Motu M2 Review (Audio Interface)

[PeteL]

What’s the output impedance of headphone out?

The lack of extra power for lower ohm load is weird.

You will have difficulty finding USB only powered bidirectional Interface with phantom power and mic preamps that does well in term of available current for headphones. Do you know any that are better in that regard?

 

[infinitesymphony]

Wish someone would create a model with a stronger headphone amp, which is the only thing holding these devices back. Even some devices that require a wall plug only seem to achieve 20 mW.

Is it possible to control the RCA and TRS outputs separately in software, or are both always controlled by the Monitor knob? Guessing you would need something with dedicated line outputs like the M4 to bus a separate full-output send to a headphone amp.

I have been using the 2i4 2nd Gen daily for almost two years. I like that unit very much as a DAC, for vinyl rip through the ADC, headphones occasionnal listening... But have had lots of drivers issues and need a lot of buffer to play seemlessly.

I was aksing to myself: would I get another one if mine was dead, or get a MOTU?

What kind of computer do you have? Some models have known DPC latency issues with USB audio interfaces, like Dell XPS and T2-based Macs.

DPC Latency Rankings (anecdotal, but a good starting point):
https://www.notebookcheck.net/DPC-L...ablets-offer-the-lowest-latency.504376.0.html

List of T2-based Macs:
https://support.apple.com/en-us/HT208862

 

[JohnYang1997]

Not sure every designer is aware of it. Acutally I think most could not care lesse. ASR seems the only place this is noted.

M2 with a more sensible filter than M4, odd..

Surprised to see the ESS hump still around these days.

Good overall package for sure though.

There are two ways of IMD hump being shown. One is actual IMD distortion component suddenly rises at around that level. That's 9038q2m 9038pro based dacs.
With these like motu m2, there's a different mechanism which is noise floor modulation where the noise floor suddenly increases at certain level. There's simply no cure for this type. It is very pronounced in 901x generation. And this is also the cause of remaining imd hump for 9038q2m when the design around the chip has very very low noise and no imd component is left.

 

[phoenixsong]

Wish someone would create a model with a stronger headphone amp, which is the only thing holding these devices back. Even some devices that require a wall plug only seem to achieve 20 mW.

Is it possible to control the RCA and TRS outputs separately in software, or are both always controlled by the Monitor knob? Guessing you would need something with dedicated line outputs like the M4 to bus a separate full-output send to a headphone amp.


What kind of computer do you have? Some models have known DPC latency issues with USB audio interfaces, like Dell XPS and T2-based Macs.

DPC Latency Rankings (anecdotal, but a good starting point):
https://www.notebookcheck.net/DPC-L...ablets-offer-the-lowest-latency.504376.0.html

List of T2-based Macs:
https://support.apple.com/en-us/HT208862

Is this model's headamp section limited by the fact that it's fed by usb? Or do usb c connections not have such a concern? Btw, a simple solution would be to feed the TRS/RCA outputs into a headphone amplifier. Only thing I'd complain about is the need for another electrical socket/outlet, ad the extra cable mess, cost and space requirements that come with it

 

[bigjacko]

Is the filter just one built-in to the ESS DAC or have Motu added their own?

Sorry I don't know much about how they implement filters, but you can ask some other people on the forums, most of them know quite a lot onthis stuff.

 

[maverickronin]

Is this model's headamp section limited by the fact that it's fed by usb?

It's limited by the fact that it's powered by USB and combines so many functions. If it was just a two channel DAC and headamp is could spare a lot more power for the head amp.

Or do usb c connections not have such a concern?

It's just the shape of the connector. The protocol is 2.0 high speed. The connector supports USB 3 and 3.1 but AFIK no audio interfaces actually use those. (There are thunderbolt interfaces though.)

USB C also supports "power delivery" which can in theory supply up to 100 watts but I don't know if any audio interfaces use that either. The problem with power delivery is that compatibility is all over the place. Essentially every USB C source which supports PD has it's own individual voltage levels and current limits which may not match what the interface needs and make it a nightmare for a manufacturer to support.

 

[PeteL]

It's limited by the fact that it's powered by USB and combines so many functions. If it was just a two channel DAC and headamp is could spare a lot more power for the head amp.



