Solid Snake-Oil Storage: This SSD Is Aimed at Audiophiles

[mbhangui]

That has nothing to do with the time axis though now, does it?

It does. As I understand the biggest problem is this time domain problem which happens at Analog to Digital conversion as well as Digital to analog conversion. Noise has direct relation to jitter.

WHAT ARE THE REASONS FOR JITTER?​

There are several factors that impact on clock deviation:

• clock generator instability
• noise in the digital audio signal line

What Is Jitter in Audio?

You hear this term jitter quite a lot in audiophile discussions, particularly with reference to digital audio conversions or DACs. Yuri ends the year with a more detailed explanation of what jitter actually is, measured,

headfonics.com

 

[superczar]

noise has direct relation to jitter.

​

So if I am reading this correctly

A - Electrical Noise resulting due to the uninformed usage of a cheap nvme drive makes its way into a fancy DAC - which in turn destabilizes its system and causes say a 50hz waveform to turn into a 49 or 51 hz which in turn results into say a noticeable deterioration of the original rhythm
i.e. distorts a waveform noticeably on the time domain for a signal in the range of 20-20x (10^-3)

vs

B - Meanwhile the mass market chipsets on your phone apply 10^-10 second scale adjustments for time dilation due to relativity on the signal received from GPS satellites (time domain precision )
while plugged into a 300 Rs charger plugged to your car's alternator (and all the electrical noise coming from it)
No wonder why my location shows as Senapati Bapat road when I am in Aundh

(or at the scale of precision required for audio signal precision, showing me in Oklahoma when I am in Delhi)

PS- I don't mean to offend, Just jousting so think of it as just a bit of a tongue-in-cheek reply

 

[raghupb]

And where exactly does the HDD come into this?

As for your argument on DACs and time domain drifts, Take a cheap DAC like the Apple USB C dongle (sub 1K INR?)
use any waveform generator to create a 50hz (or any other frequency) test tone
Feed analog output into a half decent oscilloscope - and check if you see any time domain problems
(You won't)

Believe it or not, audio DACs sit at a rather low end of the engineering complexity spectrum of digital signaling

I think over the last decade DACs have become quite good.
They have solved most of the clock issues with master clocks.
And these clocks + PLLs are actually quite good (if I read the spec correctly).
The DAC I'm listening to now also has an FPGA that supposedly has some clock related control mechanism.
Fancy stuff, sounds good though.

Every DAC has a signature sound that one may agree with or not. E.g. I don't like Sabre DACs
Most DAC chips (delta sigma or R-2R) or discrete ladder networks are fairly decent up to 16 bits.
Maybe even 24 bits. But to say that HDD or SSD is gonna make a difference in SQ if a big stretch (for me at least).

Cheers,
Raghu

 

[mbhangui]

So if I am reading this correctly

A - Electrical Noise resulting due to the uninformed usage of a cheap nvme drive makes its way into a fancy DAC - which in turn destabilizes its system and causes say a 50hz waveform to turn into a 49 or 51 hz which in turn results into say a noticeable deterioration of the original rhythm
i.e. distorts a waveform on the time domain

It is far fetched to say nvme drive is the biggest culprit to the noise. The computer has a SMPS which by far is the noisiest component. No solid state drive can do away without dc to dc convertors and these too will generate noise. But I doubt it will be anywhere near that of the computer SMPS. So by making a fantastic NVME drive and reducing it's noise will have almost zero effect on the performance of the DAC. Which is why I said in the earlier post that this drive is pure snake oil.

vs

B - Meanwhile the mass market chipsets on your phone apply 10^-10 second scale adjustments for time dilation due to relativity on the signal received from GPS satellites (time domain precision )
while plugged into a 300 Rs charger plugged to your car's alternator (and all the electrical noise coming from it)
No wonder why my location shows as Senapati Bapat road when I am in Aundh

(or at the scale of precision required for audio signal precision, showing me in Oklahoma when I am in Delhi)

PS- I don't mean to offend, Just jousting so think of it as just a bit of a tongue-in-cheek reply

no offence taken

 

[mbhangui]

Maybe even 24 bits. But to say that HDD or SSD is gonna make a difference in SQ if a big stretch (for me at least).

