What is the added value of a streamer over a networked dedicated Computer


Hi

I see lots of sales pitches for streamers as digital sources, and plenty on this site advocating them. I get that they're a purpose-built user interface but, apart from that convenience, including a visual display on the device, (i) do they really deliver better hi-fi sound as a source over a well set up computer dedicated to hi-fibreoriduction (ii) if so, why?

Here's some background to my question(s). I currently use a dedicated Mac Mini with SSD (headlessly) and Audirvana Plus software through a USB DAC. I tend to listen to digital files on external drives (wired connections). Some are high Definition eg Flac, some are aiff ripped from my extensive CD collection. Currently I only tend to use Spotify etc to test if I like music and invest in actual downloads of the music I like.  In day to day use the Mac Mini/Audirvana Plus (virtual) player is controlled using its remote app on an iPad on the same Network. If I wanted I could add high quality online streaming from, eg, Tidal. Whilst that would expand the breadth of music I have immediate access to, it seems to me to add another potential source of interruption/corruption of data flow. The Audirvana software overrides/bypasses detrimental computer audio elements and processes keeping the data path simple and dedicated to hifi audio replay.

So what, sound quality-wise, would a standalone streamer device using NAS or other drive storage and/or online web connection bring to the party? It seems to me it's just a digital device containing effectively the components of a computer with a button (or remote) interface. I understand the old argument that it's dedicated and not doing other things simultaneously and that computers are traditionally electrically noisy environments but I'm currently sceptical that with a dedicated computer, not being used for other purposes, and running a virtual device like Audirvana Plus which effectively switches off internal functions which might compromise sound, this is a real problem. Also it seems that a "dedicated streamer" contains many elements which are effectively computing elements. Note that I have no industry connection or monetary interest from Audirvana or Apple.

128x128napoleoninrags16

@herman

The timing of the Streamer vs the DAC depends on the connection used. If you use the USB connection, then the DAC retimes the bitstream. If you use the S/PDIF or AES connection then the DAC uses the incoming timing. This is why it is important to try both.

My theory is that you should use the better component’s ability to time. I tried both the AES, S/PDIF, USB from my Aurender W20SE and Audio Research CD9SE DAC and found them all to be really… really similar, with the AES slightly better than the S/PDIF. But, I think it depends on your streamer and DAC.

The timing of the Streamer vs the DAC depends on the connection used. If you use the USB connection, then the DAC retimes the bitstream. If you use the S/PDIF or AES connection then the DAC uses the incoming timing. This is why it is important to try both.

agreed. When I'm talking "streamer" I was talking about the device that takes the digital data from the internet and then sends it on to the DAC via ethernet or wifi. Like a Roon Core or similar. If this is better termed a server then I have taken things off track, 

if others were considering it to be the endpoint that takes this stream and converts to a data stream for the DAC via AES or other input other than USB then I apologize for confusing things. With AES or SpDIF I do agree that the streamer clock is therefore very important. 

 

Thanks for all of these (very different) responses. 
When it comes to hi-fi, I do have much sympathy with the "subjectivist" approach as I think we still struggle to match what can be heard with objective measures or electrical theory. However I still regard digital and electrical understanding and measurement as important adjunctive tools, in part because they are necessary for designing kit, and also as a protection against "snake oil."

The responses above reflect the objectivist/subjectivist angles and I welcome hearing both. Ultimately over the last 40 years of owning and upgrading hi-fi I have always let my ears decide when purchasing gear as neither I nor most people are in a position to actually measure the digital data, inc timing, at various points in the signal pathway. As has been pointed out by one commenter, the "experience" of others who have invested in gear (whether digital or analogue) is likely psychologically biased toward justifying added expense, especially after the event. Yet no doubt many are quite right that their newly bought piece of kit has improved their system. 
 

It has become clear from the comments that some terms (like streamer/streaming) get used loosely and I am likely as guilty of this as others. It is also clear. That the implementation of computers varies widely, inc from my own usage - not all computers are the same, some are multitasking more than others, and there is a huge difference between outputting analogue through a headphone minijack and digital output, and indeed between the various options to output digital signal to a DAC.
 

So... as suggested  I need to provide a little more detail of my system and I also should explain further its current usage. I'll do that shortly in a subsequent comment.

