Even if frankly I'm not sure I understand the real sense of your question, I think I may answer with "no".
No, I don't know what you mean with "based".
No, it is the effect of any technological efforts that is reduced to inducing a sound deception compared to a natural event in the listening to. IMO
It is the listening to that is the point of my conversation because "listening to" is just a human activity and it knows nothing about technology.
Something really happens when a person sits on his favorite armchair, he turns on his stereo-system and then he listens to.
It is only at that precise moment that the magic occurs.
A reference: https://courses.washington.edu/psy333/lecture_pdfs/chapter12_SoundLocalization.pdf
Any perception can not be formulated with 100% accurate.
Example: perception of temperature. Here, many people consider room temperature about 21 degree Celsius is cold.
But in country with 4 season, usually they consider it as hot.
It depend of habit.
Any perception can not be formulated with 100% accurate.
Example: perception of temperature. Here, many people consider room temperature about 21 degree Celsius is cold.
But in country with 4 season, usually they consider it as hot.
It depend of habit.
No need for thought experiments while actual values can be measured. 🙂
Effect of one channel load on the CRC PS (33mF – 0.33Ω – 33mF) output at 1W/8Ω at 1 kHz. There is about 20 mV ripple @ 100 Hz + harmonics. Load produces some 2 mV rms @ 1 kHz rails modulation. It is true that drum kicks or other LF signals would produce more rails modulation, as CRC PS impedance rises toward low frequency and I measured more there, but soundstage is not at the LF region.
Effect of that voltage modulation on the second channel is close to nothing. Even if amplifier would have zero PSRR and all rails modulation would be passed to second channel output, that would produce signal more than 60 dB below the first channel. As amplifiers usually have 30 – 100 dB PSRR, signal in the other channel will be 90 – 160 dB below the first one. Contrary to the claims of various superhumans present in abundance at this forum, nobody can hear that.
Case 2. seems to be counter the usual perception that amplifier with 2 separate power supplies ‘sounds better’ as difference between channels would be larger than in case 1.
Though, ‘sounds better’ was always general assessment with no specific mention of soundstage. I believe that ‘better sound’ with the two PS, so called dual-mono, is simply from improved capacity or better transient performance under load. There was an interesting observation about practical experiment with increasing PS capacitance up to extreme levels, mentioned here (post #21):
https://www.diyaudio.com/community/...or-solid-state-power-amps.374100/post-6706920
IMHO, ‘holographic soundstage’ is first and foremost created by loudspeakers, second comes room acoustics, remote third place belongs to everything else. DAC, phono, preamplifier and amplifier. They must be simply not bad enough to spoil the reproduction.
I have a good soundstage illusion, where sound emanates from wide and deep stage several meters behind loudspeakers, with artist and instruments being at various positions, depending on how much sound engineer was playing with recording. Loudspeakers simply don’t exist. Sound is never perceived to come from them and that effect is property of loudspeakers. All supporting equipment is just providing good enough signal. That is enough of holographic reproduction for me.
Comparing the sound reproduction with two pretty different amplifiers:
- 0.03% distortion @1W with H3 at -20 dB to H2, SR 12 V/us, FR 87 kHz, common regulated PS with rails modulation in the uV range under load
- 0.002% distortion @1W with H3 at -20 dB to H2, SR 100 V/us, FR 820 kHz, simple common CRC PS for both channels with amplifier PSRR 76 dB.
Last edited:
Thank you for thorough post tombo56!
ps. my intuition also says that the amplifier needs to be reasonable, good enough performance with enough headroom so that the operation is problem free enough with typical listening levels and above for occasional fun time, and all ill effects should be trumped by issues with loudspeaker and room and positioning.
On the other hand I've recently learned that amplifier output impedance is part of same circuit with loudspeaker (voice coil) and has effect on distortion current made by loudspeaker back-EMF voltage. But this is not fault by the amp and all we have to do is address it with some other tricks, like reduce driver generated current by choosing big enough loudspeaker system with good enough drivers to have less excursion and related back-EMF and then increase driver load impedance in crossover and distortion current is now reduced never minding amplifier output impedance.
Yeah, thought experiments are fun but do not include how the stuff is perceived 🙂 It might be so that even if the thought experiment appears to have the Case 2 worse with the shifting phantom center perhaps hearing system perceives amplitude modulation on stable phantom image (case 1) worse than slight momentary shift in phantom image. Not sure how much localization relates to phantom sources but hearing system seems to set localization based on transients and hold it until better cues come about, for this reason the slight shift in pink noise location might get unnoticed as brain already figured out its on the center and holds the perception steady🙂 Anyway, very interesting stuff and I'm very glad you picked it up as there is now some material to chew on.
ps. my intuition also says that the amplifier needs to be reasonable, good enough performance with enough headroom so that the operation is problem free enough with typical listening levels and above for occasional fun time, and all ill effects should be trumped by issues with loudspeaker and room and positioning.
On the other hand I've recently learned that amplifier output impedance is part of same circuit with loudspeaker (voice coil) and has effect on distortion current made by loudspeaker back-EMF voltage. But this is not fault by the amp and all we have to do is address it with some other tricks, like reduce driver generated current by choosing big enough loudspeaker system with good enough drivers to have less excursion and related back-EMF and then increase driver load impedance in crossover and distortion current is now reduced never minding amplifier output impedance.
