Sonore MicroRendu Hardware Teardown and Review

[manisandher]

This is how I would have described the differences subjectively:

#2 vs. #1: flatter/compressed, less dynamic, less focused, grayer, duller, just less interesting
#3 vs. #1: slightly sharper, slightly thinner, not quite as extended low down

But after Ray's analysis, I've got to admit that the tracks should sound identical (with a few clicks added from time to time to #3). And perhaps they do? Perhaps I've just been 'hearing' these differences in my head? Weirdly, the differences that I was so sure I was hearing consistently across different systems are harder to discern now having seen Ray's analysis. I don't know whether to laugh or cry...

Mani.

 

[amirm]

But after Ray's analysis, I've got to admit that the tracks should sound identical (with a few clicks added from time to time to #3). And perhaps they do? Perhaps I've just been 'hearing' these differences in my head? Weirdly, the differences that I was so sure I was hearing consistently across different systems are harder to discern now having seen Ray's analysis. I don't know whether to laugh or cry...

I can't tell you how many times I have gone through the same experience. It is depressing at first to be sure. But longer term, there is good teaching in that with respect to what we hear and ways we test things.

 

[RayDunzl]

Perhaps I've just been 'hearing' these differences in my head?

That's entirely possible, especially if you know which one you are listening to.

Since you are comparing digital files, an ABX test should not be too difficult to perform.

 

[amirm]

Track 3 is about seven samples longer than track one and two.

Beginning is time aligned, very close to the end they are slightly off. A few glitches in four and a half minutes, otherwise, the tracks seem to be identical.

Zoomed in to see individual samples on tracks 2 and 3.

View attachment 3796

Central Scrutinizer strikes again!

 

[RayDunzl]

I don't know whether to laugh or cry...

Here's a popular video that is often referenced as addressing the subject... At least JJ's part of it, at around the 3:00 mark.

I may have to watch the whole thing myself, since I may not have.

 

[manisandher]

Thanks Ray. I just watched the whole thing - very interesting and thought-provoking...

Mani.

 

[RayDunzl]

#2 vs. #1: flatter/compressed, less dynamic, less focused, grayer, duller, just less interesting
#3 vs. #1: slightly sharper, slightly thinner, not quite as extended low down

So, you heard a difference.

Above, I think established the data is the same on 1 and 2, and all but the same on 3.

Since you can't hear data, it has to be decoded.

Recording and comparing the redigitized analog output of the DAC might show a surprise (or not).

 

[manisandher]

Yep, I'll definitely do this.

Mani.

 

[manisandher]

I might have to capture the analogue output from a DAC with an ADC...

Here we go! The path is as follows:

microRendu (no DSP in HQPlayer) or Phasure Mach II (no DSP in XXHighEnd) -> Chord 2Qute (USB input) -> Pass Labs X1 preamp (to convert Chord's SE outputs to balanced) -> Prism AD124 ADC (balanced inputs only) -> Tascam spdif input

(I'm not going to reveal which digital transport is which just yet, in case anyone wants to actually listen to the files to determine if there are any audible differences.)

1. Original file:

https://drive.google.com/open?id=0B0PU5LO5jVjfTUdrMGtRRkZuZkk

2. Digital transport A (microRendu or Mach II):

https://drive.google.com/open?id=0B0PU5LO5jVjfQ1JYdk9mVWtEbHM

3. Digital transport B (microRendu or Mach II):

https://drive.google.com/open?id=0B0PU5LO5jVjfMEF3eURHVkU1ZWM

[EDIT: All is now good (it was already!).]

Ray (or anyone else) if you could determine which of #2 or #3 most closely matches #1, that'd be great.

Mani.

 

[Thomas savage]

Central Scrutinizer strikes again!

We need to get some ASR mugs made so after Central Scrutinizer has broken all that someone holds true we can soften the blow by rewarding them with a mug!

'ASR Central Scrutinizer approved '

 

[manisandher]

... after Central Scrutinizer has broken all that someone holds true...

Ah... not so fast. Another analysis is showing that there are indeed differences between the 3 files I originally posted...

Mani.

