• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Review and Measurements of Matrix Audio X-SPDIF 2 USB Converter

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,395
Location
Seattle Area
This is a review and detailed measurements of the Matrix Audio X-SPDIF 2 USB to S/PDIF and I²S (IIS) converter. It is on kind loan from member who also sent me the Wyred4Sound DAC-2v2 SE DAC. The X-SPDIF 2 costs USD $379 but the company is selling for reduced cost of USD $341.10. Some complicated math must have gone into that computation to come up with that extra 10 cents in the price! :)

The S-XSPDIF 2 comes in a small but very heavy enclosure. You can see it almost lost in the pool of gear I had to use to test its capabilities:

Matrix Audio X-SPDIF 2 USB to SPDIF and I2S Converter Review.png.jpg

It is the non-descript box on top of the Wyred4Sound DAC. To the right, is the Wyred4Sound PS-1 multi-voltage power supply. It has modules you buy that give you different output voltages. The owner supplied that so I tested with and without it as the X SPDIF-2 can run on USB power alone. The PS-1 costs $399 for a single output voltage. Each extra output costs an extra $125. In other words, in base configuration with a single output, it will cost you more than the Matrix Audio X-SPDIF 2 converter!

In addition to S/PDIF output, the X SPDIF-2 can also output I²S interface. It does this as commonly done using an HDMI connector and cable. Note that this has nothing to do with HDMI interface. They are simply using the connector and cable because it can carry high-speed signals and is cheap due to its mass usage. The owner was interested to know if the I²S interface works any better than S/PDIF with the Wyred4Sound DAC, hence the combo.

Let's get started with measurements and see what we have.

Measurements
I am still refining my measurements here so please excuse the continued changes in settings and graphs used. Also, we are measuring such small signals that I don't yet trust repeatability of them. So for this review, I have re-run all the measurements again so don't try to compare them verbatim to previous tests/reviews.

Fort the first test, I thought I address owner's need directly which is to see if there is a difference in Wyred4Sound DAC-2v2 SE DAC when using its own USB input, versus either S/PDIF or I²S. Here is the output using our jitter test:

Matrix Audio X-SPDIF 2 USB to SPDIF and I2S Converter Jittter Spectrum from Wyred4Sound DAC-V2...png


There is absolutely no difference! All three interfaces produce the same analog output. You know, the thing that hits your ears.

Now, there may be differences there but they are hidden in rather high jitter spectrum of Wyred4Sound DAC V2SE. It is the "weak link" as they say. This was a disappointment as I could no longer test using this DAC and had to resort to Topping D50. Unfortunately I no longer have a DAC with I²S interface so this is the end of the road for that analysis.

My Audio Precision APx555 can analyze jitter by itself. It extracts it from the S/PDIF stream and can then perform analysis on it as if it were audio signal. As such we can run our FFT analysis and see what we find as far as the spectrum when feeding the devices the "J-test" signal. J-test at 48 kHz sampling which I tested is a 12 kHz square wave with an embedded one-bit toggling 250 Hz square wave. DACs normally filter all but the primary tone converting it to a single tone as you see in the DAC measurement above. The 250 Hz toggle is playing with the rightmost bit of a 24-bit signal so should not be measurable. If it is, it shows that the device is sensitive to digital input stream. I know, a lot of technical concepts here but thought I explain it. If you don't understand it, stay with me and we can still make sense out of the data.

As a baseline, I thought we should see how good the S/PDIF digital output is out of the APx555 Audio Analyzer. Here is its jitter spectrum:
Audio Precision APx555 Jitter Measurement with cable.png


The spectrum is mostly noise-like but we have a few spikes are that correlated with our 12 kHz square wave causing spikes in jitter at some multiples of it. Worst case amplitude is only 8 picoseconds though so don't get excited over this being an audible issue. :)

Note again that what is being measured is a digital signal. No DAC is in the loop. The APx555 captures a digital pulse train and extract the variations out of it (jitter).

Next, let's analyze what is coming out of the Matrix Audio S/PDIF output when being driven naturally by the computer using its USB port for data and power:

Matrix Audio X-SPDIF 2 USB to SPDIF and I2S Converter Jittter Spectrum Measurements.png


We now have a lot more spikes there as compared to our APx555 which is comforting to me seeing how the APx555 retails for $28,000. :) There are finely spaced spikes around the larger spikes which I am assuming are the components of the 250 Hz sub-square wave in j-test. It could also be mains related power supply spikes.

Now let's juice up the X SPDIF-2 with Wyred4Sound PS-1 power supply. You would think it would make things better, right?
Matrix Audio X-SPDIF 2 USB to SPDIF and I2S Converter Jittter Spectrum Wyred4Sound PS-1 Extern...png


Wrong! If you look to the left, we see that the low frequency spikes have increased and with it, we now have a lot more of them further down the spectrum.

