MOTU M4 Loopback Measurements

[aos]

I've used the same and i got -3.2dB. Not sure what needs rechecked in your setup, unless the cables are having issues (very unlikely I'd say).
I get the same figures with both Windows MME and ASIO drivers.


Playing a 440Hz sinewave @0dB should get you at least 4.6V RMS on both 3/4 and 1/2 outputs (volume to the max. for 1/2 output).

How do you adjust the volume when using ASIO? Neither line out nor the line in offer volume adjustment and knobs don't do anything. Even when I make a WAV file from the generator at presumably -3dBF and play it back on another device via SPDIF (DAC of my own design for example) I get say -20dB on inputs and windows itself shows digital signal level well below maximum. It's like I am getting 20dB attenuation on anything coming out of the PC no matter which way.

EDIT: it does seem like my windows simply attenuates everything, no matter what output, at the digital level. I am not sure how is that possible especially for ASIO.

 

[trl]

Outputs 1/2 can be controlled from the main knob, but on outputs 3/4 there is max. output volume (what Amir was actually tested).

Double check your drivers, mabe reset the interface to defaults and don't forget to reboot Windows too, maybe it will help.

 

[RayDunzl]

I am using TRS cables for loopback from line output 3 to line input 3 on the back of the unit.

Wouldn't that create a feedback loop?

Shouldn't the output of Ch 3 be connected to the input of a different channel?

 

[trl]

Never seen such issue before, I don't think it eill create an actual loopback. On IXBT website I see that all their tested interfaces were measured this way.

 

[aos]

Yeah, rebooted twice already. Even on second PC, a laptop, I am still getting only 1.83Vrms at -3dB. That yields -17.5dBFS when running ARTA spectrum analyzer in loopback generating 1kHz sine. Funny enough if I switch to WDM, I measure -11.4dBFS! But at calibration I measure the same 1.83V. I don't see how this is possible, and on 2 different PCs as well. Perhaps it really is the cable, I will try another one.

EDIT: yes, it was the cable. I used a 1/4 plug that only has 2 contacts. I was measuring one differential end to the ground, not to the other one. Numbers are good now, thanks for the help.

 

[aos]

I do still have the issue that my desktop does not show all inputs and outputs in WDM mode (it does in ASIO). On laptop it works fine. I reinstalled drivers and rebooted, no change.

 

[trl]

EDIT: yes, it was the cable. I used a 1/4 plug that only has 2 contacts. I was measuring one differential end to the ground, not to the other one. Numbers are good now, thanks for the help.

OK, good that you found out the culprit after all. Worth mentioning that by using such a cable you actually created a short-circuit for the inverted circuit on each of the two channels.

 

[trl]

I do still have the issue that my desktop does not show all inputs and outputs in WDM mode (it does in ASIO). On laptop it works fine. I reinstalled drivers and rebooted, no change.

This is definitely something that you will need to figure it out somehow, maybe with Motu's support. You could also reinstall the OS from scratch, because sometimes existing audio drivers (previously installed or uninstalled correctly) might interfere with the new ones; same applies is multiple ASIO drivers coexists on the PC (inc. ASIO4All).

 

[AllanDavidson]

Since I'm still waiting for my MOTU to arrive, here are my motherboard integrated audio measurements.

ASUS TUF Gaming X570-Plus (Realtek ALC S1200A)
Line Out -> Line In

The Line Out Volume is set to 85, and the Line In level is set to 76, these are the sweetspots with best THD+N.


1KHz

Multitone

 

[AllanDavidson]

Is there anyway for you to measure the adc and dac independently? I would like to know which one is the limiting factor. As well as the distortion with various levels.

It is possible.

The back panel have 4 outputs, Line out channels 1/2 are tied to the internal system and monitored by the front panel, while Line out channels 3/4 are fixed volume and there's no monitoring out of them (DAC only).

The back panel also have 2 inputs, and both are likely to be straight to the ADC (no preamps), or have a low power, R2R, Delta Sigma preamp acting as buffer (yeah...)

 

[AllanDavidson]

This has been discussed already in a different thread, not sure where right now. Please check M2/M4 datasheet and look for the 1 MOhms input impedance for the M2 that means that this is not a real line input, but more like an instrument input that is able to accommodate line level signals too.

The Clarett seems to have a real line input, at least by looking at the input level requirements. However, I own a Solo Gen3 and I never had any issues feeding it will line-level signals, so this is why I am confident that Clarett should do just fine in this area.

Maybe the unit inputs have 1M Ohms for High-Z instruments like a Guitar, and they may automatically change to low impedance when using as Line In. This needs to be tested.

