Stereo imaging is affected by channel imbalance, or different frequency response between channels.
Or variation in tracking of the volume control between channels, especially when adjusting the volume to compensate
for different amplifier gains. Some circuits may have reduced or asymmetrical channel separation, at higher frequencies.
Or variation in tracking of the volume control between channels, especially when adjusting the volume to compensate
for different amplifier gains. Some circuits may have reduced or asymmetrical channel separation, at higher frequencies.
Also phase characteristics of the amplifier’s distortion profile.
Tried that too, when comparing my Groundsound PA3CC with Hypex NC502MP. Can't figure out what is going on precisely. But maybe there's a little difference in distortion, making it a little clearer to hear certain aspects of the sound, making me experience a slight change in perception. But I'm guessing here. Because I can't put my finger on a specific thing, that made me think further about potentially upgrading to this newer type of amplifier, because I could not define whether it was actually better.
Hi, if both channels on both amplifiers have LR levels and frequency responses the same, and still difference, it likely has to do with harmonics and how auditory system works.
Find Nelson pass article about aplifier harmonics phase, and howbit can affect stereo image. It might have been on the firstwatt website if I remember. Then go and read / watch youtube David Griesinger talk about auditory Proximity and Limit of Localization Distance, which is priperty of howbauditory system works and very much has to do with hifi stereo imaging as well. It's how auditory system uses original sound harmonics to lock into them, if they pop sufficiently out from noise (all sounds around us) and are considered important to pay attention to, by auditory system.
I think these things could relate, amplifier could enhance or reduce this SNR auditory system cares about. It's a long story, so better study it yourself than me writing about it with poor language and incomplete knowledge for the thousandth time 🙂
Find Nelson pass article about aplifier harmonics phase, and howbit can affect stereo image. It might have been on the firstwatt website if I remember. Then go and read / watch youtube David Griesinger talk about auditory Proximity and Limit of Localization Distance, which is priperty of howbauditory system works and very much has to do with hifi stereo imaging as well. It's how auditory system uses original sound harmonics to lock into them, if they pop sufficiently out from noise (all sounds around us) and are considered important to pay attention to, by auditory system.
I think these things could relate, amplifier could enhance or reduce this SNR auditory system cares about. It's a long story, so better study it yourself than me writing about it with poor language and incomplete knowledge for the thousandth time 🙂
Ok, can't help it 😀
Here, extremely simple demonstration and fun tool to play with 😀 https://www.desmos.com/calculator/cwwofvwsqk
It's merely simple plot of some fundamental frequency and four harmonics whose amplitude / phase you can play with and try and mix/match them to get highest possible combined amplitude peak.
Cycle length is set to 1 and amplitude 1, the display plots two fundamental cycles. As you see, when the harmonics are added to the fundamental they together superimpose and every fundamental cycle make huge amplitude peak up to 4. If you start making amplitudes of the harmonics to zero all that is left is the fundamental with amplitude of 1.
Now, if you set the fundamental to 1 and all harmonics to zero, and start adding any of the harmonics the superposition amplitude goes up or down. Even order harmonics always boost the amplitude, while odd harmonics in positive phase compress!
What this means, that if auditory system considers this original sound harmonics lining up every fundamental cycle to determine whether to pay attention to it or not, iow determines how you perceive the sound, then this is critical stuff. Original harmonics means that every instrument, except sine wave generator, sound consists of fundamental tone and overtones, the harmonics, basic stuff right. What the test here kind of indicates, that if the fundamental cycle here is any of those original harmonics of original sound, certain added harmonics will boost them, while others would reduce them. This means to me, that it's possible that particular amplifier harmonics signature will help brain to detect sounds from each other on a recording, or just more broadly whether brain pays attention to sound of the recording or not. What the playback equipment can do is add noise to the room (unrelated to sound, like distortion, or excessive reflections), which reduces this SNR, could also delay some of them like group delay in loudspeakers reducing the superposition amplitude, while it also could help by amplifying harmonics. At least this is how I think it could work out.
Same thing happens just by changing listening distance, which changes ratio of direct and reflected sound, where the reflections reduce SNR like so and brain doesn't pay attention to your stereo sound unless you get close enough to speakers so that the sounds on the recording have sufficient SNR over the "noise", all sounds in the room bombarding the ear, so that your brain locks in to the direct sound. This means there is particular listening distance where your brain switches state: if you are closer to speakers than this distance, your brain pays attention to sound and you get very nice perception of what is on the recording. If you take step further from the transition your brain loses the attention and all you perceive is "noise" basically, the sound of recording fused as one with all noises in your room, clarity and localization reduces and so on. It's just very different perception of hifi stereo either side of the transition depending whether your brain pays involuntary attention to the music or not 😉 This is the Griesinger Limit of Localization Distance.