It's just the shape of the connector. The protocol is 2.0 high speed. The connector supports USB 3 and 3.1 but AFIK no audio interfaces actually use those. (There are thunderbolt interfaces though.)

USB C also supports "power delivery" which can in theory supply up to 100 watts but I don't know if any audio interfaces use that either. The problem with power delivery is that compatibility is all over the place. Essentially every USB C source which supports PD has it's own individual voltage levels and current limits which may not match what the interface needs and make it a nightmare for a manufacturer to support.

I am trying to find if usb audio class 2 specify a max power on the bus, for some reason I can quickly find the spec, but I certainly wouldn't feed 100 W next to a non error corrected stream of Data, there is no way it could comply, this spec should be somewhere tough. In any case, power delivery thing don't apply, it's if you don't do anything else than supply power, but even then, doesn't apply to charging your phone neither, I think a computer can supply like half an amp to a USB device, but I could be wrong.

 

[Pearljam5000]

The Benchmark DAC 3 has 4 D/A chips ( 2 per channel i assume)
Theoretically if the M2 had more chips, it would have sounded better?

 

[maverickronin]

I am trying to find if usb audio class 2 specify a max power on the bus, for some reason I can quickly find the spec, but I certainly wouldn't feed 100 W next to a non error corrected stream of Data, there is no way it could comply, this spec should be somewhere tough. In any case, power delivery thing don't apply, it's if you don't do anything else than supply power, but even then, doesn't apply to charging your phone neither, I think a computer can supply like half an amp to a USB device, but I could be wrong.

UAC2 is more or less on top of the base USB spec. 5V, max 500ma for 1 and 2, max 900ma for 3.

Power delivery allows the source and sink to negotiate voltage and current requirements. IIRC the spec allows for anything between 5 and 20 volts at up to 5 amps. The interface chips on each end are powered by the base 5V so they can talk to each other.

The sink specifies a voltage range it can accept and the maximum power it will draw. If the source can supply something within that range it will go ahead and do it, and the sink device will power up. If not then it just doesn't work, but doesn't break anything either.

The problem is that the source device may not have enough power or support a specific voltage. The sink can request any arbitrary voltage between 5 and 20, but that doesn't mean the source and regulate at that specific voltage. It's something I ran into when I was working on this little project. My USB3C PD powerback would support more voltages than an AC charger I had. The charger only supports 5, 9, 15, and 20 volts output while my powerbank also supports 12V out, which is a much more common voltage for desktop equipment with external power supplies.

A designer could always increase compatibility by accepting any voltage above the minimum required and regulating it down, but that will increase cost by requiring over-speced components and more robust thermal design.

Even though the ADI-2 DAC's manual says you can run it at anything between 9.5 and 15VDC I would prefer to run it at the standard 12 (per its included supply) just to keep the heat down.

 

[DDF]

There are two ways of IMD hump being shown. One is actual IMD distortion component suddenly rises at around that level. That's 9038q2m 9038pro based dacs.
With these like motu m2, there's a different mechanism which is noise floor modulation where the noise floor suddenly increases at certain level. There's simply no cure for this type. It is very pronounced in 901x generation. And this is also the cause of remaining imd hump for 9038q2m when the design around the chip has very very low noise and no imd component is left.

John Siau from Benchmark seems to disagree and states that the imd is common mode and removable if a differential amp is added at the dac chip o/p:
https://www.audiosciencereview.com/...nts-of-benchmark-dac3.3545/page-16#post-90519

 

[Severian]

I have the M4, which has a few more inputs/outputs and some extra monitoring features. I vastly prefer it to the multi box situation I had before, for controlling headphones and monitors at a computer. I have the monitor outs connected to powered monitors and keep my headphones plugged in. I control which one I listen to by using the separate volume controls.

This functionality, combined with the immense convenience of the additional inputs/outputs/beautiful meters/super low latency has sworn me off so called audiophile dacs/headphone amps. If you want to get audio into or out of a computer this is probably the best way to do it for most people.

Also I like that they design these things in my backyard in Harvard Square.

Totally agree. I love my M4; it's the heart of my stupidly complicated PC setup. Although I do connect a headphone amplifier to the 3/4 line outputs. The M4 headphone output is fine for easy-to-drive headphones but some of mine require more power.

