|
Post by bencat on Apr 4, 2024 9:05:49 GMT
Macca the fact not the suggestion we all hear differently is something Audiologists are all very well aware of . As each and every one of us hears via a different route we all have different responses to sounds . I wish I could find the BBC programme that was shown about this so i could provide a link I was not aware of much that was shown in it . Not long after this I started a set of appointments to try and deal with my tinnitus and I had lots of conversations with my Audiologist who confirmed much of what the programme said and that we are quite a way off from understanding how and why we hear as we do . I also asked if things are so different why in some cases there is often a consensus of people liking a certain set of sounds . Her explanation was that there is a generality of what we hear so often many will agree but also there are often individuals who have a very different view against the majority . In tests these individuals have often been found to have very extreme hearing paths .
|
|
|
Post by macca on Apr 4, 2024 19:17:52 GMT
Macca the fact not the suggestion we all hear differently is something Audiologists are all very well aware of . As each and every one of us hears via a different route we all have different responses to sounds . I wish I could find the BBC programme that was shown about this so i could provide a link I was not aware of much that was shown in it . Not long after this I started a set of appointments to try and deal with my tinnitus and I had lots of conversations with my Audiologist who confirmed much of what the programme said and that we are quite a way off from understanding how and why we hear as we do . I also asked if things are so different why in some cases there is often a consensus of people liking a certain set of sounds . Her explanation was that there is a generality of what we hear so often many will agree but also there are often individuals who have a very different view against the majority . In tests these individuals have often been found to have very extreme hearing paths . yes no question that there are plenty of people with variations in their hearing which are large enough to matter. That's not the same as 'we all hear differently. ' We're not as unique as we like to think When they blind tested loudspeaker preference 80 percent preferred the same type of speaker. But 20 percent didn't, and that's a lot of people across a population. Not sure that is because they hear differently but becuase they are looking for different things in the sound. For example I was in a room with about 10 people listening to some huge Avantgarde multi-way horns. Me and one other thought they were dreadful, the other 8 said they would buy them if they had the money/space. Personal preference can differ massively and it can't be argued about.
|
|
|
Post by stevew on Apr 5, 2024 7:15:22 GMT
Macca the fact not the suggestion we all hear differently is something Audiologists are all very well aware of . As each and every one of us hears via a different route we all have different responses to sounds . I wish I could find the BBC programme that was shown about this so i could provide a link I was not aware of much that was shown in it . Not long after this I started a set of appointments to try and deal with my tinnitus and I had lots of conversations with my Audiologist who confirmed much of what the programme said and that we are quite a way off from understanding how and why we hear as we do . I also asked if things are so different why in some cases there is often a consensus of people liking a certain set of sounds . Her explanation was that there is a generality of what we hear so often many will agree but also there are often individuals who have a very different view against the majority . In tests these individuals have often been found to have very extreme hearing paths . yes no question that there are plenty of people with variations in their hearing which are large enough to matter. That's not the same as 'we all hear differently. ' We're not as unique as we like to think When they blind tested loudspeaker preference 80 percent preferred the same type of speaker. But 20 percent didn't, and that's a lot of people across a population. Not sure that is because they hear differently but becuase they are looking for different things in the sound. For example I was in a room with about 10 people listening to some huge Avantgarde multi-way horns. Me and one other thought they were dreadful, the other 8 said they would buy them if they had the money/space. Personal preference can differ massively and it can't be argued about. If anyone is going to argue about it, seems that you will.
