Camplo this an image of a single source impulse response correct?
so if there where two sources then you should be able point to where amongst the bumps and wiggles and pretty colors where the image of oh let say the snare drum lives? and better yet the blurring that reverb supplies,no?
so if there where two sources then you should be able point to where amongst the bumps and wiggles and pretty colors where the image of oh let say the snare drum lives? and better yet the blurring that reverb supplies,no?
Attachments
No, that isn't an Impulse, to the best of my knowledge. It is simply an image used to convey a concept that Sound has a literal image, that can be viewed using various tools. The more advanced we get, the better pictures we generate but essentially monitoring the same thing, pressure waves which create the sound field. Just because we as humans cannot literally see the sound field, don't put that limitation on technology, it looks like in the future the equipment will, as well, the software. Pretty sure software is already doing this to some degree. We can't see sound unfortunately, we are just staring at it, using our ears to perceive what's there. It has a very distinct "image", it being every unique situation you can think of. We cannot see thermal energy per say but with technology we can...Camplo this an image of a single source impulse response correct?
And to no lesser extent, acoustical cameras are being developed currently. Why are we saying we can't measure the image of an acoustical event? Its not about the ability of the current technology its about the concept of what an image is.
So while you guys are wondering about what image is and where it might be and how to measure, I believe that it is measured using simple tools like microphones lol... and technology that generates images like above. How I perceive the image is another conversation, another part of science even, called psychoacoustics. The very basics, of pressure and vector seem to be at play? Harmonics? I'm sure it gets pretty deep and fast... but it has a physical manifestation and it is measurable or should we say its character and aspects are?
Last edited:
Not a measure.... of if... it can be, measured, but one of, where to measure and which measurement, and how to interpret them.
What is it actually an image of?Camplo this an image of a single source impulse response correct?
sorry to still take you to task but you earlier emphatically stated that IR can be used to measure "image"...if you intended to say "might be used to measure" i'll back down. you where inferring that it could be done
while we both wait for technology to advance to the point where image can shown/ measured can you agree that currently we're not there?
while we both wait for technology to advance to the point where image can shown/ measured can you agree that currently we're not there?
Last edited:
Floyd Toole started the "circle of confusion" as a counter to the never ending barrage of complaints from subjectivist audiophiles who had beefs about the use of test signals vs actual music as a means of evaluating loudspeaker performance.
now it's taken on a life of it's own...
now it's taken on a life of it's own...
while we both wait for technology to advance to the point where image can shown/ measured can you agree that currently we're not there?
What is it actually an image of?
A simple microphone can be used to investigate the 3d sound field its just no fun. In actuality, it takes multi microphones to really get into it, does it not? I am not the expert on the topic.
I'd have a better chance simulating this stuff than taking the time to learn and physically execute measurements of this nature.... but simple measurement in increments of the XZY could be done with a measurement microphone... Then you'd have to talk about how to interpret the measurements but a person in DIY can get that far.
At the end of the day you have a multitude of IR's.....
Last edited:
lol ya, sound can travel through other mediums (other than air), and it seems they can measure the sound there, too.... Just another representation of technology allowing us the visual of what we cannot see, otherwise.
Last edited:
Thats cause I am not an expert on this stuff, that doesn't equate to wrong. I have some holes in my game, but I am not completely lost =)The more you post, the more you mess up camplo.
If you even think thats the message I am trying to convey, I have failed to make my point but I am not able to explain myself properly anyway so its my fault.Total sound pressure does not equal stereo image. Please understand that.
Imagine you are sitting in the concert hall infront of an orchestra. You have best seat, in the middle, in the sweet spot.
When soloist on the podium sings, standing in the centre, both of your ears receive the same audio information, first direct sound, than reflected sound, brain clearly process central stage signal source.
When singer on the left starts singing, your ears receive different audio information. Left ear sound level will be higher, there will be slight time delay to the right ear as sound travers around your head, and there will be some high frequency attenuation to right ear. Brain has no problem, processes all into nice left stage localization.
If more singers join, situation is the same, brain creates easily image of what is going on the stage. Because each signal source comes to each ear with different level, phase and frequency response.
Not all recordings succeed in capturing perfectly stereo image, which has been discussed, but many do. We still do not have way to measure stereo image.
IR my ***.
When soloist on the podium sings, standing in the centre, both of your ears receive the same audio information, first direct sound, than reflected sound, brain clearly process central stage signal source.
When singer on the left starts singing, your ears receive different audio information. Left ear sound level will be higher, there will be slight time delay to the right ear as sound travers around your head, and there will be some high frequency attenuation to right ear. Brain has no problem, processes all into nice left stage localization.
If more singers join, situation is the same, brain creates easily image of what is going on the stage. Because each signal source comes to each ear with different level, phase and frequency response.
Not all recordings succeed in capturing perfectly stereo image, which has been discussed, but many do. We still do not have way to measure stereo image.
IR my ***.
You are talking about the quality or approach to a recording. I am talking about how the loudspeaker could or should treat the recording.. we are not talking about the same image.
To capture is to measureNot all recordings succeed in capturing perfectly stereo image, which has been discussed, but many do. We still do not have way to measure stereo image.