Indeed and hence it is snake oil. But the point is 0s are not zeroes and 1s are not just ones as many are claiming. It is also about perfect timing of the arrival of the 0s and 1s. Listening to music is totally different than copying a file where time distortion just doesn't matter.

 

[firearm12]

So if I am reading this correctly

A - Electrical Noise resulting due to the uninformed usage of a cheap nvme drive makes its way into a fancy DAC - which in turn destabilizes its system and causes say a 50hz waveform to turn into a 49 or 51 hz which in turn results into say a noticeable deterioration of the original rhythm
i.e. distorts a waveform noticeably on the time domain for a signal in the range of 20-20x (10^-3)

vs

B - Meanwhile the mass market chipsets on your phone apply 10^-10 second scale adjustments for time dilation due to relativity on the signal received from GPS satellites (time domain precision )
while plugged into a 300 Rs charger plugged to your car's alternator (and all the electrical noise coming from it)
No wonder why my location shows as Senapati Bapat road when I am in Aundh

(or at the scale of precision required for audio signal precision, showing me in Oklahoma when I am in Delhi)

PS- I don't mean to offend, Just jousting so think of it as just a bit of a tongue-in-cheek reply

Your point self explains why technology and equipment is important. GPS satellites are equipped with atomic clocks to be precise. They don't do that with after market chips and they can't do that with after market chips. For time to be precisely synchronized among devices, almost all devices on internet or cellular devices are synchronised to global time servers (which are synced to reference clocks) which can calculate time more precisely than after market chips. Normal day to day devices have tendency to drift time calculation as clocks are not precise and one can easily observe the drifts, no fancy equipment needed here. So it means that good equipment is neccessary everywhere where precision is needed.

 

[raghupb]

I feel some of our FMs who have experimented extensively with computer based playback should collaborate (or conspire) and write a white paper
@keith_correa, @venkatcr, @superczar, @mbhangui
What say?

Break it down as:
1. HDD/FS/Processor/OS (and gobbledy-gook)
2. DACs (and gremlins)
3. Pre-amps/Amps (and boo-boos)
4. Cables & capacitors (just to add some spice )

Cheers,
Raghu

 

[venkatcr]

I am very very happy with the music I get from 6TB of mechanical drives played through ordinary USB into a Topping D50s and amplified. I have never heard any jitter, noise, etc. From absolute silence when nothing is playing, to a satisfying sound stage when Zubin Mehtha is conducting an Orchestra, it works extremely well for me.

And yes, when an Orchestra is playing, I dont hear any Rafi or SPB singing. Clock or no clock.

Thanks, but I am a great believer in music played from a digital system with no fancy equipment.

Cheers.

 

[raghupb]

<snip>
And yes, when an Orchestra is playing, I dont hear any Rafi or SPB singing. Clock or no clock.
<snip>

If this happens at my place, I'm either stoned or the "greats ghosts" have blessed my home.

Cheers,
Raghu

 

[mbhangui]

I feel some of our FMs who have experimented extensively with computer based playback should collaborate (or conspire) and write a white paper
@keith_correa, @venkatcr, @superczar, @mbhangui
What say?

Break it down as:
1. HDD/FS/Processor/OS (and gobbledy-gook)
2. DACs (and gremlins)
3. Pre-amps/Amps (and boo-boos)
4. Cables & capacitors (just to add some spice )

When using computer audio, the ideal thing is to have two systems
1. The system which has the hard disk, smps, etc shouldn't have the dac. This system should be configured for NFS, Samaba, etc.

2. The system which has the DAC should ideally be able to run on a low power supply, have a good clock. There are reclocker boards which do that. This system should be powered by a very good power supply. e.g. of this is the allo usbridge signature which comes in two flavours. one with a clean usb and another with a clean usb and clean SPDIF out.