|
|
|
Post by stryder5 on Apr 5, 2024 10:07:51 GMT
2 hours ago via mobile Arke likes thisQuotelikePost Options Post by stevew on 2 hours ago macca Avatar 14 hours ago macca said: bencat Avatar yesterday at 10:05am bencat said: Macca the fact not the suggestion we all hear differently is something Audiologists are all very well aware of . As each and every one of us hears via a different route we all have different responses to sounds . I wish I could find the BBC programme that was shown about this so i could provide a link I was not aware of much that was shown in it . Not long after this I started a set of appointments to try and deal with my tinnitus and I had lots of conversations with my Audiologist who confirmed much of what the programme said and that we are quite a way off from understanding how and why we hear as we do . I also asked if things are so different why in some cases there is often a consensus of people liking a certain set of sounds . Her explanation was that there is a generality of what we hear so often many will agree but also there are often individuals who have a very different view against the majority . In tests these individuals have often been found to have very extreme hearing paths . yes no question that there are plenty of people with variations in their hearing which are large enough to matter. That's not the same as 'we all hear differently. ' We're not as unique as we like to think When they blind tested loudspeaker preference 80 percent preferred the same type of speaker. But 20 percent didn't, and that's a lot of people across a population. Not sure that is because they hear differently but becuase they are looking for different things in the sound. For example I was in a room with about 10 people listening to some huge Avantgarde multi-way horns. Me and one other thought they were dreadful, the other 8 said they would buy them if they had the money/space. Personal preference can differ massively and it can't be argued about. If anyone is going to argue about it, seems that you will MACCA’S LAW......LOL
|
|
|
Post by macca on Apr 6, 2024 7:09:14 GMT
yes no question that there are plenty of people with variations in their hearing which are large enough to matter. That's not the same as 'we all hear differently. ' We're not as unique as we like to think When they blind tested loudspeaker preference 80 percent preferred the same type of speaker. But 20 percent didn't, and that's a lot of people across a population. Not sure that is because they hear differently but becuase they are looking for different things in the sound. For example I was in a room with about 10 people listening to some huge Avantgarde multi-way horns. Me and one other thought they were dreadful, the other 8 said they would buy them if they had the money/space. Personal preference can differ massively and it can't be argued about. If anyone is going to argue about it, seems that you will. And yet you're clearly an avid reader
|
|
|
Post by macca on Apr 6, 2024 7:30:37 GMT
Okay back to in room measurements and their utility above the transition frequency. As promised to Jason I've found the relevant research:
Pg 371 section 13.2.3
'The claim is that measurements of the steady-state sound field using an omnidirectional microphone, and signal processing by an algorithm, can repair imperfections in unknown loudspeakers in unknown rooms.
There is no doubt that such a process can yield improvements at low frequencies for a single listener, but above the transition frequency to claim that a smooth, steady-state room curve derived from an omnidirectional microphone is an adequate substitute for the timbral and spatial perceptions by two ears and a brain is absurd. However it clearly can be good business.'
Pg348 Section 12.2.3
if the measured curve deviates from the target, does that mean that applying equalization to make it match the target will ensure satisfaction? Unfortunately not...because in-room measurements include acoustical interference artefacts that may be automatically and inappropriately 'corrected' there is a very real possibility that a good loudspeaker may be degraded.
As shown on Figure 4.12, above the transition frequency the most reliable basis for equalization is anechoic data on the loudspeaker.'' (My bold).
|
|
Arke
Moderator
Posts: 1,004
|
Post by Arke on Apr 6, 2024 7:46:14 GMT
Okay back to in room measurements and their utility above the transition frequency. As promised to Jason I've found the relevant research: Pg 371 section 13.2.3 'The claim is that measurements of the steady-state sound field using an omnidirectional microphone, and signal processing by an algorithm, can repair imperfections in unknown loudspeakers in unknown rooms. There is no doubt that such a process can yield improvements at low frequencies for a single listener, but above the transition frequency to claim that a smooth, steady-state room curve derived from an omnidirectional microphone is an adequate substitute for the timbral and spatial perceptions by two ears and a brain is absurd. However it clearly can be good business.' Pg348 Section 12.2.3 if the measured curve deviates from the target, does that mean that applying equalization to make it match the target will ensure satisfaction? Unfortunately not...because in-room measurements include acoustical interference artefacts that may be automatically and inappropriately 'corrected' there is a very real possibility that a good loudspeaker may be degraded. As shown on Figure 4.12, above the transition frequency the most reliable basis for equalization is anechoic data on the loudspeaker.'' (My bold). Brilliant, thanks Martin. Yes, that all makes sense. The limitation I mainly perceive is that mic is a point source and our ears are not. Even below the Schroeder frequency the low frequencies can vary by 10-15db with a 20-30cm fore/aft shift of the mic. If anything (IME) they are even more affected by mic position than mid and high frequencies. Easy to demonstrate too. I have found in-room listening position measurements are most accurately achieved by taking about 10-20 measurements in a grid pattern around the listening position. These are then spatially averaged.
|
|
|
Post by bencat on Apr 6, 2024 9:18:28 GMT
It is noticeable Jason that the grid measurements you have settled on is very similar to how DIRAC Live works . As I use DIRAC in my active systems it is nice to know that the method is one that works in practice .
