Stello U3: async USB to coax & AES/EBU transport <Impressions updated> - Page 6

Impedance of what? the cable? the connectors? the PCB of the DAC board that goes up to the DAC chip inputs? and how is the USB signal encoded/transmitted? m2tech's tech support always advises to use NEC chipsets when ppl have problems with AMD controllers or so, and look at those HDMI eye jitter patterns: http://translate.google.com/translate?sl=fr&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&layout=2&eotf=1&u=http%3A%2F%2Fwww.erenumerique.fr%2Fqualite_des_connexions_dvi_nvidia_et_ati_assurent_ils_-art-820-11.html

Most of these graphic cards are out of specs! Luckily HDMI has an embedded ECC(unlike USB and S/PDIF) so apart from the fact that long cables will work in some graphic card/display combinations and not in others, the impact of those sloppy TMDS encoders is rather limited. Anyway, the HDMI specs call for optical fiber for >5m cable because the electrical attenuation gets too high.

I would need to find the URL again, but an audiophile myths debunker blog was showing that when everything is out of specs in a S/PDIF rig(and from what they said, it pretty much always is ), it's completely normal that every cable/adapter would sound different. I can certainly attest that I believe hearing SQ differences between my Intel USB 2.0 controller and a PCI-E USB 3.0 NEC chip with the Firestone Bravo....which one was better is hard to say. I guess this hobby boils down to finding a sonic color you will enjoy on as permanent of a basis as possible, because EVERYTHING colors the sound IME.

And I really don't think that all S/PDIF receivers provide the same subjective sound either, you can look at the reviews on those DIR9001/WM8805/CS8416 boards for the Audio-GD gear...I also read comparative reviews of DIR9001/CS8414 and I would definitely agreed with them based on my own experiments w/ the old Firestone Spitfire and its newer MK2 counterpart. It would only make sense that all those receivers would react differently to different USB transports

My fav is still WM880x but it's merciless and you'd better have a beefy output stage to compensate for its very clear sound. The Asus Essence One is using the AKM4113, it's hard to characterize its sound as it's mandatorily reclocked by the onboard Sharc DSP anyway...but I really enjoy how it doesn't make nasty glitches when you switch sample rates

I was having more thoughts about cables, the Lacie FlatCable has the great feature that the power and data lines are as discrete as can get(even more so than on the WW)...I guess we could cut the cable in two and wrap each side into ERS/aluminium foil+black electrician tape? It looks so ugly in sky blue anyway, couldn't get any worse

Chomerics have EMI shielding wrap, but they advise 50% overlap and only give away 5" samples: http://www.chomerics.com/products/shieldwrap.htm

Originally Posted by BrainFood 

Mine just arrived and plugging it in now.  If it compares well to the $300+ cables, it will be a bargain.  It's replacing a very basic one though.

TBH, I didn't like it on the Firestone Bravo(+Supplier PSU), but neither did I like the WW Starlight and Ultraviolet...my best sounding cable on this combo was a freebie that came w/ an ebay DAC, the DAC sounded like **** but the cable was most excellent on the Bravo(far less colored SS). I think there are too many factors into play:

-accepting to get robbed by greedy OEM sellers who buy the cables for a few bucks in Shenzhen, make stellar markups and bribe all (UK) magazines and websites to sing along that it's teh best cable EVR and that you need it so badly

-USB chipset tolerances to jitter, rise times yada yada...they can't possibly be all equal. You get what you pay for IMO...I see the TE7022L chip as a slightly more advanced Realtek, when the XMOS seems far more advanced, with two discrete clocks, truly async operations etc.

-whether the USB power is in use, it wasn't on the Bravo and it very much is on the U3

so we got the usual infamous synergy matter between cable inner and outer interferences/USB chipset tolerances/etc etc, the subjective bias and so on...all I know is that this $6 Lacie cable provides a very clear soundstage and a highly refined subjective SQ to my ears on the U3. I just find it sad when ppl spend twice more on the coax cable than on their hiface, or up to 50% of the price of their V-DAC in a WW cable.

But it's the same story w/ everything, how much does a LCD-3 cost to manufacture? 3K on an ortho phone sounds just as silly as $300 on a mere silver plated copper cable

moonboy told me a long time ago that he would prefer a cheap phone off a killer source/amp combo than the opposite, and I tend to agree. Spend the money on your source/amp more than on overpriced cables or phones.

44.1-only is really ghetto...no 48kHz movies, no HD audio, and they don't say anything about 24bit either for an improved SQ of DSP(EQ/crossfeed) in your media player...back to the stone age really.

Oh, I've got nothing to sell here...if you don't care for either movies, DSP(crossfeed is mandatory for me on headphones, otherwise my brain gets extremely upset) or HD audio...then yeah, I guess you should be fine.

Saying that X transport is better than Y out of its context is meaningless IMHO...it depends on the S/PDIF receiver chip built into the DAC device(they're far from all equal when it comes to clock recovery:CS8416 in 192kHz mode has more than half a nanosecond of jitter at 44.1kHz ), whether it's being reclocked, isolated, how long the cable was, how long the I2S path to the DAC chip is, the quality of the PSU, etc etc.. There's no hard rule, and w/ a male/male adapter(such as on the AP2, but you can also use them on any other transport), a good reclocker, etc etc...the diff between all those transports can get thinner and thinner. And at this rate, 44.1kHz will still be a major bottleneck for the ppl who don't only listen to CDDA. Either way, the only way to know which of these transports is "better" on YOUR own rig is to try them all...I'm sure you could obtain them from companies w/ return policies. My experience is that no fancy graph or cyber-hearsay can be solely trusted when it comes to synergy matters and personal preference.

The nice thing about the U3 is that it's USB Audio Class 2.0 compliant, hopefully W8 will support it natively and there won't be any need for proprietary drivers whatsoever...this is as future proof as you can get. You also get native ASIO from the get go

BTW, just for the record, someone on a local forum compared the Lacie Flatcable to the WW UltraViolet on an EVO+external battery+jplay4 and claimed to hear a very obvious SQ improvement...so indeed, I don't see any good reason to make WW even richer than they already are by selling copper at the price of gold

And I rest my case that wrapping its two data/power portions into discrete ESR paper/aluminium foil could improve its shielding even further...another cheap experiment to conduct down the line ^^