Lowering the noise floor

I have spent more than $10k on treating both my home and system to reduce all types of contributors to the "noise floor": shot noise, EMF/RFI (from land, sea or air), EMR, grounding issues and so on (all Alan Maher Designs). My system only cost me $6.5k, but that $10k was spent gradually over the last 8 yrs, starting off very slowly and only progressing in price as confidence increased accordingly. But, for me, it just goes to show how important the role of this kind of problem does in fact play in music reproduction. I kept pursuing these solutions because I kept finding the ones I that I was using were working...if they didn't I would've given up on all that long ago. Everything improves. The modest $6.5k system I have is certainly modest no longer.

I could have spent that $10k on better gear, but it actually, I think, would've been a mistake, really. Sure, I could have had the inherent *potential* for better sound, but I would still be dealing with all the usual suspects that everyone else seems to be dealing with: EQ that doesn't really work well enough to truly solve problems, digititis, ss sound, "harsh" or "bright" recordings, best listening only at 4 a.m., upgraditis, insufficient timbre, tonal color, presence or sound staging, dependency on things like $$$ cables, critical speaker positioning, room peaks and nulls/room treatments...it even tamed how noisy that fridge is...all of that and much more has been either greatly reduced or eliminated. And that doesn't even take into account the transformational gains in both sound quality and presentation, all the way around.

Finding a way through this particular issue, for me, has proven to be the single biggest factor in finding that actual point at which I'm truly done with changing system hierarchy - exit strategy without compromise. I believe something like that might be the missing link for a lot of people. I suspect for most people that if they fail to investigate this aspect of music reproduction in their room space, they may be subject to the need to continually upgrade - and in vain perhaps because they might have reached their goals with less expensive gear, as I did - not by settling or compromising, but by finally solving all the right problems that were the ones that were actually standing in the way all along - and that would have continued to even after upgrading.

I think it's just possible we don't Rilly need the 'latest and greatest brand X', we just may need to focus more on what the longstanding audio problems actually are and how they can be fixed and what it might take to do that. Not for everybody for sure, but I can say that it was well more than worth it for me. No going back here. 

Cheers,
John

Once again a lot of voodoo science, and some suggestions that may have to do with sound quality but not necessarily with noise levels.
If you want lower noise levels (i.e. greater S/N) to increase dynamic range and detail resolution there are a few things you can do:
1 lower background noise (see my post above). In real life this is probably the most effective option because it is backgroud noise that masks the usually pretty good S/N performance of modern electronics. There is no point in listening to low noise gear if you are living next to an interstate with the windows wide open.
2 If you still think you can benefit from lower noise levels, look at the parts of the chain that have the worst S/N: FM radio, Phono inputs and rca interconnects, in that order.
FM radio is terrible with typically perhaps some 70 dB S/N, and there is nothing really you can do to get it close to the quality that can now be achieved by e.g. internet radio.
Vinyl and the necessary phono inputs are much worse than digital sources and line level inputs. Go digital and you have a far better S/N ratio. On most amplifiers the difference in S/N between phono (80 dB if you are lucky) and line level inputs is at least some 10 dB, which is not surprising given the amplification that is required, combined with the massive RIAA equalization.
Using balanced vs rca interconnects usually improves S/N by some 5 dB or more. Since these are typically used between a DAC/preamplifier and a power amplifier, we are already dealing with noise levels better than some 95 dB which is so good that the improvement may not necessarily be audible. But if you want the best, this is the way forward.

willemj

FM radio is terrible with typically perhaps some 70 dB S/N, and there is nothing really you can do to get it close to the quality that can now be achieved by e.g. internet radio.

That's silly. There's no way that low-resolution, lossy-compressed Internet "radio" can compare to the resolution of the best FM radio. Of course, the fidelity of the best Internet "radio" can be better than an awful FM signal. But the best FM signal - on a good tuner with a proper antenna - can offer incredible fidelity. That so few stations achieve this degree of fidelity accounts for the misguided notion @willemj states here. 

No it is not. FM has serious limitations, even if internet radio has as well, and they are of course hard to compare directly. FM once was a big improvement over and above MW, but it has quite a few limitations. First, there is S/N, second there are relatively high levels of distortion and third there is only limited channel separation. As a consequence, no FM broadcast is done without (these days automatic) dynamic compression, in the best cases only to a quite limited extent, but quite often rather invasively.
Compare that to internet radio. There is no need for dynamic compression, nor does it suffer from any of the other compromises inherent to FM. On the other hand there is of course data compression, sometimes rather badly, depending on the bit rate that is chosen. However, the algorithms that are used are marvels of psycho acoustic technology with exceptionally small reduction in sound quality, particularly at the higher bitrates. BBC Radio 3 is broadcast in 320 kbps and their and others' research has shown that this is virtually indistinguishable from full Red Book CD (there are quite a few blind tests on the internet to take yourself - few people can identify above 256 kbps). Even so, the BBC is now experimenting with full red book CD internet streams in FLAC (i.e. some 600 kbps depending on the type of music). So, the comparison is between the kind of signal degradation from the FM technology and the degradation from data compression of the digital stream. In my experience (with  an excellent FM tuner and good signal) the sound quality of internet radio at the higher bit rates that are increasingly common is significantly ahead of even good FM. That is the conclusion of the BBC, and I could not agree more. Moreover, the good news is that bit rates have been and are going up all the time.

In all seriousness, the greatest contributor to the noise floor of your system just might be the refrigerator in your kitchen when its compressor is running.  In my old rental house, I would unplug that janky fridge for listening sessions... right up until my wife found melted ice cream in the freezer.  Now we've got an LG that's whisper-quiet.

Likewise, your HVAC system, when circulating air through its vents, could be an order of magnitude noisier than your power supply.