I speculate the transition is available with any speakers in any room, actually any sound source in any space, because this is property of our own auditory system, and not property of speakers or the room. Listening test to find it is relatively easy, I've found easiest just to play mono noise, like pink noise, for maximally strong dry phantom center image. Then move yourself back and forth on center line between speakers staying equidistant to both trying to listen where perception of the phantom center changes. Eyes closed listen "size" of the noise, when you think it gets focus, shrinks to very small you have it. Contrary, when you are further away, brain not paying attention, the phantom center is hazy spacious blob somewhere in the room where the speakers are. If you don't get these differences to appear, try toe-in the speakers and bring them closer and try again. You'd always have the hazy sound otherside of the room, and should always have the focus when you are very close to speaekers. Trick is try to notice where, at which listening distance, the perception between the two seems to change. I speculate that with typical speakers in reflective room this distance can be quite close, while on some high directivity speakers and managed acoustics it is further away. As a reference, I have relatively high directivity system with waveguide tweeter and cardioid mid in a typical family living room and the transition is about 2.2m from speaker to ear. Yours could be closer or farther. Now, if you want to listen your amplifier effect on music and to stereo image, find the transition and move yourself both sides of it to AB test stereo image, indirectly control your own auditory system to toggle stereo image by moving a little 😉
Drawing from the same phenomenon, you likely hear not much difference between two good amplifiers if you listen too far, because room early refections ruin the information anyway and makes your brain not listen. So, find the transition to make sure your brain pays attention, and then listen for the small differenses like swapping a good amp to another good amp.
Here, extremely simple demonstration and fun tool to play with 😀 https://www.desmos.com/calculator/cwwofvwsqk
It's merely simple plot of some fundamental frequency and four harmonics whose amplitude / phase you can play with and try and mix/match them to get highest possible combined amplitude peak.
Cycle length is set to 1 and amplitude 1, the display plots two fundamental cycles. As you see, when the harmonics are added to the fundamental they together superimpose and every fundamental cycle make huge amplitude peak up to 4. If you start making amplitudes of the harmonics to zero all that is left is the fundamental with amplitude of 1.
Now, if you set the fundamental to 1 and all harmonics to zero, and start adding any of the harmonics the superposition amplitude goes up or down. Even order harmonics always boost the amplitude, while odd harmonics in positive phase compress!
What this means, that if auditory system considers this original sound harmonics lining up every fundamental cycle to determine whether to pay attention to it or not, iow determines how you perceive the sound, then this is critical stuff. Original harmonics means that every instrument, except sine wave generator, sound consists of fundamental tone and overtones, the harmonics, basic stuff right. What the test here kind of indicates, that if the fundamental cycle here is any of those original harmonics of original sound, certain added harmonics will boost them, while others would reduce them. This means to me, that it's possible that particular amplifier harmonics signature will help brain to detect sounds from each other on a recording, or just more broadly whether brain pays attention to sound of the recording or not. What the playback equipment can do is add noise to the room (unrelated to sound, like distortion, or excessive reflections), which reduces this SNR, could also delay some of them like group delay in loudspeakers reducing the superposition amplitude, while it also could help by amplifying harmonics. At least this is how I think it could work out.
Same thing happens just by changing listening distance, which changes ratio of direct and reflected sound, where the reflections reduce SNR like so and brain doesn't pay attention to your stereo sound unless you get close enough to speakers so that the sounds on the recording have sufficient SNR over the "noise", all sounds in the room bombarding the ear, so that your brain locks in to the direct sound. This means there is particular listening distance where your brain switches state: if you are closer to speakers than this distance, your brain pays attention to sound and you get very nice perception of what is on the recording. If you take step further from the transition your brain loses the attention and all you perceive is "noise" basically, the sound of recording fused as one with all noises in your room, clarity and localization reduces and so on. It's just very different perception of hifi stereo either side of the transition depending whether your brain pays involuntary attention to the music or not 😉 This is the Griesinger Limit of Localization Distance.
