Brand of SD Card in Raspberry Pi makes a difference to sound quality?

[LongtimeListener]

Hi All,

 

I have the impression that the brand of the SD Card used makes the Raspberry Pi sound different. Now I came across that topic on the Volumio Website.

 

https://forum.volumio.org/brand-card-used-makes-difference-music-quality-t13351.html

 

What is your thought about this?

 

 

[bob_m_54]

It doesn't seem logical to me. The SD card holds the OS, which is loaded into RAM. Test it out with a few different brands and see if you can pick any difference.

[davewantsmoore]

On 30/11/2019 at 12:28 PM, LongtimeListener said:

What is your thought about this?

[aussievintage]

Hilarious

[keinesorge]

On 30/11/2019 at 9:28 AM, LongtimeListener said:

Hi All,

 

I have the impression that the brand of the SD Card used makes the Raspberry Pi sound different. Now I came across that topic on the Volumio Website.

 

https://forum.volumio.org/brand-card-used-makes-difference-music-quality-t13351.html

 

What is your thought about this?

 

 

Hi Longtimelistener,

 

I've been running a pi based setup for a couple of years and when i was putting it though its paces, the micro sd card used did make a difference.  While i did struggle to effectively conceptualize things (an ee), they did sound different enough warrant further testing.

 

Similar to the OP on the volumio thread, i've tested a few sandisk ultras (with and without the A1 logo), and a couple of samsung cards.  I'd be happy to give the toshiba unit a go but no where in Perth retails them.

 

I've found a dick smith / kogan listing (i think its the same seller) who has the item without its standard packaging (which i'm reading as possibly non genuine).

Edited December 2, 2019 by keinesorge

[LongtimeListener]

On 02/12/2019 at 7:35 PM, keinesorge said:

the micro sd card used did make a difference

God thanks, I am not crazy as some here think .

I have seen many times sound differences in digital hardware where I also came from the sceptical side of things that 1 and 0 should sound all the same, but they don't.

I suspect jitter to be the culprit and why a different brand of SD Cards can make it better or worse, I don't know.

 

There is the possibility to boot the OS on a RPI from a hard disk ignoring the SD Card. I will check out this and hope to see a difference in the sound quality.

Edited December 4, 2019 by LongtimeListener

[keinesorge]

11 hours ago, LongtimeListener said:

God thanks, I am not crazy as some here think .

I have seen many times sound differences in digital hardware where I also came from the sceptical side of things that 1 and 0 should sound all the same, but they don't.

I suspect jitter to be the culprit and why a different brand of SD Cards can make it better or worse, I don't know.

 

There is the possibility to boot the OS on a RPI from a hard disk ignoring the SD Card. I will check out this and hope to see a difference in the sound quality.

 

While i cannot say with certainty what gives with the difference in micro sd cards...  but when the position of the sd card is slightly changed with the use of use an extender (for ease of use), the differences remain.

 

 

 

Edited December 5, 2019 by keinesorge

[aussievintage]

When the OS is loaded and a song is playing, it's all happening in memory (anything on the card is buffered in memory)  The sdcard cannot make a difference.  

[bob_m_54]

Maybe the Pi was moved when switching cards, and the position of the cables was changed in relation to each other?

[keinesorge]

If the system is resolving enough the difference is clear.  The possible reasons for the variation in sound is not something i care to get into.

 

I've been building and voicing gear for quite some time and all my electronics (not speakers) are voiced at the component level.  The change with the cards can be as significant as a change in a signal / decoupling cap etc (think Elna Silmic II to Sanyo WG).  It does not seem to change with duration of use  / quality of DC power fed to the Pi / small change in distance of card from socket.

 

I have noticed this aftifact on later versions moode 3.8x with a RPi 3B.  I have not tested this with other installs / hardware.  The variation in cards could also be heard with all software kernels used on the older versions of moode (real time / low latency etc).  Ver 6.x has not been tested for this.

 

The differences in what is heard is not bloom / warmth / instrument seperation etc.  Some make of cards sound thiner and with added grain from the midrange up.

 

@aussievintage, the OS is loaded into memory for picoreplayer.  Not all of them.

