EEVblog #719 - Sony Low Noise Audiophile SDXC Memory Card

[Wytnucls]

I never mentioned pianos, only strings instruments and violins in particular. Once again, there is no controversy about pianos. Old Stradivarius violins have a reputation for sound quality above modern instruments, which is not supported by double-blind testing, that's all. Leave pianos out of this, please. if you want to talk about pianos, start another thread.

[Fungus]

"You don’t want some crappy unbiased cable, because it’ll “slow down parts of the signal differently” and that’s “a big problem for very time-sensitive multi-octave audio.”

THIS IS A NETWORK CABLE!!! Can someone try to explain me how the hell audio quality, that comes from a digital stream, carried on UDP packets, and probably encoded by a lossy  MP3 algorithm, will affected by the cable???

It's pretty obvious if you think about it: Each instrument is transmitted along a different wire in the cable. That causes timing problems in the music if those wires aren't carefully matched.

[ivan747]

It's 100% correct in the fact that someone who has just paid 5x the amount for a memory card WILL hear the difference because they have told themselves that it is better, much like anything!

There is something similar going on in the classical music world, where famous performers and music critics are convinced that old famous string instruments sound better than new ones. (A Guarneri or Stradivari violin sells for millions of dollars at auctions).
This was thoroughly debunked in double blind experiments, but the myth lives on.

You know what? I'd still pay a million dollars for a Stradivari if I had a million dollars, because it's a historical instrument made by a prodigy family. Unlike audiophools.

[AF6LJ]

This is a real head shaker;
This doesn't happen in the situations we are discussing.

"You don’t want some crappy unbiased cable, because it’ll “slow down parts of the signal differently” and that’s “a big problem for very time-sensitive multi-octave audio.”

THIS IS A NETWORK CABLE!!! Can someone try to explain me how the hell audio quality, that comes from a digital stream, carried on UDP packets, and probably encoded by a lossy  MP3 algorithm, will affected by the cable???

It's pretty obvious if you think about it: Each instrument is transmitted along a different wire in the cable. That causes timing problems in the music if those wires aren't carefully matched.

[AF6LJ]

Perhaps, but the piano is classed as a keyboard instrument. You're muddying the waters, pitch tuning doesn't influence the sound quality of the instrument.

It is my understanding a Piano is a Percussion instrument, given that the strings are struck by a hammer. Now if you are John Cage(RIP) that can be carried to it's logical extreme. 

[AndyC_772]

They claim: "You don’t want some crappy unbiased cable, because it’ll “slow down parts of the signal differently” and that’s “a big problem for very time-sensitive multi-octave audio.”

THIS IS A NETWORK CABLE!!! Can someone try to explain me how the hell audio quality, that comes from a digital stream, carried on UDP packets, and probably encoded by a lossy  MP3 algorithm, will affected by the cable???

Just for completeness... the cable plugs into an ethernet Physical Layer IC, or PHY for short. More often than not, the PHY connects to the host processor via an interface called RMII, which has only one clock. That clock is locally generated from a crystal oscillator.

So: the timing on the cable is even more irrelevant than you might expect, even if it were carrying audio as a series of (uncompressed) samples. Any mismatch between the source and destination is taken up by a FIFO in the PHY, which empties during the gaps between packets.

Never mind that the RMII clock probably isn't derived from the same source as the sample clock anyway... 

[Fungus]

Perhaps, but the piano is classed as a keyboard instrument. You're muddying the waters, pitch tuning doesn't influence the sound quality of the instrument.

It is my understanding a Piano is a Percussion instrument, given that the strings are struck by a hammer. Now if you are John Cage(RIP) that can be carried to it's logical extreme.

Times have changed.

I once corrected the Wikipedia page which had pianos in the "strings" section  (I moved it to "percussion").

"Strings" was obviously wrong but since then the orchestra has been given a whole new section called "keyboards".

http://en.wikipedia.org/wiki/Orchestra#Instrumentation
http://en.wikipedia.org/wiki/Keyboard_section

Yes, there's five sections now.

[Wytnucls]

It's 100% correct in the fact that someone who has just paid 5x the amount for a memory card WILL hear the difference because they have told themselves that it is better, much like anything!

There is something similar going on in the classical music world, where famous performers and music critics are convinced that old famous string instruments sound better than new ones. (A Guarneri or Stradivari violin sells for millions of dollars at auctions).
This was thoroughly debunked in double blind experiments, but the myth lives on.

You know what? I'd still pay a million dollars for a Stradivari if I had a million dollars, because it's a historical instrument made by a prodigy family. Unlike audiophools.

And that would be a sound investment (no pun intended). Their inflated value comes from the false aura around a limited number of instruments, not so much their historical value. Some of the Stradivari violins were actually shoddily made, depending on the period, and modern luthiers, like Peter-Stefan Greiner, can make superior instruments, using traditional methods.