Regarding localization, I practically observed that I can perceive 0.5 dB difference between channels as a position shift. It was possible with use of the Muses 72323 volume control, that has 0.25 dB steps. 1 dB difference was immediately clear and for 0.5 dB, careful listening was required.
So, my take is that there should be at least some 0.5 dB amplitude difference to perceive amplitude change as soundstage change. However, time delay and phase at various frequencies should be more critical. It seems that loudspeakers designed as time aligned one-point source, fare better at the soundstage illusion recreation. Coaxial loudspeakers like Tannoy or Fyne Audio are good in the soundstage department. I suspect that Tannoy Gold’s, not the F6, are responsible for what Bare describes. I have some sounds reproduced at position far in the house backyard. 🤣
So, my take is that there should be at least some 0.5 dB amplitude difference to perceive amplitude change as soundstage change. However, time delay and phase at various frequencies should be more critical. It seems that loudspeakers designed as time aligned one-point source, fare better at the soundstage illusion recreation. Coaxial loudspeakers like Tannoy or Fyne Audio are good in the soundstage department. I suspect that Tannoy Gold’s, not the F6, are responsible for what Bare describes. I have some sounds reproduced at position far in the house backyard. 🤣
Nobody has mentioned cross talk between channels. When I rebuilt my HK CIT I preamp several years ago
I changed the power supply from the RC voltage drop to FET regulators. This reduced my crosstalk a 1Khz
from -40db to -75db and made a great improvement in soundstage.
I changed the power supply from the RC voltage drop to FET regulators. This reduced my crosstalk a 1Khz
from -40db to -75db and made a great improvement in soundstage.
https://www.t-linespeakers.org/oddsends/biamp/Index.html
On bi-amping, but the first diagram is illustrative.
If one channel perturbs the other, smll bits of information critical to producing the illusion of soundstage/image are masked, lost.
dave
Isn’t that exactly what tmuikku is taking abou tin the part of his post i just quotes?
dave
Huh, what?
Amplifier at the clipping level as an example of one channel affecting other and ‘losing small bits of information’?
What really happens, outside the extreme cases, was measured and presented.
There is no crosstalk through power supply to speak of, except in case of circuits with zero PSRR. Any good designer would use adequate PS there, better than simple CRC.
I dont like how often I forget to turn the damn things off, an event happening across multiple decades of audio systems where I happened to own tube gear. Every time I have a tube system, the tubes have far, far more hours burning idle than they do listening to music.
One of the reasons I use class D amps is because I can forget to turn them off indefinitely and it doesnt matter! In one system, the PC running Daphile is on indefinitely, the analog power supply doesnt even have a power switch and the amp is powered on continuously.
Without an audio signal present, its outputs go tri-state, i.e. open circuit.
I have a few tube amplifiers, but keep them around for circuit experimentation and nostalgia, as I grew up in the 60's. The only way I could even connect one into my current system architecture would be to drive it with the speaker output of one of my class D amps...
Good point! I have read from the late Allen Wright that a lot of tube gear is actually designed to be left on, believe it or not. Apparently it stresses the tubes more to be switched on and off all the time than it does to just leave them on. Either way... much less of a hassle to go class d.
Tube amplifiers - for me - would need an intelligent tube management functionality, where some circuit would monitor what's happening and be able to recognize "This Guy Did It AgAiN" and automatically turn things down (bias, B+ voltage values...) to put the amp in a state it could endure being energized indefinitely. The filaments could still be lit, even partially, in such a state.
Acoustat servos are in standby at 1/2 operating voltage until the amp senses signal. Mine(2 pairs) have been plugged in operating this way for years without issue except for one 6HB5 that quietly died about a year ago which I replaced. They function as they should.
It's not about supply voltage modulation. Advantage comes from separate grounds of the dual supply.
You should accompany your claim with some technical explanation. 😉
However, you quoted wrong part of discussion, as quoted text is related to the question if rails change under load could affect stereo image.
I second that. 3d image with the center locked in in the middle so much that i can actualy draw a picture of it. 2x monoAmps fed from a Dorati DAC. The speakers (Fane 12-250TC) very agresivly toed in. Massive sound stage. Sound pleasant even at very high volumes with lots of detail and air around instruments.
Hello,
I had a similar question years ago and I think I can answer it the most directly. I have personally found that matching the components for each channel as precisely as possible led to the best possible soundstage. That has been true with both my solid state and tube amps. I took extra care to make sure all the resistors, capacitors, tubes, transistors, or whatever active or passive component was as close as possible in both channels. I have found that the most useful piece of equipment to match components are my LCR meter and oscilloscope. They have allowed me to test the qualities of each component at more than one frequency to make sure they were matched as best as possible. Sure enough it provided me with the holographic and immersive soundstage that I think you are looking for.
Good luck on your project I hope that helps.
As a past purveyor of appliances that did the acoustic holography thing pretty well, here's my list of most-importants:
: Well-matched L & R channels.
: Linear-phase speakers, especially ~ 5kHz and below.
: Speaker dynamic linearity (freq response does not change with level)
: Linear Z and Zphase power suppplies in the electronics
: A good room and setup within it.
: Recordings that preserve and/or artfully recreate the space.
: Well-matched L & R channels.
: Linear-phase speakers, especially ~ 5kHz and below.
: Speaker dynamic linearity (freq response does not change with level)
: Linear Z and Zphase power suppplies in the electronics
: A good room and setup within it.
: Recordings that preserve and/or artfully recreate the space.