Sunday, December 19th 2021
An "Audiophile Grade" SSD—Yes, You Heard That Right
A company dealing with niche audiophile-grade electronics on Audiophile Style, a popular site and marketplace for the community, conjured up an SSD that it feels offers the best possible audio. Put simply, this is an M.2-2280 NVMe SSD with a fully-independent power delivery mechanism (one that's isolated from the motherboard's power delivery), and an over-the-top discrete clock-source for its controller. The drive has its own 5 V 2-pin DC input and switching hardware onboard, including [get this] a pair of Audionote Kaisei audio-grade electrolytic capacitors in place of what should have been simple solid-state SMD capacitors that are hard to even notice on any other drive. It doesn't end there.
Most NVMe SSDs have a tiny 2 mm x 2 mm SMD oscillator that's used by the controller for clock-generation. This drive features a Crystek CCHD-957 high-grade Femto oscillator. These oscillators are found in some very high-grade production or scientific equipment, such as data-loggers. For the drive itself, you get a Realtek DRAM-less controller, and a single 1 TB TLC NAND flash chip that's forced to operate in SLC mode (333 GB). On a scale of absurdity, this drive is right up there with $10,000 HDMI cables. Digital audio is stored in ones and zeroes, and nothing is accomplished through an isolated power delivery or clock generation on the storage media. It's nice of the designers to include jumpers that let you switch between the discrete power source and motherboard power; so listeners can see the snake-oil for themselves.
Sources:
Audiophile Style, HotHardware
Most NVMe SSDs have a tiny 2 mm x 2 mm SMD oscillator that's used by the controller for clock-generation. This drive features a Crystek CCHD-957 high-grade Femto oscillator. These oscillators are found in some very high-grade production or scientific equipment, such as data-loggers. For the drive itself, you get a Realtek DRAM-less controller, and a single 1 TB TLC NAND flash chip that's forced to operate in SLC mode (333 GB). On a scale of absurdity, this drive is right up there with $10,000 HDMI cables. Digital audio is stored in ones and zeroes, and nothing is accomplished through an isolated power delivery or clock generation on the storage media. It's nice of the designers to include jumpers that let you switch between the discrete power source and motherboard power; so listeners can see the snake-oil for themselves.
160 Comments on An "Audiophile Grade" SSD—Yes, You Heard That Right
I’m about ready to build a wire coat hanger USB cable and get a oscilloscope to show people, but some still would be convinced their feelings are more right than science and post more about how it feels like it sounds better cause it cost more…..
If your argument is that the influence isn't audible thats fine but a USB cable is still susceptible to the noise in an environment and introducing it into the signal path.
Your ideas are still wrong the further down this rabbit hole you go.
Moreover you assume that digital buffers and bit correction are 100% effective which is not the case. And the accuracy and effectiveness of those measures too is susceptible to being affected by external noise.The test I linked to included both headphones and speakers.
The result; the two lower quality by sources (by spec) were statistically identified as sounding worse, it became particularly more apparent the higher up the scale the rest of the system was. In other words the better the rest of your gear was the more likely you were able to identify the differences.
Which totally makes sense given that most DACs are good and you need a system with enough resolution to reveal a difference, and goes a long way to explaining why there is a large consensus that they don't matter.Of course it is. We live in a analog word (until the metaverse consumes us all).
The point is with digital video the signal is bit perfect until it hits the pixels. If you were to draw an analogy to audio you would need an all digital pre-amp, and a digital amplifier (class D amps are still analog) and only the trandcucers pushing the air would be analog. That simply doesn't exist, the closest thing we have is servo subs even those are driven with a analog signal.
en.wikipedia.org/wiki/Cyclic_redundancy_check
For a DAC/amplifier the power source will introduce more errors than USB will allow before faulting out.
Again you are missing the point, the USB protocol has nothing to do with anything in the analog domain of a DAC yet that is going to susceptible to noise from a digital cable.
