Measurements of Motu UltraLite mk4 (AKM DAC Version)

[Blumlein 88]

Conclusions, well as mentioned the Ultralites A to D isnt as good as the 8A, however the D to A is a tough one. The performance is different, the Ultralite showing less spuria in some tests but worse jitter noise. Is any of this obvious in listening, I doubt it. Im going to have to listen and give it a try.

You could record some music with the ADC in the 8A from both the 8A and Ultralite and let us have a blind listen for differences. Or make it more obvious by doing multiple loopbacks. Maybe 4th generation loopbacks for each device using music and let us listen for differences.

 

[March Audio]

You could record some music with the ADC in the 8A from both the 8A and Ultralite and let us have a blind listen for differences. Or make it more obvious by doing multiple loopbacks. Maybe 4th generation loopbacks for each device using music and let us listen for differences.

4th gen loopbacks it is. Excellent idea

 

[pos]

Great review, thanks!

It looks like you have been testing the "main" output only. As this device has 10 outputs for (supposedly) two ES9016 converters there might be a difference between the main and the 1-8 outputs. Have you compared them?

Any plan to test the 24 Ao?
(If 8 outputs is not enough for your currents needs, then 10 will not cut it for long )

 

[March Audio]

Great review, thanks!

It looks like you have been testing the "main" output only. As this device has 10 outputs for (supposedly) two ES9016 converters there might be a difference between the main and the 1-8 outputs. Have you compared them?

Any plan to test the 24 Ao?
(If 8 outputs is not enough for your currents needs, then 10 will not cut it for long )

Will check them when I get a chance.

 

[March Audio]

OK, here are 3 files. One is the original the other 2 are 4th generation copies made on the Ultralite and 8A respectively.

Which is which?

Please PM me with your answer and dont post your thoughts in the thread so as not to influence others

http://alanmarch.asuscomm.com:5000/sharing/7rHB6YkMm

 

[t129]

Would you think there is any value to do a thd/frequency sweep to see if either diverges from flat? I don't seem to recall a reliable source where it was postulated that such effects could be audible (it is one of the audiophile myths for the cult of no global feedback though), but at least Bruno Putzeys seems to believe it.

Second thing is the the imd/level hump that seems to plague the later Sabres. (John Siau says (in the Benchmark DAC3 thread) it comes from the lacking common mode rejection of the buffer/IV-stage, not the dac chip (I might also have gravely misunderstood the explanation)) I wonder if MOTUs are immune? Be it due older dac or better buffer design.

But I digress, thank you for measuring these multichannel interfaces. I find the results and the devices highly interesting. (DSP, usb interface and higher quality multichannel dacs in one box, thumbs up for the diy dream devices at reasonable prices)

 

[March Audio]

It's certainly worth looking further into the thd. I saw Johns comments.

I think these professional interfaces are offering good performance and probably do better VFM than much of the HiFi domestic products.

 

[Blumlein 88]

Be sure to listen to the 3 files. One is an original, and the other two are 4th generation loopbacks. Surely any coloration for the devices compounded by 4 trips thru ADC and DAC would be made more noticeable. Listen and see what you think. See if the digital original is audibly different.

 

[March Audio]

Just had a bit of a surprise. I took the lid to take some pics for those of you that are into your electro porn.

Now here is the Motu Website. Note the references to ESS Sabre DACs.

Now take a look at whats actually inside

AKM 4413EQ

https://www.akm.com/akm/en/file/datasheet/AK4413EQ.pdf

This simply doesnt meet the claimed specs of 112dB THD+Noise and 123dB dynamic range. Its only 104dB and 120dB.

Motu will be getting an email with a please explain and a WTF?

 

[March Audio]

Email sent, waiting reply from Motu.

 

[Blumlein 88]

Wow! Now that is interesting. What was actually in the 8A? Nevermind, I looked back and saw it was an ESS DAC and Cirrus Logic ADC.

BTW, that is the DAC that is supposed to be in my Focusrite Forte though I've not opened it up. The Forte uses an AK5388 ADC.

 

[March Audio]

The 8a has what it says

The ADC in the ultralite is a Cirrus Logic CS5368. https://www.cirrus.com/products/cs5364-66-68/ Nothing to write home about, but it wasnt the A to D performance I was interested in on this product, but at least it meets spec.
The CS5381 in the 8A is significantly better. https://www.cirrus.com/products/cs5381/ which is on par with your AK5388 ADC


Just guessing, but this is a rev 1 board and they must have changed it to a later revision with the ESS chips. Still, if thats the case, with such different performance it should be a MK5!

