No, it really isn’t. You do not have the physiological tools to hear that far down into the mix. Otherwise you’d be listening to the vibrations of air molecules.
It’s important to note that in a house the baseline noise level can be in the 50-60dB range so perceiving noise at -120dB when playing music at a reasonably loud volume (90dB) will be extremely challenging.
1 Like
That’s understatement, right there 
@pwright92 Noted- but when you listen to music on, say, a set of Shure SE846 In Ear Monitors, which isolate the listener from environmental noise, you can hear everything.
50-60dB is remarkable for a baseline noise level. I whipped out my Radio Shack SPL meter and my JL Audio iOS app and got a ~35dB ambient noise level- and that’s with a open concept living room. (I can hear the fridge humming along too.)
Further, with FLAC/CD’s 96dB of signal to noise, I am already swimming in 35dB ambient, thus I would have have to crank it up to like 131dB (!) to even hear the noise floor on a CD/ 16bit FLAC file.
You make an excellent point with your suggestion of 90dB program volume and 60dB ambient volume, you’d be just fine with anything from 1955 LPs all the way to today’s downloadable lossless files. Cars? PPFFF!
@anon55914447, Mark, you are unclear, to me, here. “physiological tools??”
I have a background in analysis, math, and tools. More often then not, I use the numbers to justify a reason/hypothesis/viewpoint, eyes glaze over; even the smartest of the smart wander off to toon town. Big decisions are made in this world through intuition, instinct, and intangibles… …and if we had it all figured out we wouldn’t have to discuss it.
Yes just to add my 2 pennies worth here. I have a raspberry Pi with a switch mode PSU, I have a Devialet 400 (soon to be 800) which has a switch mode PSU and Magico S5 speakers. If I turn the volume up to +30 on the Devialet with nothing playing and put my ear against any of the drivers there is utter silence. Considering that I run an IT business from home with servers, router (with SM PSU’s) NAS PC’s with noisy internals all running on the same mains network I begin to doubt the noisy PSU assertion.
If I actually played music at that level in my room I would literally be pinned to the wall with the sheer volume of sound, even at 0 on the volume my wife comes in, gives me one of those looks and shuts both the doors!!!
The background noise level is about -36 dB according to my Netatmo.
Well, in my experience noise does not only exist as a noise floor, and this can be quite low when nothing is playing. It can exist as well, or let’s say it develops, when something IS playing and puts itself onto the audio signal. This is hard to measure I guess but is audible as part of those artifacts we all know.
For me the measurements that we can do proof if something is wrong, but do never proof if something is sounding good. The latter judgement is coming from the ears, or more correct, our brain. And this works differently from one to another, biased on different experiences and expectations, even depending on a given moment, that’s where all the different opinions have its origin.
I should add that there’s no discernible difference when using these vs the switching PSUs I’d previously used. The only reason I decided to use the linear PSUs is I found them in a cupboard recently, having purchased them many, many moons ago for use with Squeezebox Classic players and never got as far as trying.
In reply to my own post above I want to add:
Whether or not someone’s hearing an improvement in reduced noise floor of the source or any other component depends on the rest of the components used. In my experience the noisiest component determines the noise level of the whole chain we are hearing.
In other words: if you try to improve your chain on component A while component B being already noisier within itself than component A, you will not have much success. That’s why some cannot hear a difference for instance of a “insane high end” cable while others do. It’s a matter of whether the chain able to scale the difference. It’s as strong as its weakest link.
I moved my Pi/HifiBerry over to a battery (one of those portable power for your phone units). The change in power made a very noticeable improvement. One of the things that the better manufacturers do is employ the newer power regulator chips in their equipment. These chips are magical for DC, about 0.8uV ripple compare with 10mV on a crappy wall wart. These regulators are actually quieter than some batteries. iFi is using them in their iPower devices as far as I can tell and I’m certain that companies like Devialet is using them in their power circuits. An example of the regulator is the Linear Technology LT3042 which cost about $7 in single quantity. At that price they should be showing up all over the place.
Most “insane high end” cable is just window dressing and marketing, all “insane high end” cable is targeted at gullible people. At least use a more credible example to make your point.
My house was built circa 1875 so 50’s LP are ultra-modern here! 
1 Like
@o0OBillO0o, what I mean is your ears and brain are not biologically capable of detecting signals at -140dB.
1 Like
I do not disagree with your point, but that doesn’t mean that there are no differences in cables. For sure there are. I will give you another example that has nothing to do with cables and hopefully explains my point a little bit:
Just some months ago I was looking for a new DAC replacing my vintage Mark Levinson 30.5 DAC. I’ve listened to two other DACs for around two weeks in my chain. Those were the Totaldac d1-dual and the MSB Analog DAC with external power base. With the Totaldac I had a separate Totaldac reclocker which reclocks the signal of any source. I’ve routed the reclocker to all three DACs including the ML. While the best improvement in using the reclocker was with the Totaldac d1-dual, the improvement with the MSB came close (with the d1-dual being better overall), both improvements were significant and easy noticable. And here comes the point: the reclocker didn’t make any noticable difference to the ML. The transparency within the ML itself was limited, not able to scale the better input signal.
I think this drifts a little bit off-topic now and I apologize for that, but it’s a good example that it makes not much sense to discuss about “how noisy is my PC and do I hear this” when the bottleneck might be somewhere else.
Ok but in my case I hear zero noise at a volume level that if playing music would rival a Saturn 5 taking off, so where is the bottleneck here?
I don’t know your chain and I haven’t heard it, so of course I cannot tell. But keep in mind that noise can appear when a signal is present (playing music) as I wrote before, means when the circuit is actually processing audio data, what it does most likely (almost) not when there is no signal.
Ups sorry I’ve missed that. Congrats to your system, although I’m not familiar with the rasbperry Pi.
What I can tell you is that since I’ve improved my system regarding clean power there was is no difference in apparent noise just with turning volume up, no music playing, and getting my ears on the speakers as you’ve described. But there is a clear difference when music is playing.
But all we’ve got to go on is an anecdote. It’s great that it worked for you but you can’t make global assumptions based on your experience.
Who knows, perhaps you had mains problems, or like the vast majority of people (inc. me) you are susceptible to placebo and confirmation bias.
What I can tell you unequivocally is that you wouldn’t be able to detect the difference between a -140dB noise floor and a -132dB noise floor.
1 Like
You keep repeating supposed facts that explain what others can’t hear. Well, apparently you are looking at the wrong facts.
Lots of examples where improving PC signal quality makes an obvious difference in SQ, and it isn’t something at the level of -120db, but much less, definitely within the range of hearing. Even you could hear it if exposed to it.
So maybe entertain the possiblility that the conceptual box you are putting everything into is incorrect, and something is going on that you don’t understand or have the tools to define.