M-Track Solo budget audio interface review

[mcdn]

This is a review and measurements of the M-Audio M-Track Solo audio interface. It is very much a budget offering, costing around $40 on Amazon US. I bought it for myself as a cheap interface for my bass guitar, but as you'll see from the results it can't do that job. On the other hand it is usable as a cheap low resolution digital-only interface so long as you recognise the specs are complete hallucinations.

No panthers here in Australia, but I will give it a two quokka rating out of five just because it's so cheap.

What do you get?​

• USB in and out at 48k/16bit
• Two analog inputs, available as L/R on the USB interface
  • Input 1/Left is a combo XLR and TRS, with the XLR being higher gain than the TRS. There's also phantom power available on the XLR input
  • Input 2/Right is a TRS with switchable gain via a "Line/Inst" switch
• Monitoring:
  • Headphone out on the front from a 3.5mm TRS
  • Line out on the back from 2 x RCA which are in reality the same as it's not a stereo device
  • A switch to choose between using the USB input for monitoring or "Direct" monitoring of the input signal
• Input gain knobs
• An output level control

Build quality​

The overall package is lightweight but solid. The case is all plastic with screws tying the top/back to the main body. This isn't going to fall apart in a hurry. Taking it apart reveals decent circuit board construction for the price, with sockets secured with through-hole soldering and/or pegs. On the other hand I had to get a second sample because the first one developed a fault in the left input, but that might have been down to abusing it a bit during testing.

Performance as an ADC​

Unless otherwise specified all measurements are taken at 48KHz/16bit input which is the limit of the USB interface. A Topping D10 balanced is the signal generator, a Cosmos ADCiso records analog outputs, and REW running on a MacBook Air is the software that does the work.

M-Audio quote between -94dB and -86dB THD+N (A weighted) for the various inputs, which is wildly optimistic. Noise alone is -93dBFS(A) a minimum gain, which limits us to 14.5bits of resolution out of the gate. After sweeping them all to find the optimal scenario I was able to get -87dBFS(A) from the right TRS input at 1.2V/+4dBu, but it was at a -9dBFS signal level so the SNR was only -78dB. You can increase the level but distortion goes up, or increase the gain but noise goes up. Here are the sweeps at mainly zero gain, which show the domination of noise until we get into a narrow sweet spot, then followed by a rapid increase in distortion.

Here's the 1kHz FFT for input 2 at the best settings I could achieve.

There are some strange wobbles in the frequency response on all inputs - only +-0.1dB at 2kHz-8kHz, but then getting bigger and leading to a -2dB drop at 20KHz:

This odd behaviour also shows up in the multitone test:

Overall the ADC performance is disappointing.

Performance as a DAC​

The headphone output driven from USB is as expected for the price - it's acceptable but no better than that, being almost entirely noise dominated. We can probably say it meets the 60mV spec at a gain of 7 into 30ohms.

The RCA output gets to 750mV/0dBU so it meets its spec for output voltage but is noisy:

Performance of analog in/out is questionable​

The M-Track Solo is almost unusable for direct monitoring. Switching the monitoring to "Direct" leads to results that I needed to check multiple times across two units. RCA output drops to 250mV best case but with no input headroom, and headphone out power is less than 5mW. This happened on both units so it's not a one off result. This was reflected in my experience when I tried to use it as an interface for my bass guitar, going direct from guitar->instrument in on the Solo->RCA out->bass amp it was unusably noisy. Switching to a Focusrite 2i2 solved all those issues.

Conclusion​

As a digital in/out interface it is just about acceptable for the price. Do not try to use it for direct analog in/out though.

 

[amirm]

Nice. Promoted to home page.

 

[MCH]

There are some strange wobbles in the frequency response on all inputs - only +-0.1dB at 2kHz-8kHz, but then getting bigger and leading to a -2dB drop at 20KHz:

View attachment 449003

Thanks for the review.
I also observed the same frequency response behaviour, Julien Krause also got the same result:

Post in thread 'My quest to eliminate the noise/rumble of my SVS SB-1000' https://audiosciencereview.com/foru...e-rumble-of-my-svs-sb-1000.46977/post-1860768

 

[JSmith]

No panthers here in Australia, but I will give it a two quokka rating out of five just because it's so cheap.

JSmith

 

[GWolfman]

Thanks @mcdn. I was considering this, glad I didn’t waste my time!

 

[Scrivs]

Thanks dude.

I came for the gut-shot, but stayed for the thorough analysis, rad

I play my bass through my Motu M4, into an old Icepower 200ASC, had them both for about 5 years. Would never have known about Motu, RME, Hypex without people like you generously sharing what they’ve learned about their own gear along the way.

Mahalo nui loa

 

[DrZingo]

I'm a bit confused - what do you mean by using it as a "digital in/out interface"? There doesn't seem to be any digital audio connections such as s/pdif or toslink, so isn't all you can do with it analog in and out through USB (which both seem to be poor)?

 

[mcdn]

I'm a bit confused - what do you mean by using it as a "digital in/out interface"? There doesn't seem to be any digital audio connections such as s/pdif or toslink, so isn't all you can do with it analog in and out through USB (which both seem to be poor)?

Yes, I mean using the USB rather than direct monitoring

 

[AnalogSteph]

Overall the ADC performance is disappointing.