I speculate the transition is available with any speakers in any room, actually any sound source in any space, because this is property of our own auditory system, and not property of speakers or the room. Listening test to find it is relatively easy, I've found easiest just to play mono noise, like pink noise, for maximally strong dry phantom center image. Then move yourself back and forth on center line between speakers staying equidistant to both trying to listen where perception of the phantom center changes. Eyes closed listen "size" of the noise, when you think it gets focus, shrinks to very small you have it. Contrary, when you are further away, brain not paying attention, the phantom center is hazy spacious blob somewhere in the room where the speakers are. If you don't get these differences to appear, try toe-in the speakers and bring them closer and try again. You'd always have the hazy sound otherside of the room, and should always have the focus when you are very close to speaekers. Trick is try to notice where, at which listening distance, the perception between the two seems to change. I speculate that with typical speakers in reflective room this distance can be quite close, while on some high directivity speakers and managed acoustics it is further away. As a reference, I have relatively high directivity system with waveguide tweeter and cardioid mid in a typical family living room and the transition is about 2.2m from speaker to ear. Yours could be closer or farther. Now, if you want to listen your amplifier effect on music and to stereo image, find the transition and move yourself both sides of it to AB test stereo image, indirectly control your own auditory system to toggle stereo image by moving a little 😉
Drawing from the same phenomenon, you likely hear not much difference between two good amplifiers if you listen too far, because room early refections ruin the information anyway and makes your brain not listen. So, find the transition to make sure your brain pays attention, and then listen for the small differenses like swapping a good amp to another good amp.
Last edited:
You've already had excellent answers, but I'd like to add the following.
I'm not an expert, but if I really understood your question, to date no one knows the exact and always applicable factor contributing to imaging in amplifiers.
Nor is there a measurement that evaluates imaging properties in amplifiers.
Please also note that it's possible that if you delve deeper many will come to tell you that imaging is an illusion of the hearing system...
Generally speaking, Audio here and elsewhere is treated almost exclusively in terms of frequency responses, distortion and noise, and other few measurements.
But imaging in amps is not measurable...
AFAIK
I did similar experiments, where i kept the same speakers and amplifier. Only swapping the preamplifiers.
(I did eq after every swap to maintain flat fr response in the listening position).
Swapped about a dozen preamps. Preamps with lowest distortion were not best imaging.
(I did eq after every swap to maintain flat fr response in the listening position).
Swapped about a dozen preamps. Preamps with lowest distortion were not best imaging.
Swapping vastly different amps (classAB, classA, classD) with big differences in damping will result in fr response differences. That contributes more to sound differences.
You must eq flat if you want to compare imaging.
Adason:
No surprise here, the "audio landscape" is littered with un-loved & under-appreciated products where the classic suite of measurements look perfect
Conversely amps like the old Dynaco "Stereo 70" still receive positive accolades and appreciation in spite of relatively poor measured metrics.
Just goes to show; ears are not microphones, brains are not computers and humans is way more complex that circuits.
We have much to learn
No surprise here, the "audio landscape" is littered with un-loved & under-appreciated products where the classic suite of measurements look perfect
Conversely amps like the old Dynaco "Stereo 70" still receive positive accolades and appreciation in spite of relatively poor measured metrics.
Just goes to show; ears are not microphones, brains are not computers and humans is way more complex that circuits.
We have much to learn
Maybe it's the tonality or 'colour' of the residual distortion?
In the early days of my DIY stuff, I wanted to create a "2.1" or subwoofer + satellite system, and I found out it was extremely hard to hide the sub, acoustically. Part of it seemed to do with stray high frequencies revealing the woofer location. But finite crossover slopes are not the only source of unwanted high frequencies.
Distortion could also pin down a sound source to its physical source and collapse the perception of a virtual image. But rather than the infinitesimal internal amplifier distortion that seems to dominate numerous discussions, it's the total distortion that matters, and it could vary a lot just by changing the amplifier. (Funny how that works, right?) Because each amplifier has a different 'scheme' for how the speaker is to be driven, so the speaker distortion also changes.
My thinking is that a good stereo image is likely a soft proof of a low distortion system overall. As for how, there are many ideas... One is to devise a system so the damping factor drops off at high frequencies. Some designs roll off the OLG early, so there is less feedback at the upper audio frequencies. For a bigger impact on damping factor, a transconductance topology could be used, and/or current feedback added. But the latter options are less likely to be found commercially, probably because of EQ requirements that have to be fine-tuned to each speaker.
In the early days of my DIY stuff, I wanted to create a "2.1" or subwoofer + satellite system, and I found out it was extremely hard to hide the sub, acoustically. Part of it seemed to do with stray high frequencies revealing the woofer location. But finite crossover slopes are not the only source of unwanted high frequencies.
Distortion could also pin down a sound source to its physical source and collapse the perception of a virtual image. But rather than the infinitesimal internal amplifier distortion that seems to dominate numerous discussions, it's the total distortion that matters, and it could vary a lot just by changing the amplifier. (Funny how that works, right?) Because each amplifier has a different 'scheme' for how the speaker is to be driven, so the speaker distortion also changes.
