Measurements: Raspberry Pi 3 as USB Audio Streamer

I wish there was a ‘Dislike’ button for voting. Maybe even 20 ‘Dislikes’ and you are suspended from the forum. I wonder how long you would last before suspension.

EDIT - I never noticed before posting, but he has already been banned. Well done to the moderators or whoever saw that the comments were beyond reasonable behaviour.

4 Likes

Yep. Only thing it lacks is library support which I hope will come in time. Really don’t like having to rely on having a phone and tablet to hand to play music. It’s the only thing from Squeezebox I miss.

Yeah, I noticed that too, but don’t understand what it means since he is still posting. What does “Banned” mean if you can still post?

I think those are historical posts left to preserve the sense of discussion threads…

1 Like

Ah, so he JUST got banned since that last post was 23 hours ago!

A more relevant experiment might be to test the same PI feeding the same DAC via USB and S/PDIF.

Thanks for that link. I was curious whatever happened to JH after Stereo Review went defunct.

In the 1960’s one Gordon Holt became dissatisfied with the laboratory style of audio reporting and the disconnect with how stuff actually sounded, and started Stereophile magazine, dedicated to the “subjective” style of reviewing. He called it “subjective”, but he attempted to develop a consistent vernacular for audio sound like “liquid”, “warm”, “detailed”, “glassy”, similar to what wine reviewers do. These days subjective reviewers write mostly about how their favorite rock albums make them feel, information that is useless for anything but entertainment. They typically hear 6 impossible things before breakfast, like night and day differences between AC line cords.

1 Like

I just did a test today with 2 PIs (3) feeding the same DAC (Marantz NA8005 -> Marantz PM8005 -> KEF LS50)

  • One PI motherboard only with USB out to DAC
  • One with a Hifiberry Digi+ board connected via Coax to DAC
    Identical power supplies.
    Grouped them together in a ROON zone so they played the same thing and I could switch on the Marantz from USB to Coax. Tested with Handel, Beethoven, London Grammar, Keith Jarret Trio, and a bunch of others.
    It’s probably me but I was not able to tell the difference…
6 Likes

My brother is messing around with this Streaming business and it has proven frustrating for him over the past few months.

He’s amassed a superb Hi-Fi system ye t has yet to find a suitable, consistent high quality sound- Streaming service; Tidal (with/without AudiRvana, Pure Music etc.) Roon etc.
A recent conversation with him revealed both some drop-out oddities (all correct settings) and that Roon are a truly professional organization.

He’s diligent (and desperate? lol) to figure it out and settle on something consistently excellent. From what I’ve heard and discovered, this “consistency” appears to be an issue with Streaming HiFi ?

From a guy (me lol) that took nearly 20-years before I found CDP’s to be musical and engaging, I suspect this “arena” is also experiencing a steep (although much quicker) learning curve ?

Rewinding ten years, it was believed tat computer audio would /should be a great (if not sound quality) library system. Yet, getting sound quality that compares to a decent CDP appears to be either hit-miss or inconsistent.

The sheer volume of discussions on this topic is astonishing; what once was envisioned to be super-easy and great-sounding (CA), in fact requires considerable knowledge/understanding and patience in order to deliver the expected “premium” sound.

I’ve decided to begin my “Streaming” journey with the acquisition of a Raspberry Pi 3B -using (eventually lol) the moode/MPD. OS. (I still can’t get moode/MPD up & running !)

I look forward to incorporating Roon once I’m “connected” and have gained some much-needed experience with this category of Hi-Fi.

I just realized the length of my reply here -but seeing the author above stating (accurately) " file based music is so easy, isn’t it " … got me thinking.

Any suggestions or recommendations to help out (ease-me-in) would be most welcome.

pj

Hmmm. Try exchanging ANY AC power cord that you may have lying around. For discerning the easiest, look for a chunky #12 gauge and some smaller #14 / #16 ones.
(Look for thicker insulated ones with a “rubbery” insulation, and ones thinner with a harder outer insulation).

Listen for low-frequency clarity/definition (bass boom -or articulation). Any music will do (a decent recording) but any genre. You will be surprised at what you discover. By no means am I suggesting running out and buying a commercial one, but distinctions in SQ can be both obvious, and remarkably and (inexplicably) profound. Yet, power cords are very component dependent.

It appears the (detachable) power cord is seen (by the device) as part the components internal power supply -and thus interacts differently. I suspect it may be some type of resistive (ohms) element between the cord and the units internal transformer (or SMPS) impedance characteristics.

pj

Well, it could conceivably make an audible difference. Timing of the bits and so forth. But if you’re not hearing it it’s probably very subtle, and maybe not worth bothering about. Like moving one row back in the hall.