Not a real surprise. The PCM29xx is about on the level of 2004 era onboard audio. The wiggles (not the band) are digital filter periodic passband ripple.

At least they seem to be getting more out of the thing than Behringer, so there's that. Even if the "plastic fantastic" case can bite you in terms of RFI.

I think a major use case for units like these is podcasting, as input noise is actually decent and well-suited to dynamic mics (think XM8500 or SM48/58). Dynamic range primarily tends to be limited by microphone self-noise and preamp noise then, and fidelity requirements for speech aren't particularly stringent either. Nothing running its analog stage on +5V is going to handle guitar pickups with authority, the input level handling just isn't there - if in doubt, go DI.

What I would still like to see is frequency response and multitone or THD(f)/IMD(f) at max gain, 10 dB below, 20 dB below. Maybe around -6 dBFS, which would be a common target peak level for speech recording. I suspect that a lot of inexpensive audio interfaces have various warts in the preamp department that often go partially undetected, and then you wonder why people hate the preamps (Arturia Minifuse comes to mind). The much-maligned Steinberg interfaces ought to be pretty good in this regard.

 

[Jimster480]

It is amazing that companies actually produce products like this in the 2020s.... when the specs weren't good for 20 years ago.

 

[mcdn]

@AnalogSteph you mean this? 2mV seems typical for dynamic mic output, so backing the input gain knob off a bit to around 9.5/10 gives decent headroom.

FR is unchanged at high gain on the XLR input, no new issues there. Here's the THD+N/freq results. REW doesn't do automated IMD/F tests yet.

 

[mcdn]

Thanks @mcdn. I was considering this, glad I didn’t waste my time!

The thing is I'm not sure what else there is in the truly budget end of the market. A Scarlett Solo has no real flaws but is twice the price.

 

[dasdoing]

2 x RCA which are in reality the same as it's not a stereo device

wtf? lol
are you sure?

 

[AnalogSteph]

FR is unchanged at high gain on the XLR input, no new issues there.

That's more than a lot of other budget audio interfaces can say for themselves. Listen to the SM7B samples in the ESI Neva Uno/Duo review... including the noise floor, this is approaching telephone quality.

2mV seems typical for dynamic mic output, so backing the input gain knob off a bit to around 9.5/10 gives decent headroom.

I vaguely remember complaints about the gain dial being very compressed up top, looks like they weren't kidding. I have encountered this before, but perhaps not as extreme. Maybe 20 dB between 3 and 5 o'clock. This position normally requires a fairly special reverse log pot, maybe a plain linear one was used here.

FR is unchanged at high gain on the XLR input, no new issues there. Here's the THD+N/freq results. REW doesn't do automated IMD/F tests yet.

Hmm, I guess this is dominated by the +N part. Can REW do automated THD sans N?

 

[mcdn]

wtf? lol
are you sure?

Actually no I’m not, it could per input or it could be mixed. I’ll check

 

[willyampz]

Maybe 20 dB between 3 and 5 o'clock

Mine's like that sensitive between knob settings 9.5-10, where it jumps from much less to more gain than the focusrite in that range. I had heard about this issue but was like "nah who cares, it has an ASIO driver". I can't get enough gain with my condenser mic. and I think there could even be gain modulation if any slight vibrations happen on the table or something when set to gain I want. Not accceptable. I was able to get same latency with asio4all as well. Returning for maybe the behringer. Oh well, I am learning, I am new to this stuff.

 

[mcdn]

Actually no I’m not, it could per input or it could be mixed. I’ll check

I checked and in direct monitoring mode the analog outputs are a mono mix of both analog inputs. In other words output L and output R are identical.

In USB monitoring mode the analog outputs are L/R as per the stereo (thank you for making me check) output from the laptop.

I haven't tried running both at the same time!

 

[mcdn]

Hmm, I guess this is dominated by the +N part. Can REW do automated THD sans N?

My mistake that graph is in fact THD, not THD+N, as shown by the label in the bottom left

 

[AnalogSteph]

Then -6 dBFS must already be in the "clipping" range of quickly rising distortion, does that seem correct? Or where about in the previous THD+N/level graph would these points have been?

A similar interface for comparison:

Quick measurements of Behringer UMC22

I've had this little interface for many years and it's been a champ. It currently sells for about $56 and I use it for room response measurements and other tasks involving a microphone. The unit is USB-powered and provides 48v phantom power to the mic, as well as instrument input, headphone out...

www.audiosciencereview.com

That one seems to be missing its -2 dBu spec for mic input level by a long shot then, 265 mV is barely over -10 dBu. The Solo makes it to -1(ish) dBu at least.

 

[mcdn]

Then -6 dBFS must already be in the "clipping" range of quickly rising distortion, does that seem correct? Or where about in the previous THD+N/level graph would these points have been?

A similar interface for comparison:

Quick measurements of Behringer UMC22

I've had this little interface for many years and it's been a champ. It currently sells for about $56 and I use it for room response measurements and other tasks involving a microphone. The unit is USB-powered and provides 48v phantom power to the mic, as well as instrument input, headphone out...

www.audiosciencereview.com

That one seems to be missing its -2 dBu spec for mic input level by a long shot then, 265 mV is barely over -10 dBu. The Solo makes it to -1(ish) dBu at least.

Per the chart in the review post, you can get -2dBu on the XLR input at minimum gain. Around -3dBFS for what it's worth, but YMMV depending on gain levels