Audient EVO 16 - Experiences

[cupakm]

No, but I can measure it if someone tells me how to (big plus if there is an Ubuntu package).

Thanks! You will need a loopback cable - connect physical output with a physical input. For example take a jack-jack cable, insert into Mic/Line input 8 and Line output 8.

Then there are two options:

1) Ardour - when starting up, usually Audio/MIDI Setup window opens, choose Audio system "ALSA", Input device Evo16, Output device Evo16, Sample rate 48 kHz, Buffer size as low as possible without getting sound artifacts, Periods 3. (You may want to experiment with sample rate, buffer size and periods and see how they change the latency); then -> Advanced settings -> Calibrate Audio -> Playback channel choose Line Output 8, Capture channel choose Mic/Line Input 8, then click Measure button. After a brief beep-tone you'll get systemic latency and round-trip latency (RTL) in [ms]. Let us please know the numbers
Then you may want to go ahead and use your measurement to let Ardour compensate the latency.

Note the the actual number of Input / Output Latency samples may change when you reconnect the interface.

2) JACK only - let me know if the above mention won't work, this one is I think less convenient and can result in a tad bit higher RTL

Chipset, do you mean my PC's processor?

No, your USB controller's chipset. That one where you connect the Evo into. For example Renesas uPD720201:

Code: Select all

marek@desktop:~$ lspci | grep Renesas
03:00.0 USB controller: Renesas Technology Corp. uPD720201 USB 3.0 Host Controller (rev 03)

...or Fresco, Intel, AMD, VIA...

[icaria36]

Thank you. I have checked some combination, without really knowing what I'm doing. Following instructions.

32 samples
Detected roundtrip latency: 255 samples (5,312 ms)
Systemic latency: 159 samples (3,312, ms)

2 periods: 223 (4,646) & systemic 159 (3,312)

64 samples
375 (7,812 ms)
183 samples (3,812, ms)

96
472 - 9,812
183 - 3,812

128
567 - 11,812
183 - 3,812

192
759 - 15,812
183 - 3,812

1024
3255 - 67,812
183 - 3,812

3256 - 67,833
184 - 3,833

2 periods: 2232 (46,500)

USB controller: Intel Corporation Alder Lake-S PCH USB 3.2 Gen 2x2 XHCI Controller (rev 11)

I'm curious about why you ask and what does this data tell you.

[cupakm]

I'm curious about why you ask and what does this data tell you.

Thanks a lot!

What the data do tell us:

• Evo16 can go for very low buffer size (32 samples)
• it achieves sub 6 ms round-trip latency, around 5-ish. One most probably could comfortably record overdubs even without Ardour's latency compensation.

USB controller: if you're still curious whether you could squeeze even lower RTL, you may want to experiment connecting the interface to other USB controllers. Ideally reserve one USB controller for the interface only, so no other USB device could interfere. In case of a PCIe expansion card with multiple ports and a single controller (for example one offering 4x USB 3.1 port per controller), you'd connect only the interface and nothing else to that card. Trying out another controller will vary the results.

Even simpler method how to further lower the latency (but increase computing requirements) is using higher sample rate. I suppose you did the measurements at 48 kHz. Try doing the same at 96 kHz.

What could be interesting too is measuring RTL in stand-alone mode, which would mean connecting it via ADAT ports to a host interface and measuring latency of both host and Evo16 through that host, and subtracting the host's latency. Why would one do that? To see how much latency will be added expanding another interface. I guess Evo SP8's latency would be pretty same, as I suppose they're internally almost identical, but this is only my speculation.

Anyway, given you did not notice any sound artifacts, pops, noises, etc. on lowest RTL settings -and- cost of the unit, I would deem the interface's latency quite usable.

[icaria36]

Thank you very much, I'm impressed with this knowledge!. Since I'm happy with the performance of my setup, I won't touch anything. The only one lagging and making cracking noises in my studio is... me.

[cupakm]

Just one more question...

Since I'm happy with the performance of my setup, I won't touch anything.

...maybe you could how hot is it running?
If you'd have a contactless thermometer at hand, could you give it a try? After like an hour or so of usage, or at least idling.

[icaria36]

Sorry, when you asked I didn't have a way to measure this and, to make things more unreliable, I had a synth sitting on top. Now the synth is gone, I still don't have a thermometer, but I can tell you that the unit (in my use) just gets a little warm around the power port on the back right corner.

[damusicman29]

Also sound output starts to only be audible if you put it to 40%+, both on speakers and headphones. On the inputs there also seems to be some kind of amp "threshhold" until it really starts working but im not so sure about that yet, since I still need to gather moroe experiences with this device. Could also be a problem with my configuration.

I'm not on Linux, but I was searching for a solution to the issue I'm having. I just got this interface and read what I quoted from you ^. I'm on a Mac. Nothing fancy in installation, just plugged it in and started listening to some drum samples. I noticed that the verrrrrry small few milleseconds of the beginning of an attack of a transient on a drum are cut off until I really prime it by repeatedly playing the sample over and over again, by which it comes out of the headphones properly. I recorded a video to test this and found that in the video, the sound was normal everytime I played it back. I don't know why the headphone amp has to be "warmed" up if there's no sound. It's weird. For comparision, I plugged my little cheap Scarlett Solo in and it works like normal, so this is definitely the Evo. Thoughts?

[cupakm]

I plugged my little cheap Scarlett Solo in and it works like normal, so this is definitely the Evo. Thoughts?

Hi there. You could try plugging the interface(s) to another machine (be it Win / Mac / Linux...) to try if it really is the Evo, or some config mishap.

[icaria36]

I haven't noticed any noise coming from the audio interface in my setup. If it's there, it isn't bothering me or the couple of producers I work with.

Also sound output starts to only be audible if you put it to 40%+, both on speakers and headphones.

I hadn't noticed anything strange with the volume, and I had just tested this. I have Ubuntu and Bitwig. I set the test Tone device in one track to Gain 0db, and the track and Master with volume maxed out. I also max out volume in my monitor speakers. I'm in a normal, untreated room.

• On the monitor speakers (PreSonus Eris 3.5), I can start hearing the test tone at 8% volume between the expected hiss from the monitors. Maybe someone with better ears and better monitors can hear it below.
• On the headphones (Beyerdynamic DT 990 Pro), I can start hearing the test tone at 11% volume. Same thing, maybe others can hear it before in other conditions.

Sounds normal to me, and in any case I rarely go below 50% in my setup.