Problem is that using 2V RMS or 1V RMS on the inputs 1/2 (because inputs 3/4 are not gain adjustable) will increase the noise a lot, no matter I'm using Line or Mic from Inputs 1/2.

View attachment 93371
Line In 3/4, Line Out 3/4 @2V RMS, SINAD of 93dB
​

​

Yup. The same thing happened on Amir's review, noise rises as frequency rises on the ADC. ¯\_(ツ)_/¯

 

[lizhuoyin]

Is there some sort of attenuator in MOTU M4's LINE IN 3/4? The reason I ask is that if I connect LINE OUT 3 to LINE IN 3 and send signal to LINE OUT 3, the M4's IN/OUT meter bars are not at the same level. IN is around 3.8 dbFS shorter than OUT. Verified in REW.

 

[trl]

That VU-meter is relative to dBFS and not to dBU or dBV. Basically, there should be no correlation between output volume and input sensitivity. You should see that VU-meter as a simple dBFS indicator, nothing else; it will get to red when approaching 0dBFS, just read the manual to see output and input voltages of this card for a better understanding (Max Level Out +16 dBu which is 4.88 V RMS, Max Level in +18 dBu which is 6.15 V RMS).

 

[lizhuoyin]

Thanks, @trl. So, you say it's normal that OUT 3 is different from IN 3 in M4's VU meters when connecting them thru trs balanced? Like showing in following picture:

Now I get it now. IN 3 has different range from OUT 3.

 

[lizhuoyin]

Also I set -1dBFS in generator and there is a 3.84 shift.:

 

[isabido]

Hi friends!, I have made two measurements in loops. Through the balanced TRS outputs.

1st OUT 1 to IN1
2nd OUT1 to IN3

I have been surprised that the IN3 and IN4 inputs have a better response in frequency. I leave you the values collected in REW performing the sound card calibration procedure.

Is this normal?

Input device: MOTU M Series
Input: 3: In 3
Channel: Left
Output device: MOTU M Series
Output: 1: Out 1
Input RMS target: 0,0 dB
Actual RMS at 1 kHz: -9,1 dB
Sample rate: 48000 Hz
Input volume: no control, Sweep level: -6,0 dB
20 Hz .. 20 kHz flatness: +0,0, -0,4 dB
-3 dB points: 2,2 Hz, 23,024 kHz


Input device: MOTU M Series
Input: 1: In 1
Channel: Left
Output device: MOTU M Series
Output: 1: Out 1
Input RMS target: 0,0 dB
Actual RMS at 1 kHz: -7,7 dB
Sample rate: 48000 Hz
Input volume: no control, Sweep level: -6,0 dB
20 Hz .. 20 kHz flatness: +0,0, -2,3 dB
-3 dB points: 16,8 Hz, 22,968 kHz

​

 

[trl]

Doesn't looks quite OK to me, so do find an external ADC/ audio interface, so you can remeasure your Motu. This way you might be able to find out if the issue comes from the Inputs or Outputs of the Motu or, why not, your measurements might have a flaw in the cabling or plugs.

 

[isabido]

Doesn't looks quite OK to me, so do find an external ADC/ audio interface, so you can remeasure your Motu. This way you might be able to find out if the issue comes from the Inputs or Outputs of the Motu or, why not, your measurements might have a flaw in the cabling or plugs.

Hi Trl, It can't be, in both measures it's the same output and the same cable.

I see the problem in the lower part of the frequency from 0 to 20hz.

On the front inputs IN1 and IN2, the response -3 dB points: 16.8 Hz, 22.968 kHz (20hz -2.3 dB)

On the rear inputs IN3 and IN4, the response -3 dB points: 2.2 Hz, 23.024 kHz (20hz -0.4 dB)

From what I've read the XLR/TRS combo front inputs have a very high impedance of 1Mohm for guitars, I don't know if this is the problem. And if this would be repeated when a measurement microphone was connected, since -2.3dB at 20hz I see it as quite significant.

 

[trl]

I do see your point now, sorry I didn't got it in the first place. It's kinda big roll-off @20Hz, maybe I'll give it a try to mine and check if it has the same behaviour too.

 

[isabido]

Ahora veo tu punto, lamento no haberlo entendido en primer lugar. Es un gran roll-off a 20 Hz, tal vez lo pruebe y verifique si también tiene el mismo comportamiento.

Thanks TLR for your quick response, In Amir's tests, I see that he does everything with the rear inputs, 3 and 4. Maybe that's why no one noticed this.