|
|
|
Post by macca on Apr 6, 2024 10:01:48 GMT
Okay back to in room measurements and their utility above the transition frequency. As promised to Jason I've found the relevant research: Pg 371 section 13.2.3 'The claim is that measurements of the steady-state sound field using an omnidirectional microphone, and signal processing by an algorithm, can repair imperfections in unknown loudspeakers in unknown rooms. There is no doubt that such a process can yield improvements at low frequencies for a single listener, but above the transition frequency to claim that a smooth, steady-state room curve derived from an omnidirectional microphone is an adequate substitute for the timbral and spatial perceptions by two ears and a brain is absurd. However it clearly can be good business.' Pg348 Section 12.2.3 if the measured curve deviates from the target, does that mean that applying equalization to make it match the target will ensure satisfaction? Unfortunately not...because in-room measurements include acoustical interference artefacts that may be automatically and inappropriately 'corrected' there is a very real possibility that a good loudspeaker may be degraded. As shown on Figure 4.12, above the transition frequency the most reliable basis for equalization is anechoic data on the loudspeaker.'' (My bold). Brilliant, thanks Martin. Yes, that all makes sense. The limitation I mainly perceive is that mic is a point source and our ears are not. Even below the Schroeder frequency the low frequencies can vary by 10-15db with a 20-30cm fore/aft shift of the mic. If anything (IME) they are even more affected by mic position than mid and high frequencies. Easy to demonstrate too. I have found in-room listening position measurements are most accurately achieved by taking about 10-20 measurements in a grid pattern around the listening position. These are then spatially averaged. yes - this is why he says LF correction will work for one position, traditionally we measure from the listening position because what matters is what it sounds like from where we listen, and not what it sounds like standing over in the corner. But we should be aware of the pitfalls of using those measurements to correct the FR above transition. Of course if we are happy with the sonic results, then we are happy and nothing further needs to be said or done. This whole discussion started because you questioned my methodology in correcting above transition without in-room measurements, hopefully you now appreciate that I do have some justification for doing it that way?
|
|
Arke
Moderator
Posts: 1,004
|
Post by Arke on Apr 6, 2024 19:37:00 GMT
Brilliant, thanks Martin. Yes, that all makes sense. The limitation I mainly perceive is that mic is a point source and our ears are not. Even below the Schroeder frequency the low frequencies can vary by 10-15db with a 20-30cm fore/aft shift of the mic. If anything (IME) they are even more affected by mic position than mid and high frequencies. Easy to demonstrate too. I have found in-room listening position measurements are most accurately achieved by taking about 10-20 measurements in a grid pattern around the listening position. These are then spatially averaged. yes - this is why he says LF correction will work for one position, traditionally we measure from the listening position because what matters is what it sounds like from where we listen, and not what it sounds like standing over in the corner. But we should be aware of the pitfalls of using those measurements to correct the FR above transition. Of course if we are happy with the sonic results, then we are happy and nothing further needs to be said or done. This whole discussion started because you questioned my methodology in correcting above transition without in-room measurements, hopefully you now appreciate that I do have some justification for doing it that way? Based on those excerpts, yes, you shouldn't bother equalising based on listening position measurements. Which does beg the question: `what is the point of fully active speakers with tuning through all frequencies?` I will continue doing what I am doing in my room (and potentially helping clients to do so), which is NOT using equalisation above the Schroeder frequency. I shall be using the active equalisation for bass. Midrange and higher frequencies have been tweaked and optimised in my room using an iterative process of measuring AND listening. Any changes to HF roll off, and mid range peaks/troughs, presence region etc. are achieved through tweeter attenuation, speaker positioning (and toe in, tilt), listening position, diffraction, absorption, pictures, curtains, chairs, cushions etc.. All changes and evaluation are based on weeks of listening and measuring distortion, RT60 decay and FR. In time I shall add more to my blog (sorry not had time lately). I have documented some of the room listening room development, but shall add lots more on the measurements and room acoustics tweaks in due course.
|
|
Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,071
|
Post by Bigman80 on Apr 6, 2024 21:17:07 GMT
yes - this is why he says LF correction will work for one position, traditionally we measure from the listening position because what matters is what it sounds like from where we listen, and not what it sounds like standing over in the corner. But we should be aware of the pitfalls of using those measurements to correct the FR above transition. Of course if we are happy with the sonic results, then we are happy and nothing further needs to be said or done. This whole discussion started because you questioned my methodology in correcting above transition without in-room measurements, hopefully you now appreciate that I do have some justification for doing it that way? Based on those excerpts, yes, you shouldn't bother equalising based on listening position measurements. Which does beg the question: `what is the point of fully active speakers with tuning through all frequencies?` I will continue doing what I am doing in my room (and potentially helping clients to do so), which is NOT using equalisation above the Schroeder frequency. I shall be using the active equalisation for bass. Midrange and higher frequencies have been tweaked and optimised in my room using an iterative process of measuring AND listening. Any changes to HF roll off, and mid range peaks/troughs, presence region etc. are achieved through tweeter attenuation, speaker positioning (and toe in, tilt), listening position, diffraction, absorption, pictures, curtains, chairs, cushions etc.. All changes and evaluation are based on weeks of listening and measuring distortion, RT60 decay and FR. In time I shall add more to my blog (sorry not had time lately). I have documented some of the room listening room development, but shall add lots more on the measurements and room acoustics tweaks in due course. You *should* continue doing what you're doing...the results are superb.
|
|