There is always a lot of talk on forums about subjective Vs objective; listening Vs measurement and there seems to be a recent spate of it here (although maybe that's just my perception as I haven't visited in a while?).
As a hopefully more refreshing way for people to air their views in this area (without bickering) I thought it might be useful to gather people's thoughts/opinions about what they feel makes a more plausible audio scene (I'll use "realism" as shorthand) in their listening environment & what possible measurements could be used to evaluate this?
I'm not talking about the endless debate about how accurate the reproduction system is vis-a-vis what has been recorded - instead I'm borrowing this phrase "plausible audio scene" from Linkwitz :
Now rather than speakers, my focus over time has been on power supplies & in my experience improvements in this factor alone have resulted in major improvements in "realism". This has not only been in DACs & amplifiers but also in computer transports (others have mostly done these mods). Almost every time an improvement is made in the PS in these devices (or selected parts of these devices), a similar improvement in the sonics is evidenced - a more solid sound stage, a more relaxed listening experience, more body & richness to the sound, essentially a more plausible audio scene. The old adage is true - we don't recognise certain types of distortion until we hear the absence of it.
What I take from this is that it seems stability of power supplies is a crucial element & the most crucial part may well be the grounding & not necessarily the power. I've noted sound stages collapse when ground noise has been introduced. I've noted apparent detail being increased with ground loops only to discover it was false detail (distortion). I've noted improved realism with PS treatments to computers that act as transports. An increase in realism in moving from the stock PS to a linear PS to multiple individual linear PS powering the 5V, 3.3V 1.8V feeds to using batteries (with linear PS trickle charging) to independently powering the USB port (even though the external USB DAC is self-powered & makes no use of the VBUS 5V power USB line).
All these areas where improved PS has resulted in improved performance (even though in theory, they shouldn't) has led me to a tentative conclusion that noise & stability might be the determining factor in all of this. This is by no means conclusive & I'm open to changing my mind but I see very low level noise fluctuations as the possible explanation for the above observations. BTW, I also see this as the possible explanation for how software playback operates. I don't know of any psychoacoustic studies that have been done into the level at which noise modulation is detrimental to our perception of the audio scene?
The final thing to say is that measuring such low level noise in a dynamic music signal (& this is the correct place to measure it) not in a static, repeating signal is very difficult. Once that can be done then how this noise is modulated over time is the next difficult measurement, I believe.
Thoughts?
As a hopefully more refreshing way for people to air their views in this area (without bickering) I thought it might be useful to gather people's thoughts/opinions about what they feel makes a more plausible audio scene (I'll use "realism" as shorthand) in their listening environment & what possible measurements could be used to evaluate this?
I'm not talking about the endless debate about how accurate the reproduction system is vis-a-vis what has been recorded - instead I'm borrowing this phrase "plausible audio scene" from Linkwitz :
Now rather than speakers, my focus over time has been on power supplies & in my experience improvements in this factor alone have resulted in major improvements in "realism". This has not only been in DACs & amplifiers but also in computer transports (others have mostly done these mods). Almost every time an improvement is made in the PS in these devices (or selected parts of these devices), a similar improvement in the sonics is evidenced - a more solid sound stage, a more relaxed listening experience, more body & richness to the sound, essentially a more plausible audio scene. The old adage is true - we don't recognise certain types of distortion until we hear the absence of it.
What I take from this is that it seems stability of power supplies is a crucial element & the most crucial part may well be the grounding & not necessarily the power. I've noted sound stages collapse when ground noise has been introduced. I've noted apparent detail being increased with ground loops only to discover it was false detail (distortion). I've noted improved realism with PS treatments to computers that act as transports. An increase in realism in moving from the stock PS to a linear PS to multiple individual linear PS powering the 5V, 3.3V 1.8V feeds to using batteries (with linear PS trickle charging) to independently powering the USB port (even though the external USB DAC is self-powered & makes no use of the VBUS 5V power USB line).
All these areas where improved PS has resulted in improved performance (even though in theory, they shouldn't) has led me to a tentative conclusion that noise & stability might be the determining factor in all of this. This is by no means conclusive & I'm open to changing my mind but I see very low level noise fluctuations as the possible explanation for the above observations. BTW, I also see this as the possible explanation for how software playback operates. I don't know of any psychoacoustic studies that have been done into the level at which noise modulation is detrimental to our perception of the audio scene?
The final thing to say is that measuring such low level noise in a dynamic music signal (& this is the correct place to measure it) not in a static, repeating signal is very difficult. Once that can be done then how this noise is modulated over time is the next difficult measurement, I believe.
Thoughts?