[rsjsouza]

Good primers to understand digital/analog with respect to audio:
http://xiph.org/video/vid1.shtml
http://xiph.org/video/vid2.shtml

These videos are so good they hurt!

[AF6LJ]

Perhaps, but the piano is classed as a keyboard instrument. You're muddying the waters, pitch tuning doesn't influence the sound quality of the instrument.

It is my understanding a Piano is a Percussion instrument, given that the strings are struck by a hammer. Now if you are John Cage(RIP) that can be carried to it's logical extreme.

Times have changed.

I once corrected the Wikipedia page which had pianos in the "strings" section  (I moved it to "percussion").

"Strings" was obviously wrong but since then the orchestra has been given a whole new section called "keyboards".

http://en.wikipedia.org/wiki/Orchestra#Instrumentation
http://en.wikipedia.org/wiki/Keyboard_section

Yes, there's five sections now.

Oh Well, times are a changin I suppose....
Does this mean I have to give up my HP-412 for a Fluke DMM?

The hole digital VS analogue argument has spread across all fields of interest not just audio, video amateur radio and even the crappy voice quality of cellphones are not without those who find the whole thing one big step backwards.

As for me, it doesn't really matter much, I quit my interest in HiFi when CDs came out all it took was the garbage that was generated every time a drum solo began.
Now music is just background filler.
I have hardware in my computer that will sample up to 192KHZ and the 96KHZ stuff does sound better.

Some day we will have digital sound that is HiFi without all the stupidity.
 That day doesn't appear to be very far off.

[EE-digger]

What's the BER of a $5 SDHC card?  About 10^-15 perhaps (uncorrected).
What's the BER of this card?  I'm guessing around 10^-15 or so (uncorrected).

So, bit error rate is essentially zero for most intents and purposes.  Let's move on...

Someone mentioned that the emissions graph says that it is radiated.  Ok, so if this is true then we've got some possible scenarios where the $$ card produces audio with a higher noise free dynamic range.

scenario #1 - the SDHC card is mounted direcly above the ADC section - don't laugh - if the product is tiny, there can be a lot of compromises in placement of parts, where marketing wants certain features, etc. - Maybe the ADC isn't above the SDHC card, or under it, but perhaps the emissions are enough to alter the ADC noise AND/OR produce mixing products.  Maybe that 1,2,5 or so MHz emission is mixed down INSIDE the ADC and it is now planted in your band of interest and untouchable by your filtering.  Maybe a remnant of the emissions remains on the supply rails where, if synchronous, can still take serious filtering to mostly remove.

scenario #2 - if the read clocking of the SDHC ever becomes synchronous with the ADC conversion clock, you wreak havoc on the converter noise - much worse than in #1 - you could start getting downconverted clock noise sweeping in and out of synch with the audio for horrendous effects.  Maybe this could happen as a function of storage location, maybe as a function of wear leveling activity, etc.

While I haven't experienced either of the above with audio, I was extrapolating from problems I've had to solve in the past in telecom measurement equipment.  It's easy to turn your sampled data system into a spectum analyzer if you're not careful. 

So, I'd like to wait and see some noise and distortion numbers between the cards and with the equipment clearly identified.  Maybe when you crank the volume all the way up, it is still quiet with this card 

Not a huge Sony fan but always had a place in my heart for Trinitron monitors (pro video and PC)

Den

[Richard Crowley]

Not a huge Sony fan but always had a place in my heart for Trinitron monitors (pro video and PC)

During the early days of commercial color TV production, Trinitron monitors (Sony and other brands) were unfavorable for camera monitors because the aperture grille created a picture that was artificially too "crisp". It was harder to judge whether the camera was in critical focus or not because the Trinitron made it look better than it actually was. Most broadcast TV cameras used monochrome CRTs for viewfinders to enable the camera operator to judge focus properly.

In modern (digital) times, we use software to artifically enhance the edges of things in the frame. "Peaking" is added to the viewfinder signal in a color of the operator's choice to much more easily see when the camera is in focus (the high-frequency from the transition is at maximum amplitude.)

Here is an example of a viewfinder image with artificial red peaking edges added...

[Seaton]

Good primers to understand digital/analog with respect to audio:
http://xiph.org/video/vid1.shtml
http://xiph.org/video/vid2.shtml

These videos are so good they hurt!

Totally agree. But would an audiophool take any notice, even if they had the ability to understand them. It would be like trying to convert a Jihadist to be an Atheist.

[Fungus]

Totally agree. But would an audiophool take any notice, even if they had the ability to understand them. It would be like trying to convert a Jihadist to be an Atheist.

Why would you need to learn math when you can hear the difference?

[EvilGeniusSkis]

Is somebody's calendar off? It's not April fools day yet. Ya got 1 month t go.

in aus it is march fool's day not april.