Ghmi

You're thinking along the lines of "putting some stuff on one CPU, and some other stuff on the other". That's not how any of this works. The two CPUs are linked and synchronized anyway. What matters isn't CPU1/CPU2, but the number of lanes of each PCI slot and whether they're switched or not.

Slot 2 and 6 are the only X8 PCI ports on this mobo which go straigth to the CPU. The 3 and 4 are switched, 1, 5 and 7 are x16 ports. Also, using a dual CPU board really isn't a very good idea if you're concerned about noise. Twice the CPUs = more than twice the electronic noise. Also, that's not "theory". Unlike you, I'm a practicing engineer. He most definitely won't.

Adding an external power supply to your SSD is going to : - break your motherboard ground plane design, tremendously boosting EMI - introduce a huge ground loop, acting like an antenna so that those EMI can be scattered in your whole room Rule one of mixed electronic design : proper grounding and single path of return for each load. You're only making things worse. Whatever you do, your OS will be running on both CPUs. Your player program can only run on top of the OS.

Send me one of those scams, I'll gladly test it. Or better : I can recommend the equipment you need to do that (less expensive one than the SSD itself) and write you a test plan if you're willing to do the work. One thing I won't do is give money to scammers. A "low noise SSD" isn't insane. It's stupid. It's not even an idea : it doesn't mean anything. The very functioning of a SSD is noise from an analog standpoint. The only low noise SSD you can get is a non-functioning SSD. The manufacturers your are talking about do not have any meaningful experiment to show for a very simple reason : they don't actually have products to sale and test. They're scammers. They sell you the groundless belief that the sound of your system has improved in an untestable way. All digital products with added analog filters are scams based on the fact that the people who buy them are somewhat familiar with analog electronics but have no understanding of how different digital and mixed-signals electronics are.

I am perfectly able to measure this kind of signals and noises. It's far from being as challenging as anything I do for a living. Considering the budget you put into snake oil, you are also able to measure these, you just don't want to. "Eletricity" isn't a signal, and it matters very much how we call signals on the path : digital and analog signals are fundamentally different. They're encoded in very different ways, at very different frequencies, don't suffer from the same kind of noise, and aren't processed nor carried in the same way. It's like saying "it's all just sound anyway, why can't I just use a 2$ buzzer instead of expensive speakers ?".

I'm not an expert in "sound with computers". I'm an expert in measuring very small analog signals with computers. Which is way more constraining, but has an advantage : I don't have to relay on my ears and subjectivity to "see" the effect of electronic components. I am equipped to measure this effect directly at every stage of the process. When I talk about a signal and a noise, I have a clear definition of said signal and noise, and am able to directly measure it. And based on this expertise, I can tell you that : - your aren't experiencing nor testing anything. That's not how you do that. If you want to test the effect of a component, you start by measuring this effect directly on its inputs and outputs before even looking at the "final result". - what you say makes absolutely no sense whatsoever and shows that you have absolutely no clue about what you're doing. You don't even remotely understand what sound is and how "sound with computers" works. The 700$ SSD is a good example : 1. there's no sound nor noise at the SSD level. The signal you call sound is first generated by your DAC. It simply doesn't exist before that stage. Your music at this stage is purely digital, and noise in a digital signal isn't defined and filtered in the same way as in an analog signal. Noise at the digital level is due to jitter or bit changes. The capacitors used as a filter on the scammy SSD are analog filters : the only thing they can do in terms of digital noise is make it MUCH worse. 2. data doesn't go "through PCIe and directly to the CPU". It goes from the SSD to the RAM, then the CPU cache, where it's processed. If you're worried about noise in the analog sense of the term, the CPU cache is FULL of it. It's on the CPU die, that's the worst place of the whole computer and anything you can do is negligible compared to that. 3. If you're worried about digital noise, just get ECC RAM and clock timings right. Probably won't make much of a difference considering how reliable it all is nowadays, but that's how you do it. If you're worried about analog noise at the digital stage, well, you're worried about a drop in an ocean. The very functionning of your computer IS analog noise and there's nothing you can do about it unless you shut down the whole computer. Just isolate your analog circuit and shield it properly. The only things you need to run on a separate stable power supply is the DAC and amps. 4. having a separate power supply for an SSD is a whole world of pain. You're not just adding a new power source, you're also adding a new ground. What's the potential difference between your grounds ? How are they connected together ? Did you use the star point grounding technique ? This has a huge potential to create more digital and analog noise, or even to completely mess-up the SSD. What you do isn't "empirical". It's "subjective" in the pejorative sense of the word. You don't even have the beginning of a test : you just buy expensive snake oil and decide that it works because you don't want to admit you've been scammed. If you're ok with spending enough to buy an actual music instrument in devices which can only degrade your system, fine. But you've been warned : you're buying expensive scams which in the best case have no effect whatsoever and in the worst case make your computer less reliable, much less efficient and less safe.

Also a task which doesn't just come down to buying expensive stuff. But an interesting task, at least. I work with small analog signals for a living, the kind for which the location, shape and thickness of pcb traces matters, and which sometimes requires not just filtering caps but running on batteries. When I see two caps slapped on a 700$+ ssd, it makes me nuts. Exocer : don't worry too much about us. I'll be able to enjoy quite a lot of true analog live music with 600$ not spent on 2 caps on a SSD, or 1250$ not spent on a fancy 20 MHz clock with extremely low jitter measured at 10Hz.