Camplo, when turk and adason talk about measuring image I'm a trifle confused. Do you believe they talking about quantifying the imaging ability of a speaker? ..or perhaps are they trying to map the soundstage (which if so, what does that have to do with speakers..)?To capture is to measure
I thought thats what we were debating, can we measure the quality known as imaging. I suggested an impulse response was a measurement of such a thing, then we went on from there. I bring up the sound field and the 3d representations of the direct and indirect sound, because it is the manifestation of what it is we are perceiving and it is measurable. If we can measure all the things that create the image,,, why not just say we are measuring the image, the acoustical image that is, and the mental imagery is a separate discussion as well. When I see people talk about image and refer to localization, it is all created by direct and indirect sound manifested in the physical world, representable in 3d using measurements and simulation. What the image Is, is encoded in the signal. How well the loudspeaker+room recreates it is a matter of imaging resolution (accuracy) or preference . The original image is either sharp or dull and the playback system is to follow suit. If you say a loudspeaker systems imaging is amazing, are you saying the recording is amazing or the loudspeaker is amazing for not getting in the way of the recording. The loudspeaker only does as its told and from there is at the mercy of the room. Should we thank the loudspeaker designer or the recording engineer and mix engineer or the room and system designer, etc Obviously there is flow here that at every stop can change the image. Its the dang electric signal and all the transfer functions it will pass through before hitting your ear. The most perfect representation, I thought, was the theoretically perfect IR which represent 0 transfer function. This being a performance metric that can be applied to the pressure character coming from a source in any direction and indirect sound as well.
I agree that the stereo information is what it is. What we're really talking about is the speaker's ability to give that image. In other words, how do we use measurements to ensure the speaker system itself can 'image'.
Not that it would matter, I think there's evidence to suggest impulse response could be useful in either endeavour.. but like you, I'm inclined to leave the layout of the soundstage itself for another thread.
Not that it would matter, I think there's evidence to suggest impulse response could be useful in either endeavour.. but like you, I'm inclined to leave the layout of the soundstage itself for another thread.
(Now 137 messages in thread.)brain creates easily image of what is going on the stage. Because each signal source comes to each ear with different level, phase and frequency response.
Not all recordings succeed in capturing perfectly stereo image, which has been discussed, but many do. We still do not have way to measure stereo image.
Let's define what is to be measured then, by the words imaging and soundstage. Hopefully, anyone with proper equipment and test environment ought to be able to conduct the necessary experiments. Nothing new really -- from Fletcher et al. (1934) to modern VR/AR/AI (re-) creations.
Given a recording data-stream and/or audio playback of concert "standard voices" both human and instrumental, the task would be to measure, compute, and summarize a set of source events each with timing and localization estimates (distance relative or absolute, direction including height, size and shape contour, movement vectors if any). Fletcher had in fact done this using behind-veil speech and human listeners. We do it by-ear listening to reproduced music/vids, as well as in our daily lives. Many distance cues are well understood: HF decay from travelling through air, off-axis attenuation/power, reflections etc. Directional cues from ear-head-ear transfer function merited a paper in Fletcher (1934), including discussion of combined distance with/contra direction perception. So one can measure by listening test per Fletcher, or using mic(s) FR/IR/FFT/RTA/etc. possibly like this:
Set up both speakers; test first 1ch monophonic depth perception, drawing for each sound event putative locale-paths curving away from one ear, then the other ear; separately test 2ch stereophonic directional cues phase/loudness/FR (of standard voices); finally combine to best-localize. (Simplified, LOL.)
Most people can't be bothered with this stuff. Convenience vs Hi-Fi bandwidth? BT-headphones/earbuds have enough of each to essentially replace/retire home loudspeakers for music. As for us mostly seniors on diyaudio -- anyone who actually hears 15khz bandwidth music at-the-ears at normal level and everyday casual listening pose, please say so. (I think I do.)
Many smartphones have built-in hearing test for automatic headphones PEQ-compensation; just output through speakers and take the short test.
If measuring by omni-directional mic(s), I'd fashion a simple dummy head-and-shoulders with nose and outer ears. And I cannot stress enough, the speakers are well-defined only if their 6DoF physical position/angle/distance relative to the mics'/ears' 6DoF are precisely and repeatably set up. Even after minimizing or leveraging room effect.Set up both speakers; test first 1ch monophonic depth perception, drawing for each sound event putative locale-paths curving away from one ear, then the other ear; separately test 2ch stereophonic directional cues phase/loudness/FR (of standard voices); finally combine to best-localize. (Simplified, LOL.)
And it should be quite easy to (DSP) muck with a particular audio characteristic, such as time or phase alignment, or limit/tilt bandwidth, to try to break a best-tuned stereo image. Even just turning off one channel while playing (depth should persist; sources from opposite side more fuzzy -- why?). This really isn't rocket science or voodoo.
But to tell the truth, I generally found such a strong correlation between perceived monophonic soundstage depth and eventual stereo imaging (and both correlate with transcient fidelity), that for practical purposes just mono step one is sufficient to compare, measure, judge, and tweak loudspeaker imaging.
Last edited:
- Home
- Loudspeakers
- Multi-Way
- Bigger midranges/speakers have better imaging?