 

[Music1969]

I have been using the 2i4 2nd Gen daily for almost two years. I like that unit very much as a DAC, for vinyl rip through the ADC, headphones occasionnal listening... But have had lots of drivers issues and need a lot of buffer to play seemlessly.

I was aksing to myself: would I get another one if mine was dead, or get a MOTU?

Noted. I wonder Gen3 could be difference between our experiences though?

 

[infinitesymphony]

Btw, a simple solution would be to feed the TRS/RCA outputs into a headphone amplifier.

Sure, but then you can't easily switch between speakers and headphones without changing the volume, unless you can set separate volumes for those outputs in software. I assume that it is possible with the M4's line outputs.

 

[Sir Sanders Zingmore]

You will have difficulty finding USB only powered bidirectional Interface with phantom power and mic preamps that does well in term of available current for headphones. Do you know any that are better in that regard?

What about the RME BabyFace Pro FS?

 

[JohnYang1997]

John Siau from Benchmark seems to disagree and states that the imd is common mode and removable if a differential amp is added at the dac chip o/p:
https://www.audiosciencereview.com/...nts-of-benchmark-dac3.3545/page-16#post-90519

What he said is almost completely false from what I actually measured. I was influenced by this comment for a long time.
A few things:
From my current point of view, the noise of DAC3 is way too high to reveal the imd hump of 9038pro/9028pro. It will definitely be there if you look FFT.

As demonstrated by Ben from Soncoz, IMD hump can be fixed by changing values of feedback RC of the I/V circuit. It appears on both rca and xlr output so in every way John from Benchmark is wrong.

The IMD hump is also caused by poorly designed reference voltage amplifier/regulator. It may be something to do with output stage of the dac with the output impedance of the Vref and driving the capacitance in the I/V stage.

It's also affected by the oscillator. Although I didn't find the pattern and what to look for, different oscillators cause different distortion change vs voltage. A phenomenon that's not fully discovered is when you attenuate 0.5db the distortion components change a lot. This is a much deeper effect than we thought.

Another thing is that when I measure one half of the output of my 9038q2m board, I found only 2nd harmonics increased 5db, anything else didn't change. This is in so many way contradicting with what people thought they knew. Or it's simply they didn't design it properly.

 

[Tks]

There are two ways of IMD hump being shown. One is actual IMD distortion component suddenly rises at around that level. That's 9038q2m 9038pro based dacs.
With these like motu m2, there's a different mechanism which is noise floor modulation where the noise floor suddenly increases at certain level. There's simply no cure for this type. It is very pronounced in 901x generation. And this is also the cause of remaining imd hump for 9038q2m when the design around the chip has very very low noise and no imd component is left.

I sure hope you know this because you're interested in coming up with that DAC/AMP I've been bugging you always about ;P

And also with AKM up in flames, hope that's why you've been looking at ESS stuff and know this :]

 

[trl]

Surprised to see the ESS hump still around these days.

@Χ Ξ Σ, probably someone tried to quoted or replied to you, then manually changed the username, like I did below. I'm sure it was nothing intentionally done.

Weird, that is not my words. Not sure how did another member's words mix up with my username.

 

[AudioSceptic]

Sorry I don't know much about how they implement filters, but you can ask some other people on the forums, most of them know quite a lot onthis stuff.

Sorry, I didn't intend to make it seem that I was asking you personally. I'm hoping that someone will chip in.

 

[PeteL]

What about the RME BabyFace Pro FS?

It does look to be a little better, Amir measured 56 mW in 33 ohms, not a huge difference but Yes you get a few more dBs, looks like they've been able to optimize their design a bit more. It still not great for a desktop headphone amp tough, if it's the main thing that makes you reject the Motu, I wouldn't spend 5 time more just for this extra bit of power. It is however better all around.
Edit: I see that it does have an optional dc power supply, but the review suggest it's been measured bus powered, it's likely the same but hard to know for sure. RME specs it very slightly more, but it's not very conclusive, they don't specify the load and they specify the max power at 0.01 % THD, where Amir sets the max power at linear behaviour, best THD+N

 

[Draki]

I have Focusrite Scarlett 18i20 Gen3

All Gen3 use same driver and for me it has always been rock solid reliable

I need it to be, doing 6 channel DSP crossover and room EQ . Using Win10 ASIO driver

Could you please tell what crossover and EQ? Is it a PC-only as a source? How do you control the volume? Thanks.