The only effect it can have on a music library is the slight delay before loading a file or between files. Unless you put your library on a slow SD card or USB key, latency and bandwidth won't affect the music itself, as the file will be buffered faster than it's played. Based on yamamoto2002's tests, it's in the few hundreds µs range for SSDs and ~5ms for a standard sata drive. Even a 5ms delay before a song starts would be quite hard to notice.

Off course it doesn't explain those differences in sounds quality : none of the digital parts of a computer play any role in sound quality. It's all up to your analog circuit and your subjectivity. If you notice any difference while adding filters to digital parts, it just means that at least one of those two isn't as reliable as you think it is. Maybe your analog setup is messed up and your sound quality just varies randomly. Or maybe it's all in your head and you just really want to believe that your ears are better than they actually are, some kind of snobbism, or a need to justify the purchase of expensive snake oil. Or both. This being said, I don't think a real audiophile would care this little about how his setup actually works and end up treating it like it's some kind of crystal power magic.

How computer hardware and software works isn't really all that "subjective". You can just check your OS documentation and see how RAM and virtual memory are handled. That's how it's done, not through some wildly unreliable "testing". Spoiler : unless you're using a very exotic OS, all of the data goes through the RAM before being ultimately handled by the CPU. How the music sounds to you is subjective, sure. How a computer works, not so much.

Well, RC filters for RAM have been available for quite some time already : https://www.audiophonics.fr/en/alimentations/elfidelity-axf-75-power-purification-pc-hifi-memory-ddr4-p-11163.html EVERY possible niche for audiophile scams is taken already. You can add RC filters or gold basically to every part of a computer and audio setup, and there are way enough people in the audiophile community who have more money than understanding in electronics to buy that. It's kinda sad, but well, you can see that as a form of wealth redistribution. The part which bugs me is the added power consumption and fire hasard, particularly when people start replacing switching power supplies with linear power supplies, having an unknown number of separate power sources connected in random ways etc.

This thread is a nightmare from an electronic engineering standpoint. All the efforts spent on the digital parts of the setup (most of them futile or even detrimental), to end-up with a non-isolated DAC ?!? Using regulators instead of a proper regulated switching power supply is heretical. These regulators have horrifically low yields and won't handle fast changes in power draw properly, generating MORE noise on power rails and maybe even on the ground than any decent PC power supply. The gold and silver things are absolutely unnecessary. They won't make any difference on noise for cables of this scale (it only makes a difference for very thin wires). And in the end, it's always the same issue : you're working on digital parts which job basically is to generate noise. From an audio standpoint, everything digital IS noise and what you may be able to filter out without turning your PC into a toaster oven is infinitesimal compared to the normal functioning of a computer. Just. Isolate. The. Analog. Part. Of. The. Circuit. Use an all analog amp, any connection between the PC and DAC which allows isolation, proper shielding, stable power supply for the DAC and amps and that's it. Don't try to scoop out some tiny part of an ocean of noise : get your DAC and amps out of it entirely. You're basically saying that your house is too hot, but that setting up a fancy cooler in your garden makes you feel cooler. Good for you, but that's not just a subjective improvement : it's also a huge waste of money and energy for the result you get. Isolate your house and put the cooler inside instead.

You can actually test the effect a drive has on audio data itself : write an audio file on it, read it back, and compare the binary sequences. If there's even a 1 bit difference, that's not normal : any properly working HDD will give you exactly what you write on it. Any difference in sound comes from the analog part of the circuit. Not the digital part. The analog circuit, DACs, amplifiers, cables, speakers are sensitive to noise and distortions. Not the HDD, RAM and CPU.

The disk does not produce any music. The DAC does. Up to this point, there's no concept of music or sound : it's all binary data and none of it will change with filtering, not a single bit. What can happen though is either EMI or digital noise creeping through the supply during read operations. In this case, the SSD is the least of your worries: your CPU, GPU and RAM are basically noise-production machines compared to a standard SSD. Particularly the CPU and GPU, which will generate regular voltage drops on power rails which carry several amps , which you cannot filter out. Filtering on the digital side is like trying to empty an ocean with a scoop. Focusing on the digital side makes no sense at all : it's all noise anyway, whatever you do. That's actually not unusual for digital noise to creep into small amplified signals, most PCs show signs of that. But the way we deal with that in engineering is by isolating the (external) DAC and amplifiers. That's the real solution: put all your analog circuit in a faraday cage and on a separate stable and filtered power supply (preferably a dedicated battery). Instead of trying to filter out a random potential source of noise within an ocean of noise, isolate the analog circuit from it all. It may not be as cheap as the audiophile SSD, but it makes a world of difference, not just a subjective difference only few will feel. If done properly, you can even get the same delivery as a fully analog system, providing you use high resolution audio files and DACs.

What we can hear doesn't even come close to what can be measured nowadays, in terms of signal AND noise. Human senses are nothing to write home about. If you can hear that some sounds are noisy or distorted, it WILL show up on even an entry level spectrum analyzer. And you can trust me on that : measuring small signals which are unaccessible to our senses is a huge part of my job as a researcher. From measurements down to ADC conversion.