Matrix Audio X-SABRE 3 Measurements (DAC & Streamer)

[VintageFlanker]

Thread would be a good idea,go on and I'll follow with measurements (and others,I'm sure) in time be.

I already hear subjectivists celebrating from here ! (lot of religion about the "phase" and how it makes night and day differences)

I hope my questions were not perceived as residual nitpick noise... not even below -150dB noise floor.

Hahaha, of course, not !

I am wondering how much you can gather from a crowd-funding effort to purchase yourself an AP rig!

Not much, I'm afraid. I would love to have one on hands (assuming that would next require months to learn how to use it (barely) properly). Truth is, I even sought about how much would it cost to rent one. Well, at least 500€/month for an APx525. I gave up.

 

[pseudoid]

...I would love to have one on hands..

Isn't there a software-based version of AP which requires a PC plug-in card (@10dB cheaper)?

 

[xaxxon]

Unfortunately I can't, I sent it back a few days ago...


View attachment 235726


Yep, my sincere thanks to them! Not to mention they allowed me to keep the unit waaaay longer than I initially planned.


Back to the X-Sabre 3, the insane level of its assembly continues inside. Since there was no way I dared to open the unit for a teardown, I just point to other's pictures:

From HI-FI News:

View attachment 235727

From @MatrixAudio :
View attachment 235731

If @amirm wants to test anything more on this DAC he's welcome to borrow mine. I'm local in seattle.

 

[Goodfellas27]

If @amirm wants to test anything more on this DAC he's welcome to borrow mine. I'm local in seattle.

Awesome
Lets make it happen

 

[Rja4000]

Isn't there a software-based version of AP which requires a PC plug-in card (@10dB cheaper)?

Yes, the software may be purchased separately, and used with some Audio interfaces.

But cost is, unfortunately, still nowhere near "affordable": last time I checked, for a bundle with the options required to run Amir's measurements set, it was reaching $8000.

Too high for me or any amateur, I'm afraid.
And you still don't have the low distortion analog generator and variable frequency analog notch filter for this price.

 

[Rja4000]

Here is the lowest I could measure, at -60dB volume :​

REW FFT

​

From this, you should be able to deduct dynamic range, which is the SNR with an output signal at -60dB (so your SNR figure -assuming it's bandwidth limited) + the (exact) APU preamp gain.
You just need to check the BW and to measure the APU gain at 1kHz.

NB: "SNR", so not only "-Noise", but Signal level - Noise level

 

[VintageFlanker]

From this, you should be able to deduct dynamic range, which is the SNR with is an output signal at -60dB

The graph is the other way around: This is the output signal at -60dB... on DAC volume. But now that you're mentioning it, the results were quite close when running -60dBFS tone at full volume.

You just need to check the BW and to measure the APU gain at 1kHz.

Not sure I got it.

 

[Rja4000]

The graph is the other way around: This is the output signal at -60dB... on DAC volume. But now that you're mentioning it, the results were quite close when running -60dBFS tone at full volume.

The problem in your dynamic range figure is that you are measuring the dynamic range of your DAC + ADC.
Using the APU preamp, the ADC impact is removed.

The Audio Precision is doing the same: amplify first, then measure, then add the preamp gain.

 

[Rja4000]

You just need to check the BW and to measure the APU gain at 1kHz.

Dynamic range, if I'm not mistaken, is measured with a bandpass filter, probably 20Hz-20kHz.
You have to make sure that's what REW is using.

REW shows "N", which is -SNR + Signal level.

If SNR is, say, 71.4dB, and APU exact gain is 59.4dB (that's what I measured with mine), Dynamic range is 71.4+59.4 = 130.8dB
(Closer to the vendor specs, which says >134dB)

 

[VintageFlanker]

Dynamic range, if I'm not mistaken, is measured with a bandpass filter, probably 20Hz-20kHz.
You have to make sure that's what REW is using.

Oh, I'm using it already ! (44.1Khz, bandpass filter 20Hz-20kHz).

and APU exact gain is 59.4dB (that's what I measured with mine),

I guess we should expect sample-to-sample variations (?). My APU is a pre-production sample, BTW.

Dynamic range is 71.4+59.4 = 130.8dB

Makes sense.

The AP is doing the same: amplify first, then measure, then add the preamp gain.

...with an integrated SOTA DMM. Ça change la vie.

 

[Rja4000]

with an integrated SOTA DMM

Actually, for high gain like this, you'll get a more accurate gain measurement from your ADC than from a DMM (Even from my quite expensive Keighley DMM6500).

Just measure input level on your ADC before and after the APU preamp (with preamp connected, to avoid load impedance change)

 

[Rja4000]

Actually, for high gain like this, you'll get a more accurate gain measurement from your ADC than from a DMM (Even from my quite expensive Keighley DMM6500).

Just measure input level on your ADC before and after the APU preamp (with preamp connected, to avoid load impedance change)

With the Cosmos ADC, given its very low input impedance, I'm not sure this will work.

 

[Trell]

Contrary to previous generations, the circular screen is not monochrome anymore. It seems to be IPS, and is very tiny, while the resolution is good enough to have a proper reading at reasonable distances. Should we expect OLED at this price, in this niche market? I don't know, but overall brightness, contrasts and viewing angle are all decent for an audio product.

Thanks for the review!

As use of IPS for the screen I think that is currently a much better choice than OLED due to burn-in issues showing such static images for prolonged time.

 

[AudioSceptic]

Teriffic review! I don't know if the photos were stock or you handiwork, but they are superb.

Nice to see one of these premium priced products is also performant. It is a cool looking piece of gear which may justify its large tariff with some would be buyers.

It might be premium priced, but it's bargain basement compared with products from the established audio "High End".

 

[AdamG]

Another Home Run Review @VintageFlanker. Why do I feel like I am in a Discotheque? Your lighting choices certainly make your reviews stand out and nobody is going to mistake one of your reviews with anyone else’s. Let’s get this party started!!!

 

[DonR]

So now one must ask.... why Dave?

 

[VintageFlanker]

It might be premium priced, but it's bargain basement compared with products from the established audio "High End".

Sure. I really think that @MatrixAudio is aiming a different market than Topping/S.M.S.L. Contrary to the latter, their products are more reviewed by mainstream audiophile press (as HFN posted above) and come with next-level presentation (design, packaging, website, advertising). In short: the target seems to be classic audiophile with a decent amount of money to spend.

Your lighting choices certainly make your reviews stand out and nobody is going to mistake one of your reviews with anyone else’s. Let’s get this party started!!!

Hahahaha. Don't forget to set the forum in Dark Mode for a better experience!

 

[Arnas]

The Linn Klimax DSM costs £30k and the Naim 555 costs £20k. What could they possibly improve for those insane prices?

Someones wallet.

 

[Kevinfc]

Nice review, beautiful photography

 

[Arnas]

Matrix folks, please release ess 3039 version of this whitout streamer and mqa for lower cost.
You can have 2 different versions of this with one being cheaper whitout all of the above.
I feel like you folks straying too far away to from being affordable... Not long ago Matrix x sabre costed 1600euro.