 

[Thomas savage]

Ah... not so fast. Another analysis is showing that there are indeed differences between the 3 files I originally posted...

Mani.

You don't want a mug?

Seems what you expected to hear and did hear has been informed by factors outside of your auditory system. When you thought there was no difference you found it hard to hear a difference and vice versa..

So expectation has clearly informed the perception of what you have heard, regardless of any differences that may or may not be actually audible we have identified a susceptibility to expectation.

It's a nightmare , you hear it then attach a reason for this perceived difference. A physical reason that often informs all sorts of opinions for us and even may lead to us spending a fair amount of money when all along these assumptions we have made are unreliable for the formentioned reason .

Iv done plenty of this, I think most here have at some point.

 

[amirm]

Here we go! The path is as follows:

microRendu (no DSP in HQPlayer) or Phasure Mach II (no DSP in XXHighEnd) -> Chord 2Qute (USB input) -> Pass Labs X1 preamp (to convert Chord's SE outputs to balanced) -> Prism AD124 ADC (balanced inputs only) -> Tascam spdif input

(I'm not going to reveal which digital transport is which just yet, in case anyone wants to actually listen to the files to determine if there are any audible differences.)

1. Original file:

https://drive.google.com/open?id=0B0PU5LO5jVjfTUdrMGtRRkZuZkk

2. Digital transport A (microRendu or Mach II):

https://drive.google.com/open?id=0B0PU5LO5jVjfQ1JYdk9mVWtEbHM

3. Digital transport B (microRendu or Mach II):

https://drive.google.com/open?id=0B0PU5LO5jVjfMEF3eURHVkU1ZWM

[EDIT: All is now good (it was already!).]

Ray (or anyone else) if you could determine which of #2 or #3 most closely matches #1, that'd be great.

Mani.

Great test and thank you for keeping us honest by running it blind this way .

I tested all three files in frequency domain. Here are the graphs overlaid on top of each other:

The red graph is the original. The other two look like one graph in pink because they are smack on top of each other. The only difference is in low frequency between the original and the two versions through the DAC and that is due to slight roll off of your ADC/DAC in that region.

In other words, in frequency domain Sonore MicroRendu made no difference. Which to me says the device is ineffective.

I let Ray do the time domain sample analysis as he has far more patience than me .

 

[manisandher]

Amir, thanks for the spectra.

In other words, in frequency domain Sonore MicroRendu made no difference. Which to me says the device is ineffective.

But surely "making no difference" is an absolutely good thing, isn't it? Or did you mean compared to the W10 audio PC?

Mani.

 

[amirm]

But surely "making no difference" is an absolutely good thing, isn't it? Or did you mean compared to the W10 audio PC?

I meant it as compared to W10 audio/not having microRendu.

 

[RayDunzl]

Another analysis is showing that there are indeed differences between the 3 files I originally posted...

Can you elaborate?

Will look at the 'analog' files now...

 

[manisandher]

Can you elaborate?

He's still doing his analysis, but it looks like there are deviations from file #1 in both files #2 and #3 at the LSB level, i.e. around 96dB down. I'll share more (with his permission) once he's done.

Look forward to seeing what you make of the analog(ue) files...

Mani.

 

[RayDunzl]

it looks like there are deviations from file #1 in both files #2 and #3 at the LSB level

He's right. Amplification of the diff file between 1 and 2 shows it.

The difference result that looked enough like a straight line to "call it a day" for me, when viewed at sample level, and amplifed by 90.309dB (via two amplifications in Audacity which limits a single amplification to 50dB).

My first guess (only a guess) would be dither added in file 2.

Good catch.

 

[manisandher]

Dither was exactly what he thought too. But I had dither switched off in HQPlayer (applies to #2) and XXHighEnd doesn't apply any anyway (applies to #3). Interestingly, he found that in file #2, all the deviations were 'downwards' (as you've shown), but that in file #3 they were all 'upwards'. Seems a strange way of applying 'random' dither to me ;-)

Mani.

 

[RayDunzl]

I see negative direction in File #3, and it looks the same as above, without sitting here and trying to count the transitions.

As for dither being random, there are flavors of dither.

https://en.wikipedia.org/wiki/Dither#Different_types