Noise counts on the digital side as well, the nanosecond the signal is transformed from digital to analog errors can be introduced by noise.
Try this... put a paper towel tube up to your mobo VRMs while the CPU is working. Even better, run RGB on a pulse mode too. I hear all that through my speakers when I plug most dacs under $200 into my PC via USB.
Worth noting, I never found a cable or dongle that touches it. Got a lot of people so sure it was anything other than what it obviously was though. The only workable solution was a better isolated DAC. Or straight up convert to something like spdif or aes. Isolation should be a natural byproduct, just by virtue of what has to happen to do that conversion.
I’d understand your point of view if we were talking about amps or speakers, but modern DACs are essentially completely digital.
As for the point of ’electric interference’, it of course also applies to the voltage the pixels are driven at, meaning that the brightness of pixels can be affected in the same way as in a shitty DAC the audio output can be affected. Any decent DAC (and TV) is essentially immune to that kind of shit, and the interconnect cable does not matter.
A cable can only shield the signal from outside interference / noise. With something like this no cable is going to solve it because the noise is being conducted through the cable from the PC to DAC's analog output stage. Something is really wrong in a situation like this as you are hearing things that are not even part of the signal. Nothing really to do with how (external) noise can affect things but it illustrates a point of how sensitive the signal path really is even though its "digital".Yeah, it still matters and errors could occur at the output stage of the light source or even in the final conversion of signal to analog and frankly they probably do but our panel tech isn't that great anyway and I'm not sure our vision is at the same level of acuity as our hearing to pick up on things like that even if the panels were good enough.They are not completely digital, they are half analog, its right in the name.
Look at the example above as to how wrong things can go with a DAC. Granted thats an extreme example of noise making it all the way through to the output stage but any kind of noise picked up (not isolated) by the cable can make its way into path and affect the analog domain. Even the digital domain is not immune to noise, errors happen all the time, its are not immune errors to just because its digital. Asynchronous clocks and FIFO buffers largely mitigate the issue but its not eliminated, thats a fact. Now can you hear the difference?, thats the part that depends.
Why would you even connect the USB power on the DAC to anything? As for the USB’s digital signal, it is a BALANCED input, which negates essentially all interference BY DESIGN. You could also use optoisolators on the DACs PCB to get rid of all (theoretically)possible interference from the digital signal, but that is not necessary.
The noise you hear from bad cables when listening to badly designed DACs come from the power supply, not the USB signaling, nor the cable. Use a DAC with separate power supply, and you won’t get the ”cable interference” anymore.
Please, in the future, could you be more specific when describing things like ’the USB signal’, as you seem to mix that constantly with the USB 5 volt power output. The USB power is usually of bad quality, and shielding that does not help in most instances, as it does absolutely nothing to fix the power source itself.
Edit: and just to be clear;
There is no need to shield the power cable (unless you live inside a microwave). Go look at some topping DAC measurements. Do you think that they use shielded power supply cables, to get rid of this interference you speak of? No. And they still get some of the best SNR measurements ever on a consumer product.
the scarlett devices (in the example you quoted) use USB power, and are thus at the mercy of your power supply and motherboard. If they are shit, then you might get some interference. You can fix that by using a powered USB hub that has a stable power supply, or a USB cable with a separate power input of decent quality.
The usb d+ and d- signals are just compared to each other and that comparison then drives a transistor that gives information to the rest of the IC whether it’s a 0 or 1 on a given time. NOT A SINGLE ELECTRON FROM THE DATA LINES IS TRANSMITTED TO THE ANALOG DOMAIN.I NEVER STATED THAT DIGITAL INTERFACES WOULD BE IMMUNE!!!
I even had an example of a TV picture fluttering because of a bad power supply.
What I’m trying to say is that (most) digital signaling cables either work or do not. (There are examples where the devices are given power with the signaling itself, which is IMO out of scope here)
YOU are trying to somehow obfuscate shitty powersupplies with USB cables, which makes zero sense.
