Perception/reality
Agreed! In my experience amplitude and frequency domain errors are most easily perceived and are the anomalies most reliably detected in an ABX.
Hey, I'm for just about anything to heighten the illusion. We must remember that what we are trying to do in high-quality reproduction is create a convincing illusion. We call it reproduction, but it is not: we make no attempt to build a physical space like the one the original performance was made in, nor hire the same musicians to play. By it's very nature it is not real: there are no musicians there, not the acoustic space in which the recording may have been made(or simulated), but just an illusion. It is the most precious of illusions to me: I've dedicated my life to music and audio, but as an illusion it is subject to many of the sleight-of-hand phenomena known to illusionists, some of whom seem to be making a good living in the audio business.
I think John was right on when he said:
Nor do I recommend messing with brain chemistry, but the fact is that the brain is the audio processor we all listen to/through that makes the illusion. Any changes to this processor affect our perception, making 100% confident claims of audibility difficult. Therein lies the argument for testing of some sort.
Howie
Howard Hoyt
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org
Agreed! In my experience amplitude and frequency domain errors are most easily perceived and are the anomalies most reliably detected in an ABX.
Hey, I'm for just about anything to heighten the illusion. We must remember that what we are trying to do in high-quality reproduction is create a convincing illusion. We call it reproduction, but it is not: we make no attempt to build a physical space like the one the original performance was made in, nor hire the same musicians to play. By it's very nature it is not real: there are no musicians there, not the acoustic space in which the recording may have been made(or simulated), but just an illusion. It is the most precious of illusions to me: I've dedicated my life to music and audio, but as an illusion it is subject to many of the sleight-of-hand phenomena known to illusionists, some of whom seem to be making a good living in the audio business.
I think John was right on when he said:
Nor do I recommend messing with brain chemistry, but the fact is that the brain is the audio processor we all listen to/through that makes the illusion. Any changes to this processor affect our perception, making 100% confident claims of audibility difficult. Therein lies the argument for testing of some sort.
Howie
Howard Hoyt
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org
The joke, on me, is that I'm almost completely oblivious of frequency response issues - someone could massively reshape the curve, and I would barely notice anything - it's not "my thing". OTOH, low level distortion issues really, really bug me - they scream at me, drive me batty until I've worked out how to sort them out - I've had terrible times going to stage shows, and having to endure the irksome grubbiness of the sound, having paid good money for the "pleasure" ...
Yes. The interesting thing is that I stumbled upon, that worked for me and others, a far more intense version of that - in every sense it's a large jump up from the normal standard, but the presentation of that level is very fragile, because not enough is known about all the factors that relate to making it happen. The fundamental is that certain types of distortion have to be at a very low level, and with the current state of audio engineering this is not trivial - usually, only fastidious tweaking can push the overall quality to a sufficiently high standard.
Scott, thank you for going to the trouble of finding your article.
This looks exactly like what I need.
Dan.
Perception vs measurement
My experience has been the same with virtually all of the differences I have been involved in investigating. Challenging systems with pulses and analyzing the output as is done with the TEF system has been most enlightening. The extension of this idea which has gained some popularity in the DIY community is that there are audible differences which we have not yet identified as discrete phenomena, and do not yet have equipment to measure. I don't mean to drag you into a flame war, but what is your opinion on this issue? Feel free to not reply!
Howie
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org
My experience has been the same with virtually all of the differences I have been involved in investigating. Challenging systems with pulses and analyzing the output as is done with the TEF system has been most enlightening. The extension of this idea which has gained some popularity in the DIY community is that there are audible differences which we have not yet identified as discrete phenomena, and do not yet have equipment to measure. I don't mean to drag you into a flame war, but what is your opinion on this issue? Feel free to not reply!
Howie
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org
Digital sound has had a pretty difficult birth, because the "experts" kept insisting that there nothing wrong, yet people who actually listened to what their purchased equipment produced knew there was something not quite right, no matter what the numbers said. It's been a long, drawn out "battle" to get the "experts" to acknowledge that their measuring regime was not good enough to tell the full story, bits of information about how the system was misbehaving were being missed ... and that process is still taking place, to some degree ...
I'm wondering here - why the focus on differences, rather than improvements?
Believe it or not the corner was peaking out of a pile on my desk, the proverbial engineer's desk. We once had an engineer's office declared a fire hazard due to some manager's OCD tendencies.
I think that the reason for investigating differences is to lead to improvements. There is no other way after all. Before one can focus in improvements one must focus on differences. If you do it in the wrong order, you get a bad result. Furthermore, having a natural curiosity to catalogue and investigate differences is not in itself a bad thing at all, and will likely lead to more effective improvements.
I don't follow. Why? After all an improvement is a positive difference. Why would one have to include negative differences in one's focus?
Personally, I don't concern myself with "positive" differences - these are irrelevant to me. The "nuggets of gold" are the negative differences, because these pinpoint, give vital clues as to where the improvements can be made, where there is some abnormal, audible artifact in the belly of the beast ...
The more you know about a problem, the better the solution you can make. I often find myself wishing I had written down results of experiments, good or bad beforehand so I could study them later. But of course they were bad results so I wanted to forget about them. It's a lack of foresight.
A "disease" I think we all have, 
- over the years I have got so many interesting insights from simulations, things behaving in a non-intuitive way - and later on I recall in another situation smidgeons of that "interesting result" I got years before ... can I "lay my hands" on that useful stuff, now - nooo, of course not ...
- over the years I have got so many interesting insights from simulations, things behaving in a non-intuitive way - and later on I recall in another situation smidgeons of that "interesting result" I got years before ... can I "lay my hands" on that useful stuff, now - nooo, of course not ...
Yes, the "kind" is absolutely key - unless I am 100% positive it is a negative difference I will ignore it - trying to grade differences which can be taken either way is a recipe for disaster, in my book. So how do I know it is in fact a negative? Normally by using difficult, "bad" recordings which very strongly provoke the misbehaviour - a classic example is sibilance, that people mention often - I would have on hand the recording with most extreme, in your face, version of that characteristic in the recording, and see how the system handled it ...
Difference vs improvement
When I use the word "difference" in a signal context I am referring to any change from the original. Without the original as a reference, you do not know whether you are changing the signal for the better or worse. Ideally the final signal/sound would be identical to the actual event (which we know cannot be).
Ultimately any "improvements" should be making the difference between the final sound and the input reference smaller. Quantifying the difference is at the root of my previous question to Pavel.
In general usage, the word "improvement" can be used subjectively, in an input vs output context the term "difference" should not, as it represents a quantified change from the original. However most of us here would probably view them as equivalent in a sound quality context.
Just my tired $0.02 worth...GN
Howie
Howard Hoyt
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org
When I use the word "difference" in a signal context I am referring to any change from the original. Without the original as a reference, you do not know whether you are changing the signal for the better or worse. Ideally the final signal/sound would be identical to the actual event (which we know cannot be).
Ultimately any "improvements" should be making the difference between the final sound and the input reference smaller. Quantifying the difference is at the root of my previous question to Pavel.
In general usage, the word "improvement" can be used subjectively, in an input vs output context the term "difference" should not, as it represents a quantified change from the original. However most of us here would probably view them as equivalent in a sound quality context.
Just my tired $0.02 worth...GN
Howie
Howard Hoyt
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org
- Status
- Not open for further replies.