My thinking is that a good stereo image is likely a soft proof of a low distortion system overall. As for how, there are many ideas... One is to devise a system so the damping factor drops off at high frequencies. Some designs roll off the OLG early, so there is less feedback at the upper audio frequencies. For a bigger impact on damping factor, a transconductance topology could be used, and/or current feedback added. But the latter options are less likely to be found commercially, probably because of EQ requirements that have to be fine-tuned to each speaker.
I have owned amps with variable negative feedback.For those amps too much negative feedback clearly made the imaging less deep and less layered.Generally a hint of negative feedback sounded best.
A lot of people do not seem to be able to hear image depth.That includes experienced audiophiles.And I suspect most of the chief jackboot wearing headkickers at A.S.R.
A lot of people do not seem to be able to hear image depth.That includes experienced audiophiles.And I suspect most of the chief jackboot wearing headkickers at A.S.R.
Oh man ...
Imaging. I'm going to bet the brain has more to do with it than any electronics - unless the amp is broken.
You image (or locate an object) by processing phase differences between your ears. Okay, so what can affect that? Well, if your amp has a phase difference, it is broken. Your cables are more likely to cause a phase difference, but at audio frequencies? Frequency response should be pretty flat, if not your amp is broken. Phase and frequency response would be strongly related if you can hear it.
Level difference? Some amplifiers can have a dB difference, but those are really cheap. Being honest, your position in the room will affect perceived imaging differences way more than the electronics.
Feedback in a decent design makes performance more equal between channels, more alike to the design goal. Feedback and a bad design negatively affects the sound quality, but it is worse if you remove all feedback. A bad design is a bad design, feedback isn't at fault here. Given a good design, high feedback improves things subjectively as well as measured response.
Variable feedback is a real mess. You're varying gain as well along with all kinds of things. There is no control on that experiment. Even the amplifier performance may vary A LOT. So when you twist that control, you are not just changing the feedback level.
I worked in recording studios. Image depth starts there with the recorded material. Your room and speakers have the final say. If your electronics affects imaging, it is broken.
Imaging. I'm going to bet the brain has more to do with it than any electronics - unless the amp is broken.
You image (or locate an object) by processing phase differences between your ears. Okay, so what can affect that? Well, if your amp has a phase difference, it is broken. Your cables are more likely to cause a phase difference, but at audio frequencies? Frequency response should be pretty flat, if not your amp is broken. Phase and frequency response would be strongly related if you can hear it.
Level difference? Some amplifiers can have a dB difference, but those are really cheap. Being honest, your position in the room will affect perceived imaging differences way more than the electronics.
Feedback in a decent design makes performance more equal between channels, more alike to the design goal. Feedback and a bad design negatively affects the sound quality, but it is worse if you remove all feedback. A bad design is a bad design, feedback isn't at fault here. Given a good design, high feedback improves things subjectively as well as measured response.
Variable feedback is a real mess. You're varying gain as well along with all kinds of things. There is no control on that experiment. Even the amplifier performance may vary A LOT. So when you twist that control, you are not just changing the feedback level.
A.S.R. isn't what I would call accurate, I'm not a member, nor will I be. What I say comes from a lifetime in audio, professionally. I have the test gear and several systems. Listened to just about everything out there.
I worked in recording studios. Image depth starts there with the recorded material. Your room and speakers have the final say. If your electronics affects imaging, it is broken.
There is a wealth of information from group blind reviews of all types of audio components carried out by Hi Fi Choice Magazine over many years.In most of these differences in imaging are noted and described and those components assigned a relative rating.The listeners tend to be a mix of reviewers and industry engineers/designers.Those differences occurred in preamplifiers amplifiers and sources.I bought a CD player based on those findings back in 1996 and it has very good image depth compared to most sources.Still sounds better than most modern digital.
I agree that recording ,speakers and room are the major factors and if they are not right then you will never get decent image depth.However other components can and do contribute to optimising that capability.And that ability is backed up by blind listening tests.
I agree that recording ,speakers and room are the major factors and if they are not right then you will never get decent image depth.However other components can and do contribute to optimising that capability.And that ability is backed up by blind listening tests.
In my decades-long amateur experience I easily "deduced" that no device shows the same imaging, almost as if each of them had a sort of imaging "signature".
I've never noticed that two different devices produced the same imaging.
Indeed, sometimes I changed the device just for that, to evaluate the search (and the finding) for a "better" imaging.