EMI based shit can affect only the latter, which is clearly audible if present (USB packet corruption -> missing segments of audio. There is no ’retry’ mechanism in USB audio). I.e. It either works or doesn’t.
Its not "works or it doesn't" thats the fallacy, you never get a perfect reproduction of the analog signal, distortion is present in the entire signal chain and presents as audible quality differences.
The (USB) cable is active component, physics dictates its going to have an influence for all the reasons mentioned already. You can have the opinion that the whatever differences are are not audible, but they exist.Yeah there is. All DACs have a clock that has to be maintained with the source. USB has different ways to go about it depending on the DAC and OS and driver combination though.
Blind studies show that hardly any of ’high-end’ audio crap has any meaning at all when it comes to audibility. USB cables or m.2 SSD:s definitely do not.
The USB cable is not an active component, but a passive one. It has no electronics inside it.
www.techopedia.com/definition/735/passive-component
Physics also dictate that the way you breathe across the globe has an effect on the music I listen to right now. Is that relevant to anyone? No.
You are free to disagree and to
be wrong.
Blind studies with this kind of thing hard to do with proper control. Without proper control you are not really getting any useful data and your conclusions are only as good as the data. I already linked to one that shows pretty conclusively you can pick out the differences between sources pretty easily when the rest of the system is good enough. A cable is going to have less of an influence than a changing out one source for another but if something is prone to producing errors and affecting the sound (which it is) who are you to say that someone can't hear the difference between one cable and another?Yep, you too.
Now that you know that your wallpaper affects your music experience, do you advocate for ’high-end’ wallpapers?
Just by knowing that timing inaccuracies are present does not make them audible. Audibility needs to be proven.Yes, and without blind studies you have no data.Yes, a meaningful amount of people could pick out a crappy 10 year old motherboard audio out of other options. That has nothing to do with USB cables.
Either way its a recent change in how Windows handles USB audio and I don't think its the default even if you have support for it. Unless you know otherwise.Am I supposed to take this seriously? A cable that is prone to interference and producing errors is a real thing.Proven to whom and how? Measurements don't tell you everything and often conflict or show no difference with what you hear to be superior or inferior. We have a long way to go in terms of getting a complete picture with audio measurements. Measurements are a tool to confirm what you hear, nothing more.
Blind studies sure, but its an insane amount of work to do properly. Its a ton of work to even do improperly (not enough controls) and draw incorrect conclusions from it.
From my perspective what x number of people are able to hear from one component change out of a given sample is kind of pointless when there are so many other variables. I'm more interested in why something could make a difference and understanding why and then hearing for myself. Blind studies, measurements, specsheets are interesting but even if those things were completely conclusive it dosn't meaning you or I are going to hear what they are purporting to prove.To that I would just say absence of proof is not proof of absence.Most picked out the onboard audio (which had pretty decent specs) on a pretty wide range of gear. If you read further into the test people that had higher end gear were able to hear differences between the higher-end sources as well.
Are the results statistically meaningful enough to prove it by any scientific standard?, no but that doesn't mean there isn't anything there it just means you need better tests if your goal is to prove it.I don't think thats correct. You need USB Audio 2.0 to support async and that only made its way into Windows 10 in 2017.I think you are missing the point. The conclusions you draw are only as good as the test you conduct, the data you collect and how you interpret it. Its pretty common for well conducted scientific tests to draw misleading or incorrect conclusions through no fault at all in how the test was run. It happens in all the time in much bigger well funded studies where the stakes are much higher than something as trivial as audio.Displays are electrically noisy, thats not really disputed and why notebooks are often not recommended to use as streaming devices.
The idea of 'high-end' wallpapers is absurd because if the display is causing a problem you'd just turn it off. If a cable is the problem you could turn it off by unplugging it but then well.....