Audient EVO 4 [Audio Interface] Review

[peniku8]

I needed an audio interface to connect a measurement mic to my laptop for live shows and to get audio to the mixing console. I chose the EVO 4 for its digital gain control and bought a B-Stock unit from Thomann for 105€.

The lower part of the housing is a matte black plastic, which is not scratch resistant in any way. You touch it, you'll leave a mark.
I installed the driver and skipped the account creation (the hell is up with that?) to access the UI. Sadly the gain control has no numerical values, so you'll have to either guess or count what gain the inputs are set to. Each step is 1dB of gain from 0 to 58.

Let's have a look at the DAC measurements first:

-104dB THD+N

The distortion vs level profile looks quite interesting. Is this the 'infamous' ESS hump?

Dynamic Range: 109,7dB

IMD ENOB: 13,8 bits

Distortion vs frequency:

Frequency response: -3dB @2Hz

I'm surprised by how flat it is even at 10Hz. I like that.
All in all, no glaring issues! The only oddity I noticed during my testing was the output not being differential. It's sadly just impedance balanced. This will reject noise from the cable just as well (if properly implemented), but the output itself won't be as clean as it could be. I would've expected SINAD to have been at least 3dB better if it was differentially driven.


Onto the ADC measurements:

I had to set the gain to 4dB in order to be able to max it out with the EX5 as signal generator. This resulted in the following THD vs level graph:

We see an early rise in distortion. SINAD is best at -8dB on the input side, so let's see if we can get a better result with a lower generator output:

Sure enough, SINAD improves a tad.

-94dB THD+N

Dynamic Range: 102dB

Since we have plenty of gain available, I decided to look into dynamic range at maximum gain too.
At 58dB gain I measure THD+N at -70dB. This is with the EX5's generator at -55dB, hinting at a dynamic range of 125dB for the DAC. This matches the DR value in the data sheet and is 5dB higher than Amir's measurement. Not sure what value is correct now, but the take away is that the EX5 is the limiting factor here. I plan on building a 1:100 divider (-40dB pad) to remove this bottleneck from my measurement setup.

Max gain, generator DR bottleneck:

IMD ENOB: 12,5 bits

What's going on in the low end?

Distortion vs frequency plot to reveal this issue:

THD at 100Hz is almost an order of magnitude worse than at 1KHz.


Frequency Response: -3dB@7Hz // -0,7dB@20Hz

I already had two shows with the interface and it does everything I need it to. It keeps and remembers gain values once I set them, keeps on phantom power once its enabled and the controls with the LED ring are surprisingly nice. It's simply just plug and play, exactly the way I want it. I have the gain at 11dB, which gives me headroom to 145dBSPL with my mic, which is way more than I need. The performance is also decent for the price!
The only weird thing I noticed so far was some crackling issues. I left the mixing console for half an hour to come back to loud crackling that the interface produced (while nothing should be playing). I re-plugged its usb cable and had no more issues for the rest of the show and the show the day after. Not sure where that came from.
I'll keep the interface and use it for the upcoming shows. I'll report back if I ever run into any issues.


Cheers!

 

[Blumlein 88]

Nice work. You mention using the EX5 as a source for testing the ADC. Was this the Topping EX5 DAC as there are a few devices named EX5? Also maybe I missed it, but what were you using to measure the output of the DAC?

 

[peniku8]

Nice work. You mention using the EX5 as a source for testing the ADC. Was this the Topping EX5 DAC as there are a few devices named EX5? Also maybe I missed it, but what were you using to measure the output of the DAC?

The Topping, indeed. DAC measurements were all taken with the Cosmos ADC. Scaler is on the way.

 

[msmucr]

I'm not 100% sure about design of EVO range, but preamps and DI input stages at many other Audient interfaces and standalone preamps are deliberately "voiced" to have certain house sound. This is is bit different compared to usual aim to lowest distortion from given circuitry among other interface manufactures (say like MOTU). For example at their iD range they use discrete class-A JFET circuitry for high-impedance DIs, which has subtle "flavor" compared to generic clean DIs. This can be seen as a good thing in particular context (eg. recorded instrument sound could be perceived as richer with added weight and clarity). Some of their interfaces like iD-22 or iD-44 has its first two input channels with balanced insert points, where you have fixed sensitivity (0 dBFS is 18 dBu). So if you use this balanced return instead of XLR or front DI input, you'll bypass the mic preamp and DI circuitry and gain cleanest signal. The remaining line level input channels at those interfaces are similar with fixed sensitivity.

I haven't measured nor inspected EVO4, but this might be the reason for measured increased input distortion. To me it looks quite tidy and intentional including rise of HD towards low frequencies (akin to some outboard or console pres) to be qualified as some design overlook. However I can be wrong of course. I've installed this EVO4 interface just one time for podcasts and voiceovers at friend's place and then it seemed pretty good for this application and given budget price range. Solid pres, decent headphone output, USB class compliant mode and internal loopback.

 

[AnalogSteph]

The distortion vs level profile looks quite interesting. Is this the 'infamous' ESS hump?

Sure looks the part. Which would mean that the innards have changed as a result of the chip crisis, much like in other interfaces - it started out as an AKM job (I'm pretty sure that the chip labeled "4452" next to the AK5552VN ADC is an AK4452 DAC).

Speaking of ESS:

ESS Technology Introduces New Stereo High Performance CODEC – ESS Technology, Inc.

www.esstech.com

What's going on in the low end?

Distortion vs frequency plot to reveal this issue:

THD at 100Hz is almost an order of magnitude worse than at 1KHz.

You'll notice similar in other inexpensive interfaces with digitally-controlled input amps, like the MOTU M2. If it's not the -5V inverter (which I somehow doubt), I guess it's a chip quirk.

The only weird thing I noticed so far was some crackling issues. I left the mixing console for half an hour to come back to loud crackling that the interface produced (while nothing should be playing). I re-plugged its usb cable and had no more issues for the rest of the show and the show the day after. Not sure where that came from.

Bus-powered interfaces are somewhat prone to power-related hiccups. Could be poor contact or a dodgy cable. Or maybe some USB power saving feature it didn't like.