I have done extensive work on cables on my setup and ultimately I have found out good cables have low capacitance. Over a period of time, I have discarded all cables and made my own cables using cat6 twisted pair. These give just around 10 pF per meter and the quality of HF portion is unparalled to that of most market cables. The only problem I have with my cable is a very minute 50Hz hum when you play no music and crank up the volume

I have made my own cable tester using the atmega8 microcontroller and calibrated it using a capacitance meter. I'm very satisfied with my system without using costly cables, cable risers, etc. A photo of my cable is below. Every cable I make is of the minimum size required to connect point A to point B

 

[superczar]

When using computer audio, the ideal thing is to have two systems
1. The system which has the hard disk, smps, etc shouldn't have the dac. This system should be configured for NFS, Samaba, etc.

2. The system which has the DAC should ideally be able to run on a low power supply, have a good clock. There are reclocker boards which do that. This system should be powered by a very good power supply. e.g. of this is the allo usbridge signature which comes in two flavours. one with a clean usb and another with a clean usb and clean SPDIF out.

I have done extensive work on cables on my setup and ultimately I have found out good cables have low capacitance. Over a period of time, I have discarded all cables and made my own cables using cat6 twisted pair. These give just around 10 pF per meter and the quality of HF portion is unparalled to that of most market cables. The only problem I have with my cable is a very minute 50Hz hum when you play no music and crank up the volume

I have made my own cable tester using the atmega8 microcontroller and calibrated it using a capacitance meter. I'm very satisfied with my system without using costly cables, cable risers, etc. A photo of my cable is below. Every cable I make is of the minimum size required to connect point A to point B


View attachment 65537
View attachment 65538

If you like the sound from these, by all means use them

But the complete lack of shielding on these, for a low level line signal at 1V rms that is going to be amplified many times over..
That would equate to far more induced current noise (aka electrical gremlins on this thread i guess ) on the final output from speakers than any of the other points bandied about so far - (orders of magnitude more)

 

[mbhangui]

If you like the sound from these, by all means use them

But the complete lack of shielding on these, for a low level line signal at 1V rms would also equate to far more induced current noise (aka electrical gremlins on this thread i guess ) on the final output than any of the other points bandied about so far

Using shield increases the capacitance and capacitor always shows a lower impedance for higher frequency.

Since this is a twisted pair, it does reduce the electrical noise compared to untwisted pair. In my case this 50Hz is noticeable only when not playing music and volume cranked up. But the quality of cymbals, vocals and overall clarity is better than using market cables.

I'm day dreaming about making a large diameter pipe and using some mechanical scheme of having a single wire seperated by air dielectric (air is a bad dielectric and will have very poor capacitance). It's in my todo list.

I have also purchased few keystone jacks and will connect RCA jacks to them and extend them using shielded ethernet cable and measure the capacitance.

This is what I have purchased. Need to find some time to do the project

https://www.amazon.in/gp/product/B09J4N3W58/ref=ppx_yo_dt_b_asin_title_o03_s00?ie=UTF8&psc=1

 

[firearm12]

I am very very happy with the music I get from 6TB of mechanical drives played through ordinary USB into a Topping D50s and amplified. I have never heard any jitter, noise, etc. From absolute silence when nothing is playing, to a satisfying sound stage when Zubin Mehtha is conducting an Orchestra, it works extremely well for me.

And yes, when an Orchestra is playing, I dont hear any Rafi or SPB singing. Clock or no clock.

Thanks, but I am a great believer in music played from a digital system with no fancy equipment.

Cheers.

A person loving music will listen to music only. But some of us might get too obsessed with audio and then suddenly believe that they can hear strange noises and distortions and some can even calculate the distance from singers vocal cord to microphone in mm or distance of his chair from the performer. Then suddenly again one day in future we realise that any of this doesn't matter, only thing that matters is music only.

 

[raghupb]

A person loving music will listen to music only. But some of us might get too obsessed with audio and then suddenly believe that they can hear strange noises and distortions and some can even calculate the distance from singers vocal cord to microphone in mm or distance of his chair from the performer. Then suddenly again one day in future we realise that any of this doesn't matter, only thing that matters is music only.

