Is a USB stick directly into ones streamer (renderer) better than streaming from a PC/NAS?


I have a cambridge Audio 851N which has a couple of USB slots, one of which is occupied by a small 32Gb USB 'stick' with some of my music on it.

I see a LOT, a huge amount of discussions going on re streamers/computers/NAS etc etc, and some spend thousands on these items. A LOT of money can be spent with expensive LPS and specially built computers, some also costing thousands, but WHY? Why, when one can dispense with the PC/NAS, rip off switches and their cables etc etc altogether and listen directly from a USB stick attached to a streamer, so no PC etc is required at all (for music playback).

Am I missing something here? Surely a USB memory stick (drive, whatever) costs almost nothing, as is connected directly to my streamer. I've no need to stream from my (expensive) PC or my NAS, with all the pitfalls that that entails. All I need to be powered up is the 851N and my power amp connected to my Maggie 1.7i's - bliss.

I don't use Tidal/Spotify or any of the other expensive sources, nor have any need for the rediculously over priced (IMO, I've not tried it) programs like Roon, as the USB stick does it all.

I've compared direct streaming from the USB stick with NAS and PC listening, both using 'good' cables and power supplies, and the direct USB input beats them both.

128x128daveteauk

And that is quite logical if you consider that a network cable is included in a digital chain. A chain that relies on clean energy to keep everything clocked tight. A chain that works with noise-levels of -140dB. A chain where small deviations are immediately audible in staging, focus and smoothness.

No DACs can do -140db noise floor, and if they could there is not a person who could hear it .

 

What we observe almost immediately is that the version with the shielding fixed on both sides focuses better. The version with shielding on one side seems to play a bit larger, but that is not true: the effects in the song Perfect Life by Steven wilson are placed at the same spot in the room, but are more tightly framed with the cable where the shielding is fixed on both sides.

 

So shockingly low levels of noise caused things to "play a bit larger", or are "more tightly framed" ... and better when both ends are connected to shield. If this person had the first clue how our hearing works, or the electronics they are working with, they would realize how silly this is. Grounding the shield at both end is to reduce EMI susceptibility primarily in noisy commercial environments with high speed transfer. However, if you are doing analog electronics, it is  great way to create a ground loop. BRILLIANT!  Let's take the fully isolated nature of Ethernet and introduce the potential for a ground loop then call it better?

p.s. The indicated pretty clearly I thought that every test was sighted. Given their lack of knowledge of how things work, they had already prepped themselves for the answer.

@akg_ca - I agree this can force one 'down the rabbit hole', but that is the reason for my original post, to get people to go down it, and explain their reasoning. Also, your idea that there's no 'noise' before the DAC, is wrong. We're not talking of musical noise here, but noise genereated by the device itself, the switch eg.

The link you've supplied from the network engineer, is, basically, irrelevant, as he's talking of data transmission and it's integrity along the way, which is not the issue here. He also doesn't seem to understand the issue we're talking about. He says that it's ridiculous to suggest that this can colour the audio signals - THAT statement proves he's not on the same page as most on here, as no-one is talking of audio signals being coloured in a switch or ethernaet cable. nIt's the JITTER that's produced which is the issue, I believe.

Do you realise that your post has several duplications within it?

I agree with what cindyment is saying, I'm afraid akg, although we're not talking of analogue electronics here. Everything in this post is iin the digital domaine

Whether the digital Data is sourced from USB or ethernet the clock for the DAC is the couple from the data source. There is no evidence that the incoming data connection causes jitter on the output. That would be very very difficult to occur on ethernet and while theoretically possible due to electrical noise on USB there is little evidence that actually happens. For USB isolators are relatively inexpensive. The noise generated internally in the DAC is going to exceed external noise generally by a significant amount. The edge speeds in modern logic are quick meaning that small amounts of noise generate insignificant amounts of jitter.

@cindyment - It's not jitter on the output of the DAC I'm wondering about, it's on the input, along with electrical noise, which is/may be created by the ethernet links - switch/cables/PS.

USB 'isolators', not inexpensive IMO, as they vary, in rip off terms, from £50 to £250, and have been proven, with in depth measurments, to do absolutely nothing to anything passing through them. Legalised stealing is my opinion of them, as are 'special' mains input fuses.

There's been a lot of discussion in this thread, but little in a way of answering why people spend so so much on the equipment mentioned, when a cheap USB stick plugged into your device gives as good, or better SQ.

Forget your NAS (for music), forget your PC (for music), expensive Audiophle PC or not, forget your expensive switch and associated expensive LPS's to power each of those things, and not forgetting to forget your expensive ethernet cable. You can dispense with ALL that noise/jitter creating expensive hardware and just spend c£10 on a USB stick - job done. There, I've just saved you (and me) £1000>£10,000. No-one has said anything to convice me, or suggest otherwise.

I believe this thread is finished now.

Thanks for all your contributions and opinions.