Reason is simple: yet another power supply, yet another source of ground loop, mains leakage. That low level "hum" rides inside the X SPDIF-2 converter, modulating clocks and such. But again, we are talking very small signals here with worst case being 18 picoseconds. So don't go asking, "how come I don't hear it." You are not going to hear, but we sure as heck can measure it.

Let's now compare the Matrix Audio X SPDIF-2 to our budget recommendation DAC and USB to S/PDIF converter, the Topping D10 (USD $90 including shipping):
Matrix Audio X-SPDIF 2 USB to SPDIF and I2S Converter Jittter Spectrum Wyred4Sound USB Power c...png


Lot more spikes are visible now and some peaks up to 40 picoseconds or so. This may seem to be bad thing but it might be reverse of that: the peaks are so close to each other that may resemble noise than individually identifiable spikes. But again, all of this is at such low levels that I don't think either will be audible. Still, the Matrix Audio X SPDIF-2 shows that it can technically do better.

Now let's have a real-life scenario driving the Topping D50 DAC. To start, let's compare Audio Precision APx555 driving it versus Matrix Audio X SPDIF-2 converter:

Matrix Audio X-SPDIF 2 USB to SPDIF and I2S Converter Jittter Spectrum from Topping D50 DAC Me...png


Here we see a series of spikes with the APx555 analyzer which are clearly created due to mains frequencies. The Topping D50 is powered by yet another switching power supply (one from a Samsung mobile phone) so naturally we see mains leakage there. The Matrix Audio X SPDIF-2 doesn't create nearly as much of that by itself. But power it with a PS-1 external power supply and it inherits the same problem:

Matrix Audio X-SPDIF 2 USB to SPDIF and I2S Converter Jittter Spectrum from Topping D50 DAC Ex...png


Finally, let's drive the Topping D50 with D10 as a USB to S/PDIF converter:
Matrix Audio X-SPDIF 2 USB to SPDIF and I2S Converter Jittter Spectrum from Topping D50 DAC co...png


So we get our power supply noise spikes again although probably a bit more elevated than when APx555 was driving the D50.

Conclusions
It is so easy to imagine scenarios for improvement audio performance based on what "makes sense" to one's mind. Unfortunately these are complex technologies and don't yield themselves to lay assumptions. To wit, we easily saw that the addition of Wyred4Sound PS-1 power supply did nothing for baseline performance and just served to create extra ground loops. Seeing how the the PS-1 lacks any regulator certification, it is easy to say don't spend your money there.

As for Matrix Audio X SPDIF-2 converter, it seems to do its job. Can't match the quality of my audio analyzer but in real life with a DAC, it actually seemed to do better (this may very well be a DAC thing). So from engineering point of view, assuming you don't add an external power supply to it, I don't have any issue with someone purchasing it. The price is actually "reasonable" give how much custom designed logic is in there.

Personally I would just use the Topping D10. It has a nice display telling you the sample rate, has a bonus DAC included for one fourth the price of Matrix Audio X SPDIF-2 converter.

As for I²S, we got bamboozled by Wyred4Sound DAC-2V-SE being the dominant noise and jitter source. I will have to do the comparison at a later date when I have a more performant DAC.

Finally, please note that we are seriously "measurebating here." I have to zoom in so much into performance of these signals that breathing on them may change things. So do not worry about any audibility issues here. Use USB if that comes with your DAC and be done with it. Don't string all these things together thinking you are doing something useful.

------------
As always, questions, comments, recommendations, etc. are welcome.

It is snowing here for the first time this year. I need some money to buy snowshoes. Yes, it is only one inch of snow but you don't want me to fall and get hurt, do you? So please consider donating money using:
Patreon: https://www.patreon.com/audiosciencereview), or
upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
 
Last edited:

VintageFlanker

Major Contributor
Forum Donor
Joined
Sep 20, 2018
Messages
4,940
Likes
19,692
Location
Paris
Thanks @amirm.

The X-SPDIF 2 costs USD $279 but the company is selling for reduced cost of USD $341.10. Some complicated math must have gone into that computation to come up with that extra 10 cents in the price! :)

For sale at 369€ here in France... I'm still thinking this is a lot of money for a digital interface "only".