Kishore Kumar supposedly could nail this
Cheers,
Raghu

 

[surfatwork]

Oh wow.... The cables have made an entrance. Just to remind myself - the original question, I think, was whether a ssd supposedly specifically engineered for audio was snake oil or not....
Where do the cables figure here?
Unless one is talking about the noise from the 440kV, 220kV and so on cables right down to the pesky 220V cable/wire which is hand twisted onto the ancient 220V single phase line outside my home. And screwed into the rusty fuse in my mains box... But then I thought we had the 2.5feet of Uber super power cord between my amp and power socket to solve those problems... don't we? Surely that power cord solves all cable issues?

 

[raghupb]

Oh wow.... The cables have made an entrance. Just to remind myself - the original question, I think, was whether a ssd supposedly specifically engineered for audio was snake oil or not....
Where do the cables figure here?
Unless one is talking about the noise from the 440kV, 220kV and so on cables right down to the pesky 220V cable/wire which is hand twisted onto the ancient 220V single phase line outside my home. And screwed into the rusty fuse in my mains box... But then I thought we had the 2.5feet of Uber super power cord between my amp and power socket to solve those problems... don't we? Surely that power cord solves all cable issues?

My bad!!
Cheers,
Raghu

 

[mbhangui]

Oh wow.... The cables have made an entrance. Just to remind myself - the original question, I think, was whether a ssd supposedly specifically engineered for audio was snake oil or not....
Where do the cables figure here?

Cables belong to the same category of the so called musical ssd - snake oil.
Imagine the last 1 meter of cable making a huge improvement to sound after the electrons have traveled a kilometer from the distribution transformer. Laws of physics be damned.
Regarding interconnects, a good cable (regardless of the cost) will sound the same. All cables have inductance, capacitance and resistance. An ideal cable should have 0 inductance, 0 capacitance and 0 resistance. A straight cable wthout any bends will have 0 inductance. A cable with a good conductor like copper will have almost zero resistance. But any cable running as a pair will have some capacitance. The closer the cable, longer the cable, higher the capacitance. A capacitor blocks DC but conducts all forms of alternating current. Higher the frequency lower the impedance. Hence a cable with high capacitance will have a non-linear response with the frequency of the waveform and will act like a short circuit for high frequency. This will colour the music. A bad cable will make your music warmer.

e.g. my mogami 2549 cable gives me 200 pf for 2 m length. If I use this, it gives me 50K impedance for 15 Khz. If I use a 100k pot, that would rob around 50% of the 15Khz audio, making the music sound very warm. Companies exploit the gullible by claiming exotic material used in the cable to make it warm, bright, etc. And there are people who actually hear the difference, when told about the special quality of the interconnect, power cable, ether cable, hard disk, etc. Our hearing is not just the ear. It includes the perception created by the brain. I too am susceptible to this. e.g. I used to find my costly tube amp better sounding than my sub 10 K amp from allo. But after a blind test I couldn't distinguish between them.

If anything cannot be explained by measurement, science or physics, it is snake oil. In such cases, it is better not to rely just on your ears.

Capacitor Impedance Calculator - Engineering Calculators & Tools

This tool calculates a capacitor's reactance for a given capacitance value and signal frequency.

www.allaboutcircuits.com

EDIT: This was another level of snake oil that was popular around 10 years back. The cable cooker

The Cable Cooker

 

[Decadent_Spectre]

A bad cable will make your music warmer.

Then it is a good cable .

 

[mbhangui]

Then it is a good cable .

solder a 100 pF capacitor across the center pin and the ground of good cable, you will get the same effect

 

[Decadent_Spectre]

solder a 100 pF capacitor across the center pin and the ground of good cable, you will get the same effect

Good to know, thank you.

I will be relying on my ear as always, not on "measurement, science or physics". My money will always go to whatever sounds good to me, people shouting snakeoil makes no difference to me

I see no point to such threads. Those who will buy will buy regardless and those who won't, will not buy regardless.

Carry on!