"Audiophile" SD Card Transport?

[barrows]

Optical Network connection works great, and makes everything upstream of the optical connection irrelevant in terms of noise etc.  No limits on sample rates either.  One can do whatever they want upstream (powerful computer oversampling to DSD 512 running HQPlayer, for example), and then what goes on in the actual audio system can be simple, and not compromised by any computer based "problems" (RFI, etc).  No need for expensive audiophile "servers" or endless server and Network tweaks.

opticalRendu to USB DAC, or optical Network input DAC, simple, and no tweaks needed for amazing sound quality.

 

I have listened to the Mirus Pro DACs via SD card, and it the SD card was no revelation in comparison to the USB input using an excellent USB source.  And navigating the card was tedious and tiresome at best.

[barrows]

@The Computer Audiophile

 

I was attempting to speak generally here, is that a problem?  Please contact me privately if so.  For example, Luminary's has a DAC with optical Network input...

[barrows]

On 1/29/2019 at 11:37 AM, Doak said:

Honestly, I have gotten blaze' regarding streamers/servers/transports and am thinking more seriously now about the beauty of "simple."

 The closest thing to a "truism" I have encountered in 40+ years of "messing with" music reproduction systems is (to paraphrase Einstein): Simple sounds better - as long as it works. 

The above is a quote from the first post.  Looks to me like this thread is about whether (or not) an SD card transport could provide a simpler data path and better sound.  I was responding to that, from the basis of having some experience with an SD card transport, and various other transport methods.

Without comparing and contrasting there is no discussion at all, or learning.

[barrows]

12 minutes ago, Ralf11 said:

what is the benefit of using a SD Card to store music files?

Considering that the file has to be removed from the SD cards and then cached in some kind of memory before playback, it is hard to imagine there being any benefit (and vs. what?)

Those over on the "massive" thread using NUCs as Renderers insist on removing the SD card after loading the OS from it, insisting that it is a cause of unwanted noise... although I would prefer to see some objective proof of that before believing it!

 

This stuff is much more complex than armchair engineers (including this analog power supply designer) understand.  Software playback writers have tried many different approaches, and all have their own takes on which approach achieves "best" sound quality.  Is it better to pull large chunks of data at once, or to more slowly pull smaller chunks more often... at what rate?  Or does it matter?  How many layers of noise isolation are necessary, what is the largest source of noise and is one addressing that first...  It is indeed a rabbit hole.  And, on the hardware side, howe to best isolate the audio system from that noise...

[barrows]

6 minutes ago, Doak said:

While SD cards have been noted to be an excellent performing form of music file storage this previous post of mine in this thread should explain the rest:

 

But playing files from an SD card still requires a "computer" and a playback software system.  An SD card cannot directly output an I2S, real time, data stream to a DAC.  What this does is put a computer running playback software inside the DAC, exactly where you want the least noise...

[barrows]

2 hours ago, Doak said:

I am seeking to simplify the process of playing back digital music files with a possibly simultaneous result of simplicity and enhanced sound quality.

Not here to prove anything.

Not here to argue.

“Why’s” are nice to know, but it’s the “What’s” that really matter. 

Besides, I’d rather be listening to music. 

OK, I am confused now.  I already have a system which is super simple to operate and playback any music in my library.  Everything  is accessed from a tablet and with a couple of taps I can play any music in my library in a coupe of seconds with no issues and no agonizing over anything, and the sound quality of the source is fantastically good.  And my library can be as huge  as many TBs, or for those who enjoy streamed music from the Internet (which is not me though) I hear Quobuz offers almost unlimited access to music of all genres with equal quality to home based playback (I have not tested myself, i play only locally stored files here).

What could be simpler, fiddling around with tiny SD cards?  i am just not sure i understand what "problem" SD card based playback would solve, and how?

I have no issue with whatever you might like to do/try with your system, but your stated desire for simpler process and better sound does not seem like something which would be served by an SD card player?

[barrows]

While that power supply does not look impressive, the price is certainly reasonable.  I would make a supply much much better for that.  But of course the cost would be way higher... perhaps $400.00 or $500.00 for something capable of 1.5A @8VDC continuous and easily capable of 2A+ for peak currents.

Of course the onboard local regulators are still really important, I doubt this card player is using the best local regs, or that many of them given the price.  There is only so much a really good external power supply can do, so it may not make sense to go too far in the upgrade department if the local regulators and decoupling on the main board are not top notch.

[barrows]

12 hours ago, sligolad said:

If you google SLC v MLC you will find a lot of good information on why SLC costs more than normal MLC SD cards.

My read is that a lot more processing and electrical activity takes place getting data off an MLC card and I think many of us have discovered that less is more when it comes to reading digital files in preparation for and during playback.

 

In the journey to best sound I have found that the following are super important:

fast clean power (super-caps at end stage)

good clocks with re-clocking just before DAC (not ultra cost clocks but good and potato chip at end of I2s)

Minimum bit processing of digital audio files (stripped back control software, wav only single tread real time operation)

Storage medium with least data processing activity to deliver digital files (SD SLC card)

 

I am no electrical engineer and my conclusions are defined by a long journey of listening tests so YMMV.

I also spent a lot of money and time on USB out and while it has significantly improved with lots of add ons in recent years, after hearing SD playback done at less than one tenth the cost it became my go to for lights out  being there listening.

 

Do you have links which show measurements of less noise from an SLC SD card?

[barrows]

3 hours ago, hopkins said:

He did not claim there was less noise, just less power consumption.

That is why I asked the question: he actually speculated (or inferred if you prefer) that because there was less power consumption given a fixed level of I/O, that there is less noise.  Given the resources of high speed computing manufacturers it would be easy fro them to measure such noise.

On this site there is a lot of conflicting information regarding playback from files stored on SD cards, we have people claiming that SD cards are actually the noisiest storage medium for playback, all the way to people claiming they are the quietest storage medium for playback.  Clearly there is no consensus when it comes to subjective evaluation, perhaps some measurements might add some clarity within all the "noise".