 

@bob_m_54, Same power cable, same  interconnects etc.  The swap can be done without moving things around too much.  As bizzare as it might sound, the change on the software hardware combo was significant.

[sir sanders zingmore]

Does the capacity of the card make a difference? I mean two cards of the same brand but different capacity ?

[Ittaku]

No.

[mwhouston]

I’ve been a Pi user for a number of years now and my Pi is on and producing music at least 30 hours a week. File based music or streaming.  And for some time now under the latest Mo0de. 
 

My qualifications are electronics and microprocessing and I have worked in technology all my life, if not longer. I do not think different SD cards change the way the music sounds. I also don’t think different USB cables maker a difference. But some very respected reviewers would disagree. Not sure they have hard core electronics grounding. Maybe they are sniffing fairy dust. 
 

I don’t think  jitter comes into play when a SD card is read or SD cards have a “jitter rating” as processors may. 
 

This is how it goes; if you think it sound better with one SD card over the other, then it does, whether it really does or not. Because it is your personal option. 
 

I’m not having a crack at you here I’m just trying to be funny; get three new razor blades and lay them flat on a small makeup mirror. Make a triangle shape with them. Place the SD card in the middle of the triangle and place them outside in full view of the moon but only on the night of a full moon. This should sharpen the sound. 

[aussievintage]

1 hour ago, keinesorge said:

@aussievintage, the OS is loaded into memory for picoreplayer.  Not all of them.

No, but all executables and data are loaded into memory (or processor registers) for access by the processor in ALL operating systems.  Whenever a process reads the drive, it is loading it into memory ram/registers/cache memory.  This is different from Picoreplayer which runs the OS from a memory "drive".

Edited December 6, 2019 by aussievintage

[keinesorge]

1 hour ago, sir sanders zingmore said:

Does the capacity of the card make a difference? I mean two cards of the same brand but different capacity ?

I dont know.

[keinesorge]

.

Edited December 6, 2019 by keinesorge

[March Audio]

On 30/11/2019 at 9:28 AM, LongtimeListener said:

Hi All,

 

I have the impression that the brand of the SD Card used makes the Raspberry Pi sound different. Now I came across that topic on the Volumio Website.

 

https://forum.volumio.org/brand-card-used-makes-difference-music-quality-t13351.html

 

What is your thought about this?

 

 

Its not possible.

[LongtimeListener]

Hi everybody, I revive this old topic I started a couple of years ago since it still baffles me after using different SD-Cards and they sound different to me.

 

Now, before somebody declares me crazy, I have to say that I am a very rational person. I am strongly in the 0 and 1 camp, believe in error correction and clock synchronization and master clocks and eliminating jitter etc etc..., with other words, if the digital information is lost it will be re-read and then processed, if this is not possible then it clearly crackles...

This is the rule for everything in the digital sound chain till it is converted in a DAC and then when it becomes an analog signal things look different.

 

So this out of the way: I have seen differences over the years in using different laptops, netbooks and raspsberry pis and all made a difference in the sound. So how, I am asking, can this be if there is no change in the quality of sound since all 0 and 1 are treated equally?

My sound chain has never changed during those times, except RPI 3/4, Laptop, netbook connected over the years -> same cheap USB cable -> same highend DAC -> same highend amplifier -> same highend loudspeaker in the same room.

 

Now to the SD Cards: due to the lifespan of SD Cards in the RPIs and trying different sound server software like Moode or Volumio I came across the issue that SD Cards sound different even when installed the same Moode version.

 

Sorry to say, seeing all those differences in sound quality in the digital sound chain, where can it be that all 0 and 1 are equal?

 

 

Edited June 25, 2021 by LongtimeListener

[sir sanders zingmore]

Funny I was just inadvertently experimenting with this very thing. I set up a pi to stream to my desktop setup just yesterday and somewhere in the process I misplaced my SD card. So I grabbed another one and flashed Roopiee onto it.

 

Then I found the original one and put it back in. 

 

To be fair, I wasn't listening for a difference, but I didn't notice any difference whatsoever when I changed cards… sorry

Edited June 25, 2021 by sir sanders zingmore

[davewantsmoore]

 

On 25/06/2021 at 12:58 PM, LongtimeListener said:

`if the digital information is lost it will be re-read and then processed, if this is not possible then it clearly crackles...