There are actually three issues when making comparisons: is there an audible difference? Which is more accurate? Which is more satisfying? The third is especially subjective. If one can be more satisfied for less money, I say that’s a good thing.

I defy anyone to identify such obvious and profound differences under blind conditions. In fact, I’d be willing to place a wager on it.

2 Likes

Hi David: Your say it right; do the test, listen and evaluate. If there is an audible distinction (and there often is), determine its impact. If it’s subtle and not perceived as an improvement, let it go.
If there is (a clear distinction), it really should be in the “oh my” category. If not, leave all as it is. It’s wise to consider long-standing, respected company’s (in their respective fields) if considering a premium cable/cord evaluation.

And, as you wrote, quote:

“There are actually three issues when making comparisons: is there an audible difference? Which is more accurate? Which is more satisfying? The third is especially subjective. If one can be more satisfied for less money, I say that’s a good thing.”

Perfectly said !

(Although I strongly suggest one “focuses” on LF (low-frequency) performance; articulation, definition, resolution/transparency. Simply put, “clean” bass performance results in enhanced lower-midrange performance; in fact, the first seven octaves (20 Hz. to 2,560 Hz.) is critical in defining … well, everything.
Those (cables/equipment) that display (over-bloated) bass power or extension (sometimes incorrectly referred to as “rich”) often corrupts the upper-bass and lower midrange -the fundamental frequency range of virtually all (90%) of typical music instruments and vocals.)

pj

David: That would be way too easy. And repeatable. Time and time again.
Yet for this, we must enter into some pricey and sometimes truly spectacular-priced cables (if we’re discussing cables). Unless you’ve experience this kind of thing (you clearly have not), one has no idea of the enormous contribution (impact) it has upon sound quality. It is (as obvious as), day and night.

No need to be uptight. Nor to lose your money. And make no mistake about it, you will lose it.

pj

Though I don’t over analyse these things myself I believe that expectation bias is a major hurdle and level matching is considered necessary. If I wanted to do some serious comparison I would take these factors into account.I’ve listened to a range of highly priced cables through my own system. These have been borrowed or given. I’ve yet to hear significant differences or even improvements with these products. I sold these on.The biggest differences have come with speaker changes.

While I appreciate much of what the guy writes about, that TEAC DAC rates as one of the worst I’ve ever auditioned, so the results don’t surprise me. I don’t agree with the idea of running some amateur-grade tests (RMAA is amateur grade in my book) on a single device and making sweeping declarations. That is poor science at best.

A proper experiment (since someone is likely to ask) would be to run a full noise profile (from Hz to MHz) of the USB output of a variety of devices with a variety of power supplies, using an AP555 at least. Then take multiple types of measurements (regular, multi-tone and varied types of IMD) of a 20 or 21-bit capable DAC with and without those devices connected so that one could get an idea of the affect of noise (which is the primary problem with USB) on the analog output.

Ironically, from my experience, the worse the quality of the USB receiver on the DAC, the more the differences in transport will make a significant difference. The three new ~$2000 DACs I have here seem to be fairly unaffected by USB transport though. Any differences between the RPi 2 I have been playing with (using DietPi as a Roon endpoint) and my Mac Mini are small enough that I wouldn’t consider myself a reliable judge of their significance. A proper set of experiments would be far better than USB converter and transport pot luck which we end up playing.

2 Likes

Amos: Excellent! And well written.

pj

Very sensible. If nary a thing can be detected, let sleeping dogs lie.

However I must note; it is not necessary to either level match or consider ‘expectation bias’ -have someone make the changes (delicately).

If one (cable/change) is not consistently and obviously better, save any further time and money.

What one should be mumbling under their breath should be this: " OK you newcomer (example cable) hot-shot pos, let’s see what you’re all about, prove some things to me …" lol -but true !

pj

1 Like

I wouldn’t dream of being uptight. Whatever floats your boat is fine, for you. But being by nature a curious person, when someone claims to hear things for which there is no known physical explanation I have the urge to get to the bottom of it. One possibility that can’t be overlooked is that in a situation like that the listener, when fully aware of what he is listening to, is deluding himself.

I’ve spent a fair amount of time comparing speaker and interconnect cables, and concluded that any differences I thought I might be hearing were not worth chasing.

I don’t buy the quasi quantum mechanical arguments against blind listening, that getting valid results is as fundamentally impossible as determining through which slit in a diffraction grating a single photon passes. If audible differences do exist then under some condition, with some program material, some listener who doesn’t have his ego attached to hearing them or stand to lose his reputation as an audio critic (like the wine critics in “Bottle Shock”) must be able to hear them reliably under blind conditions.

4 Likes

[Moderated]

Discuss the subject please, not the person posting.