 

[t129]

My computer setup at the moment is not even remotely transparent enough to say anything really conclusive about the comparison files (integrated Realtek and a weird English op amp head amp using some video chips whose type number escapes me + HD650). But I did test them. It was fun. There might be an audible difference.

For some reason, even the manual lives in the past and claims measly -101dB THD+N and has no mention of the Sabres, so there probably has been some kind of board update. I probably would not mind the AKMs though with the measured performance so far, if only they were upfront about them... @BE718 What does it say on the retail box? I seem to remember somebody posting a picture of the cardboard box boasting the Sabres. Perhaps that is the way to differentiate the old and the new: box art.

 

[pos]

Just guessing, but this is a rev 1 board and they must have changed it to a later revision with the ESS chips. Still, if thats the case, with such different performance it should be a MK5!

Looks like new revisions have the ESS DAC explicitly mentioned on the box:
https://audiosciencereview.com/forum/index.php?threads/motu-8a-measurements.2342/page-2#post-83002

Did you buy it new?

 

[Blumlein 88]

Looks like new revisions have the ESS DAC explicitly mentioned on the box:
https://audiosciencereview.com/forum/index.php?threads/motu-8a-measurements.2342/page-2#post-83002

Did you buy it new?

That may be the explanation. A complete change of DAC chips and ADC chips sure seems like a reason to call it an Ultralite mk 5 or at least an Ultralite mk IV version 2 or Ultralite mk IV ESS or something to make it clear.

I've a similar situation with my Antelope audio Zen Tour. There are two revisions. They claim increased dynamic range on the ADC side between them. I've the first revision (which I knew when I purchased it). They don't tell anywhere what they've changed. If you know where to look they show 120 db DR for the ADC on the old version, and 124 db DR on the new version. Yet all update to the same software, and the same FPGA based effects.

Then you have the Focusrite Scarlett series. Now in Gen 2 versions. They all use the same components and ADC/DAC chips. Two things changed. Gen 1 series only worked to 96 khz though the chips could do 192 khz. Gen 2 goes to 192 khz. There is a new software interface between the PC and the device. Otherwise some minor cosmetic improvements on level displays and knob markings. Minor adjusting of gain ranges. One might even complain too little was changed. It beats having two versions that aren't made known however.

 

[March Audio]

OK to be fair Motu got back to me very quickly. Apparently all the mk4s shipping to distributers are the new Sabre versions and it looks like this one is old stock held by the store. They were very apologetic and said they would help me sort it out. Non specific as to what that help would be but they have the stores contact details and I have suggested they ship a new version to the store where I can exchange it.

Waiting for their response.

 

[March Audio]

Good news, Motu are arranging a replacement Ess version so all is good with the world

It sounds like the old units should have been returned and swapped out by the stores, this one being a straggler. So it will be interesting to test the new version.

Again to be fair Motu have been extremely quick in response to do the right thing, so ultimately no complaints from me as this is " just one of those things" or "shit happens "

Ess sabre 32 versions have ES in the serial.

 

[Blumlein 88]

Good news, Motu are arranging a replacement Ess version so all is good with the world

It sounds like the old units should have been returned and swapped out by the stores, this one being a straggler. So it will be interesting to test the new version.

Again to be fair Motu have been extremely quick in response to do the right thing, so ultimately no complaints from me as this is " just one of those things" or "shit happens "

Ess sabre 32 versions have ES in the serial.

Well sounds like they are making it good on it. Can't complain about that, and as a result we'll have measurements of both versions. That will be interesting in itself. Knowing about the ES in the serial number is useful info for purchasers too.

 

[March Audio]

Oh also I have found a video that shows the box has an ESS Sabre 32 label clearly on the front which mine obviously doesn't have.

 

[March Audio]

OK, someone needs to do a sanity check on the following results. Struggling to believe them, but I cant see whats wrong.

Someone asked about IMD v Level results for the Motu Ultralite so I decided to make a graph similar to Amirs. I copied data from one of Amirs graphs for the Schiit Yggdrasil, Topping DX7 and Benchmark DAC3 and proceeded to measure my Motu 8A in loopback to check the process.

I used REW RTA to measure the distortion at discrete 5dB points, so it isnt as high granularity as Amirs plots, but should suffice. I used http://www.sengpielaudio.com/calculator-thd.htm to convert to dB values.

The results were not as expected! Hard to believe, however if you look at the above -20dBFS plot there genuinely is bugger all IMD.

Thoughts anyone?

@amirm , I note on your plot it says SMPTE / DIN. Can you confirm which was the standard used as my understanding is the tone frequencies are different.

-10 dB just for info