In realistic terms, a quite good room should require less correction than a relatively poor one.

I’ve had readings as low as 8% room correction and as high a 32%. In the last week after making some changes to my room, I’ve gone from 32 to 25 to 22. But seeing in my records that I had 8% in a previous room makes me wonder if—according my Lyngdorf’s SW, my old set up was a lot better.

 

[orange55]

@tmallin Great thread. I have had a Lyngdorf since 2018 and never looked back. Like you say the steamer is a hidden gem.

One thing I would recommend is trying the iOS app JPLAY. It sound much better than mconnect or the native Lyngdorf app. It has a 14 day free trial. Well worth trying out.