Speaking of Matrix Audio, I would be curious to see the X-Sabre Pro DAC measured. :)
 
Last edited:

graz_lag

Major Contributor
Joined
Oct 13, 2018
Messages
1,296
Likes
1,583
Location
Le Mans, France
Currently, I am testing the Singxer F-1 interface, which I bought for $/€100 second hand in pristine condition.
I'm very satisfied of that purchase so far.
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,395
Location
Seattle Area
Currently, I am testing the Singxer F-1 interface, which I bought for $/€100 second hand in pristine condition.
I'm very satisfied of that purchase so far.
That's pretty cheap. A lot of people use that one so would be good to test it at some point.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,580
Likes
38,281
Location
Gold Coast, Queensland, Australia
The X-SPDIF 2 costs USD $379 but the company is selling for reduced cost of USD $341.10. Some complicated math must have gone into that computation to come up with that extra 10 cents in the price!

10% off. :)
 

graz_lag

Major Contributor
Joined
Oct 13, 2018
Messages
1,296
Likes
1,583
Location
Le Mans, France
It looks there is a growing market for these high end standalone USB interfaces, many buy these to feed DACs that certainly are already engineered with nice USB inputs but still ...
Singxer is selling quite well of their highest model, the SU-1, priced @ $360 or so ...
The lobbying from Xmos surely counts a lot in this process ... They can sell two chips per end user : one in the DAC USB infeed, the other in the interface ...
Clever, simple, effective approach !
 

jj_coffee

Member
Joined
Nov 25, 2018
Messages
10
Likes
12
There was a time when I was searching high and low for a reliable, high quality spdif to USB converter. I can only vaguely remember why I was searching for one, but I think one reason was that USB driver support for legacy products used to be a big concern when we didn't have as much class compliant chipsets out there, and another was probably that USB input among high-end gear used to be rare. But whatever the reason, I presumed that such a converter would be quite straightforward, and remember being baffled at the lack of options out there.

As to why such a device might be necessary today, I think it's worth pointing out that not all USB implementations are equal, and it can be nice to be able to remove it from the equation when choosing a DAC (in case you missed it, there were some interesting comments in this thread regarding USB vs SPDIF differences here: https://www.audiosciencereview.com/...ments-of-holoaudio-xeme2ve-usb-to-spdif.6386/). Also, a good USB to SPDIF interface opens up options for going with an older DAC, which could be a benefit in some instances.

As such, I'm happy to know that Amir's testing is highlighting some great options at various price points :)
 

soundwave76

Addicted to Fun and Learning
Forum Donor
Joined
Dec 28, 2018
Messages
730
Likes
1,365
Location
Finland
Thanks for the review! I am using the previous version of this product, without the I²S interface, to convert USB->AES to feed a pair of Genelec 8331.

One question @amirm: the Matrix driver software offers 16 and 24 bits for output format. If I use 24 bits, the Matrix upsamples what I am listening, if I've understood correctly. Tthe Genelecs apparently also upsample the incoming digital signal to 24 bits before the DSP calculations and DA conversion. Now, should I use 16 or 24 bits in the Matrix driver, when listening mostly to Tidal Hifi?
 

gvl

Major Contributor
Joined
Mar 16, 2018
Messages
3,427
Likes
3,981
Location
SoCal
Matrix doesn't upsample anything, it is a pass-through device. Tidal HiFi has MQA albums that get unfolded to 24 bits, so 24 bit is the setting you want to use.
 

gvl

Major Contributor
Joined
Mar 16, 2018
Messages
3,427
Likes
3,981
Location
SoCal
XMOS driver bit-depth setting on the Format tab simply controls how the device is presented to Windows applications. It is up to application to decide what to do with data (dither or simply truncate) if the data is wider than what the device advertises it can accept. Iirc Tidal will simply not do the MQA unfold when outputting to a 16 bit device.
 

graz_lag

Major Contributor
Joined
Oct 13, 2018
Messages
1,296
Likes
1,583
Location
Le Mans, France
@gvl
With my current Singxer F-1, Linux Alsa let me control the volume of each of the two channels, so it interacts with the XMOS.
I was surprised after verifying that the other day, this is my fifth interface since 2012, I never seen the same with the other four ...
 

gvl

Major Contributor
Joined
Mar 16, 2018
Messages
3,427
Likes
3,981
Location
SoCal
@gvl
With my current Singxer F-1, Linux Alsa let me control the volume of each of the two channels, so it interacts with the XMOS.
I was surprised after verifying that the other day, this is my fifth interface since 2012, I never seen the same with the other four ...

Are you sure Alsa doesn't simply alter volume in software before sending data to the Singxer?
 

hege

Senior Member
Forum Donor
Joined
Dec 25, 2018
Messages
466
Likes
819
Location
Finland
SU-1 driver controls volume from driver/XMOS in Windows too, driver takes over the Win10 volume control, even when playing Wasapi/ASIO. You can see the driver sliders going up down. I was always curious about this, does the driver/XMOS dither properly?? Something I'd like to see tested.
 
Top Bottom