However, I've always noticed one thing, without ever knowing the real reason.
When I swapped a component of the system of any type, from transducers (not the speakers though) to the electronics and even cables (I say this quietly so as not to create a fuss), then the imaging invariably changed, while obviously keeping the speakers in the exact same position.
And so far the above is something expected.
But the real thing I noticed in particular is the following and that's the need to change "moderately and as much as enough" also the orientation of the speakers.
Almost as if it were necessary to "adapt" the two different way of recreating the imaging together of the device I was swapping.
It seems to me that replacing one piece of equipment with another one also requires an adjustment of the speakers position and I would like to know if anyone has had a similar experience.
Yes, we really have a lot to discover and learn still.
I've never noticed that two different devices produced the same imaging.
Indeed, sometimes I changed the device just for that, to evaluate the search (and the finding) for a "better" imaging.
However, I've always noticed one thing, without ever knowing the real reason.
When I swapped a component of the system of any type, from transducers (not the speakers though) to the electronics and even cables (I say this quietly so as not to create a fuss), then the imaging invariably changed, while obviously keeping the speakers in the exact same position.
And so far the above is something expected.
But the real thing I noticed in particular is the following and that's the need to change "moderately and as much as enough" also the orientation of the speakers.
Almost as if it were necessary to "adapt" the two different way of recreating the imaging together of the device I was swapping.
It seems to me that replacing one piece of equipment with another one also requires an adjustment of the speakers position and I would like to know if anyone has had a similar experience.
Yes, we really have a lot to discover and learn still.
Sigh ...
Poorly controlled testing. Expectation bias is extremely powerful and clouds most tests. This has been repeated over the decades by multiple groups.
So let's start again shall we? Imaging (depth has no meaning as it is the same thing) came about how? It is the sound location process built into a human (most animals) as a survival mechanism. Period. Nature developed this ability to keep us alive. So then, how exactly do we locate the direction and therefore location of a sound? Time difference between our ears. Frequency aberrations between the sound arriving to our ears can affect this, not surprisingly.
All right then. So what affects phase? Well, your source material, the transducers you are listening to are far more likely to differ in phase. Certainly room acoustics. For electronics to create that much phase shift, your electronics are broken, period.
People have to stop imagining stuff. Look at the time differences needed to shift an image, then look at the delay in electronic terms. Reviewers and "golden ear" folks always seem to bring up nonsense, stuff that isn't possible. And yes, we can very easily measure these things.
I'm just asking you to actually think instead of listening to what others say. Phase difference between channels in an electronic system with the same electronics in both channels will not cause imaging differences - unless something isn't working properly. Then you will also have other issues. Phase shift always comes along hand in hand with a filter response.
Strange that when these items are reviewed blind and the listeners do not know what product they are listening to ,and the test is repeated multiple times, those products found to have the best imaging in the first listening also are generally found to have the same quality in subsequent ones. And then ranked and described accordingly.
Expectation bias cuts both ways and is probably most potent in people who lack the ability to hear image depth and them believing that other people are imagining it makes perfect sense in that context.
Expectation bias cuts both ways and is probably most potent in people who lack the ability to hear image depth and them believing that other people are imagining it makes perfect sense in that context.
Hi jtgofish,
Think man!
Re-read what I posted above. Then actually consider those points. The only people who might believe these tests you are referring to are folks who have no idea how amplifiers work, no idea on how electronic circuits work.
Convert phase into time, then look at the frequency response (1/f) to see what is required. Phase shift normally indicates a filter, or a time delay circuit. This means a rising or falling frequency response which you would notice over "imaging".
People who review equipment normally don't understand electronics well (or at all). They want to be part of "the club" so they try to describe things they imagine. Some of the prose they post should be a warning sign, then often they describe the music. Very confused individuals.
So, look at the physics of the situation. The laws of physics don't take into account what goes on in the goo between people's ears, but all nature follows the laws of physics.
Think man!
Re-read what I posted above. Then actually consider those points. The only people who might believe these tests you are referring to are folks who have no idea how amplifiers work, no idea on how electronic circuits work.
Convert phase into time, then look at the frequency response (1/f) to see what is required. Phase shift normally indicates a filter, or a time delay circuit. This means a rising or falling frequency response which you would notice over "imaging".
People who review equipment normally don't understand electronics well (or at all). They want to be part of "the club" so they try to describe things they imagine. Some of the prose they post should be a warning sign, then often they describe the music. Very confused individuals.
So, look at the physics of the situation. The laws of physics don't take into account what goes on in the goo between people's ears, but all nature follows the laws of physics.