 

Jitter is much more likely than this scenario.

 

 

On 25/06/2021 at 12:58 PM, LongtimeListener said:

can this be if there is no change in the quality of sound since all 0 and 1 are treated equally?

 

Have a look at the analogue output and see if it's the case.

 

Hint:  Sometimes it is and sometimes it isn't.

 

 

TBH, I'd question your assessment of your DAC.    Why does it produce a different analogue output to such an extent that it is

audible when the incoming (USB) signal changes so slightly?

 

 

OTOH, assuming this is what is going on (the Pi is letting things down), the you need to upgrade (the source).    You could use something like a "BridgePi" from Ian Canada to add a USB audio output to the Pi, like an Amanero384.

https://github.com/iancanada/DocumentDownload/blob/master/Adapters/BridgePi/BridgePiUsersManual.pdf

 

 

[Gordon Macfarlane]

This is just one of the topics that will go round and round on the subjectivist vs objectivist carousel. I can't say I have ever compared cards but have always followed manufacturers suggestion to use good quality high speed cards including an emcc sd card replacement.

Once thing that does offer an easily discenrable difference is to use an operating system that gives you the option of loading the OS into ram. I use Gentoplayer , which is unfortunately no longer free, and the sound does improve when the os in loaded into memory.

Iuse a pi4 with a full compliment of Ian Canda hats and power supplies and the gives me better soundquality through coax than using USB.

These days you  can aslo use a usb stick to load the OS or even a usb  sata drive with the pi but while these will be faster I have no idea if the will change the sound produced.

From my perspective there are far more imporatant things to work on than what I imagine will be small diferences , if any at all between SD cards. That being said if you hear a difference between the card just choose the best sounding one and be happy.

 

[davewantsmoore]

18 minutes ago, Ozcall said:

Once thing that does offer an easily discenrable difference is to use an operating system that gives you the option of loading the OS into ram. I use Gentoplayer , which is unfortunately no longer free, and the sound does improve when the os in loaded into memory.

Iuse a pi4 with a full compliment of Ian Canda hats and power supplies and the gives me better soundquality through coax than using USB.

Same.

[LongtimeListener]

Thanks for all the input. As I said before I am in the 0 and 1 group who understands the logic of digital information is either there or not and nothing between. But many say here that changing something in the digital sound chain changes the sound. 

How can this be if the clock of my DAC is correcting everything and re-reads the data if there is a loss?

I mean, following the logic of the pure 0 and 1, there should be absolutely no difference in whatever you do in the digital sound chain till the 0 and 1 enter the DAC and the signal becomes analog...

 

At the moment the COVID makes it difficult to hold a hearing test here in my audio club since 12/19 when I first was thinking about testing this SD Card thing...

 

Thanks again and cheers

 

[MrBurns84]

Just whilst on my end of year break, came across this thread and with the missus and kids out grocery shopping.  Flashed volumio 3 onto 3 different cards,  transcend, kingston and verbatim 16gb microsdhc class10 cards and played music over 15mins on each on a Rpi4B and topping P50 LPS.  No difference whatso ever.  

[davewantsmoore]

On 01/07/2021 at 4:16 PM, LongtimeListener said:

How can this be if the clock of my DAC is correcting everything and re-reads the data if there is a loss?

It doesn't.

 

It doesn't perfectly correct errors in the timing of the data..... and it does nothing to correct incorrect data (eg. a 1 instead of a 0), although the later is (under normal circumstances) as rare as rocking-horse poop (the 1s and 0s are always right).

 

On 01/07/2021 at 4:16 PM, LongtimeListener said:

I mean, following the logic of the pure 0 and 1, there should be absolutely no difference in whatever you do in the digital sound chain till the 0 and 1 enter the DAC and the signal becomes analog...

Yes.

Errors in the timing of the 1s and 0s can affect the DACs analogue output.

 

On 01/07/2021 at 4:16 PM, LongtimeListener said:

thinking about testing this SD Card thing...

It's possible that differences in the player (computer) to do with the SD card make the data very "jittery" (timing is very inconsistent) .... or that there is significantly more / different electrical noise.....  AND that either of these things make it through to the DAC and affect the analogue conversion.