Innuos Zenith ripper/streamer

[innuos]

Here's some maths for the budget conscious SQ freak....

sT $1455
Don't bother with a LPSU for the sT, it really will not make much difference.
Drop the SOtM, use an ultraRendu = $875
LPS-1 for ultraRendu = $395
Mutec MC3+USB = $1160

Total = $3885, a little cheaper than the Zenith

Now for some speculation.....   I reckon that the above would offer significantly better sound quality into a Devialet Expert versus the Zenith.  I have not listened to the Zenith, so this IS speculation, but read back though the thread, it does not look like the Innous has done that much to optimise the USB output, add this to reports that even top notch USB sources such as the Aurender N10 sound better via AES/EBU, I am sure you get the idea.  Of course, if someone ever gets to actually try the Innous in a A/B test with something equivalent, maybe the results will prove different.

Hi,

Just a few comments with regards to the topic above. The USB output on the Zenith can be optimised in many ways that may not be as obvious as adding a separate USB board. 

First, the architecture of the USB Controller makes a difference. Typical boards have a USB controller at the CPU level and another at the motherboard level. When we compared boards to use with the Zenith, boards with two-level controllers always shown higher levels of EMI on the USB line. The motherboard on the Zenith only has a single USB controller directly on the CPU and the CPU alone has it's own separate linear power supply. If you have ever heard the difference in sound quality between a Zen (which uses a single LPSU) against the Zenith (with 3 LPSU) you'd know right away there is a significant difference. This is what differentiates the Zen and the Zenith the most, more so than using SSDs.

Second, we've worked on reducing vibration on the electronics of the Zenith. Whilst the clocks on a Zenith are not OCXO 10ppb, this vibration control and the LPSU do make all clocks on the motherboard perform their best and there's no cabling to contend with as when using external boards. 

Third, don't dismiss the difference the software and firmware can make to the sound of the system. The memory playback alone on the Zenith makes a very audible difference.

I wouldn't generally agree the linear PSU on a SonicTransport i5 wouldn't make a difference on the setup you describe. This would pretty much imply any noisy NAS would do the job just as well. That has not been our experience. In fact, a simple test we demonstrate all the time is testing a Zenith sending music via UPnP to a streamer in two scenarios:

1. The Zenith and the streamer both connected to a regular router/switch. 
2. The Zenith connected to the router/switch and the streamer connected to the streamer port of the Zenith

The same cables are used throughout. It's very audible the difference between both, particularly the size of the soundstage. Some people are really puzzled of how much a difference it can make. It just shows that ethernet cables can transmit EMI just like other cables, despite their construction. It doesn't affect the bits being transmitted but it does, directly or indirectly, affect the electronics on the other side. It seems even a dCS Network Bridge benefits from it based on some tests - you can read it at Head-Fi.

In summary, my recommendation would be: do not ignore what happens upstream of the streamer (pun not intended).

Hope this helps.

Nuno