Continuing from above, I already said in my OP that I don’t really listen to internet streaming (eg Spotify, or higher resolution internet sources) for serious listening. I appreciate the massive expansion in listening choice they offer but still prefer to have local "hard copy" and listen to that rather than be at the mercy of internet connection for uninterrupted/uncorrupted high quality audio. Perhaps my loose use of the word streamer in my question has given a different impression.

@Fuzztone re "If you had wisely spent on music reproduction equipment all of the money you handed to Apple, you would have a fine system. Details? I don’t know anyone that tricked out a Mac enough to stop there. Most stop at an Amazon Echo before they even get there."
I believe I do already have a fine system, thanks, and I gave no money directly to Apple since I bought the Mac Mini used. I did buy it as a way to get into SSD-based high resolution music files as well as convenient backup of my CD collection. That said, I remain interested in cost-effective upgrades to my system but am still unlikely to prioritise internet-streamed audio over locally stored music (or music via wifi) for serious listening in my dedicated listening room. I do send via Airplay to devices in other rooms when I’m in "party" mode but that’s not for serious listening. You (and others) asked for details. The system components are ageing but I have found that newer is often not necessarily better; furthermore earliest iterations of new technology direction have often been poor alternatives to traditional sources. My own view (open to alternative suggestions) is that the weak links in the digital chain are the 2 DACs. So my question about streamers is because I need to consider suitable interfaces/connections for digital sources/servers when upgrading the DAC, and also that I’m aware that some streamers come with an onboard DAC. I also recognise that internet-streamed audio will likely continue to improve in the future.

Analogue source: Audiomeca Romance turntable with Romeo uni-pivot arm, Van den Hul Condor Gold coil (LO) MC with suspension adjusted by vdH himself (begrudgingly!) for uni-pivot arm

Digital source 1:
Audiomeca Mephisto CD transport connected via BNC to Audiomeca Ambrosia DAC. This DAC was high end at the time of purchase but doesn’t handle high resolution audio. Nonetheless, over the years since I have auditioned other DACs and significant expense would have been required to improve on it significantly. One DAC I did demo at home which ran rings around it (but which I couldn’t afford) was a dCS Paganini.

Pre-amp: BAT VK30SE (Tube) with onboard phono stage
Power Amps: Audion Black Shadow 845 monoblocks (modified)

Loudspeakers: Coincident Super Eclipse III (92dB/14Ohm)

In addition to the above, as a weekend warrior musician, I was doing some home recording via MacBook Pro laptop and had already purchased an Edirol (Roland) multichannel USB interface for that purpose, capable of handling up to 24bit/192kHz data. This is essentially an ADC and DAC in one box with optical and MIDI in and out as well as USB. It’s connected to the preamp with audio note silver cables. When exploring high res digital audio files via the hifi, it made sense to me to begin with this device I already had. It also made sense to me at that time, if wanting to listen to such files using the hi-fi, to move from my laptop to a device not being used for other purposes and to keep those files along with my ripped (aiff) CDs on dedicated drives. Having done so, I have found the sound quality of the Audirvana/MacMini/Edirol combination remarkably good, given the price, such that I rarely listen to the original CDs. The situation is further complicated by the fact that the Edirol requires a driver and the drivers weren’t updated beyond the Mavericks OS, so the Mac Mini runs on that legacy OS and I can’t update the Mac Mini with later OS. However, since it’s dedicated to hifi listening and recording (not simultaneously), that’s not been a problem for me.

This is the context of my question on streamers, and I remain primarily concerned about the quality of handling of the music files on the drives (both the mac mini’s onboard SSD, and a couple of external drives with wired connections to the Mac Mini), as opposed to music which may be streamed from the internet. I am not expert in the technological details of rendering but was under the impression from what I read on the PS Audio website that, specifically with regard to timing issues, rendering by the computer was essentially (ahem) rendered irrelevant when playing music files from its SSD using a USB interface. Happy to be educated further by those with real expertise in this area.

 

 

Timing is usually a non-issue with USB since the advent of asynchronous mode. For a reference, see this document by XMOS, maker of USB audio interfaces used by manufacturers.  The key takeaway is on p.4:

Asynchronous mode enables external clock sources to be used as the master, or a low-jitter clock in the device.

Interfaces such as SPDIF (coax or optical) might exhibit higher jitter since clock recovery is done using the incoming digital signal though a clever scheme called biphase mark coding. However, considerable jitter needs to be introduced before audible effects can be heard.

 

Given that your DAC has multiple digital inputs and if you like the sound, I don't see a compelling reason why you couldn't use one of the other inputs for a streamer.