Who wants to help me understand Imaginary Noise?
Oct 9, 2016 at 4:44 PM Thread Starter Post #1 of 5


Formerly known as Denon5220
Jan 14, 2010
Top head fi members here have said the science behind SNR of 124db means it is essentially useless since you will become deaf if you actually listened to music at 124db and nobody does that. And that the only reason we don't have 200db snr chips is because we cant listen to 200db sounds, this audiophile went onto say the SNR is basically eliminating the hiss when you sit in a silent dark room and turn the volume up all the way. And this is apparently what higher SNR equals to.
BUT here is my problem I had a Xonar DX in the past and foolishly sold it for a Xonar DG "with headphone amp" which was just marketing since apparently the DX does have headphone amp just weaker.
Anyways there is a noticeable difference between the DG and the DX. AND the DX is 112db SNR while the DG is like 103db SNR.
There is no hiss or noise in either when volume is turned up yet somehow the DX sounded clearer. Does this mean an imaginary noise exists that we cannot hear? how would you all explain something like this?
Also get this my PC also has a fancy shamcy Asus Crystal Audio 2 system, uses fancy names like Japanese fine gold Audio Capacitors things the Xonar DG does NOT infact the DG uses cheap Taiwan capacitors. And YET the Xonar DG sounds significantly BETTER than the onboard. LOL
Help me understand what the hell is happening here it is clear to see the onboard has a type of imaginary noise even with all its fancy marketing of fine gold caps etc
Oct 9, 2016 at 6:32 PM Post #2 of 5

NA Blur

Headphoneus Supremus
Feb 8, 2011
You are picking up on some of the perceived idiosyncrasies that exist and are audible, but without proper training or knowledge are hard to explain. Let's start with the SNR you posted for each unit which is probably the actual SNR for the chip not the measurement taken through the output with the unit on and volume at some nominal level. I can't say this for certain, but as you know the human ear maxes at near -130dB for its absolute SNR so anything beyond that is completely inaudible. As you may suspect the SNR is not telling the whole story as to what you prefer one unit over the other. There are other aspects to consider like the audio damping factor, harmonic distortion, THD% + Noise, and other measurements. In may be as simple as the DAC chip and filtering is different between the two units. One major thing that alters the sound is how the output capacitors are used int he amp circuitry. You can take a look at this video which will give you some idea what I am talking about. The coupling caps really change the sound quickly. So if these caps are not the same in both units you are already getting some audible differences without even talking about distortion that may be taking place.
This is why it is important to include other measurements, which the DG and DX may already have listed, beyond the SNR that may clue you in on what is going on. Also, are there any software differences between the two?
Your comment about cheap caps rings true with audio in general. More expensive does not mean better sounding or better measuring. There are a ton of factors that go into a design and as you can tell the better sounding gear may be in fact less expensive. For now take a look at the other specs and see if you can find a difference that you think accounts for the sound changes between the units.
Oct 9, 2016 at 6:32 PM Post #3 of 5


500+ Head-Fier
Apr 3, 2012
The dx doesn't have a headphone amp. All dacs usually have some amplification, but a headphone amp implies an additional amp stage beyond that. And this goes for external dacs and headphone amps too; you'll notice people don't plug in their headphones to a standalone dac, but often use a separate headphone amp as well.
On the asus cards with headphone amps (dg/dgx, st/stx), you turn them on by setting the analog out to headphone in the asus driver panel. There are gain options as well, and you can hear a pretty big difference in loudness between the amp on with the highest gain and the amp off.
Oct 9, 2016 at 6:35 PM Post #4 of 5


1000+ Head-Fier
May 9, 2014
There's much more to it than just SNR. SNR only tells you the noise level when music in not playing, so it's not really useful for explaining why your various sound cards sound different. There are numerous other parameters such as dynamic range, THD, IMD, impedance, frequency response, etc. which could explain audible differences.

Users who are viewing this thread