Thanks for the advice KBK
But I'm not interested in 'winning' arguments - I'm more interested in truthful exchanges & truth about audio perception - I can't agree with a person's point, just to sway the discussion my way, when I don't think it is correct. I've been under no illusion that I could persuade those people who are the main debaters opposing my view but I would hope that some readers weigh up the points made on each side & judge for themselves? Hopefully, I can trigger some thinking outside of what passes for 'accepted wisdom' as I believe we still have a lot to learn about audio & the issues of measurements & DBTs are at the heart of the understandings about auditory perception & why the 'accepted wisdom' tends to be a hindrance to progress rather than a help
Sorry but that's just the way I am.
Another funny thing is that I'm not saying anything different than Jakob2
But I'm not interested in 'winning' arguments - I'm more interested in truthful exchanges & truth about audio perception - I can't agree with a person's point, just to sway the discussion my way, when I don't think it is correct. I've been under no illusion that I could persuade those people who are the main debaters opposing my view but I would hope that some readers weigh up the points made on each side & judge for themselves? Hopefully, I can trigger some thinking outside of what passes for 'accepted wisdom' as I believe we still have a lot to learn about audio & the issues of measurements & DBTs are at the heart of the understandings about auditory perception & why the 'accepted wisdom' tends to be a hindrance to progress rather than a help
Sorry but that's just the way I am.
Another funny thing is that I'm not saying anything different than Jakob2
the ear/brain combination has a pre-sound list to speed up speech cognition. Where we each hear the beginnings of a word and then overlay the known word on top and hear that, if the initial fit is good enough. We then construct the message more quickly. Part of the human sensory library system. That thing about our acts being about 80 milliseconds behind the original impulse, and so on. the body is rife with these systems. Feel, sound, touch, cognition, emotions, sight, etc. Learning is when we defeat such systems and learn anew. This is the core point of why the majority is wrong, to the tune of 97%.
Pareidolia is principally named after the concept as expressed in human sight, but it is paralleled throughout the human nervous sytem.
Placebo has some solid aspects tied to this, ie, inflammatory response, and so on. Placebo can work as medicine, even though yes, placebo is placebo. Depends on where we're each coming from in our application context of the usage of the word.
Hypnosis, for example and how it can enable adrenaline and other aspects that can make for superhuman strength attempts that will break the body. Placebo can do some amazing things, if viewed down that sort of path. As, what is hypnosis, exactly? Good luck defining it in the real world, outside of a dictionary. The more complex the question the greater the splay of required areas of expertise in order to define it well enough to solve...
The trick in hearing in audio is to not have the known sound forms overlay the heard sound and substitute for the new data that would be there..if it is not overlaid by the pre-load in the brain... of the known and patterned sound. humans and music are unique in the animal world. It is probably tied to this as it is near permanently sheep dipped in the ego I/O loop. Human existence from second to second relies upon these systems always contributing in an anticipatory act of humming a few bars.
To learn as far and as best as is possible, one has to break themselves like a frightened child across the knee of a madman. The esoteric types inform you of this basic truism, so they should never be wholly dismissed.
Due to this given above point.... at the peak of what might be described as esotericism, you will find levels of intelligence and informed intellect, that easily exceed that which has ever existed in the realm of a scientific world.
Last edited:
As an audio manufacturer, my 'bias' should be obvious. However, I want to thank 99 and Jakob for putting straight facts about ABX testing forward here in a learned and thoughtful way. Of course, it should be obvious that ABX testing gives false negatives, and we have stated that for the last 40 years, but some people are 'biased' toward wanting audio products to all sound the same, and they will promote ABX even when it gets in the way of progress.
Last edited:
The Clark tests were numerous and are a matter of public record, and I experienced many of them firsthand, having been the engineer setting them up. I have also conducted tests involving hundreds of recording engineers at AES and ITA conventions, some being well-known engineers who possibly made the material you use for references. Have you any actual experience which is the equal? These tests were conducted some 15-20 years ago and I do not have the test records, perhaps Richard Clark does but he is not in the audio industry anymore.
I have voiced my objections to aspects of AB and ABX testing, and agree it can deliver a preponderance of negative results if not conducted correctly, however my actual experience (not conjecture) says when there is indeed a difference between A and B these tests do deliver positives.
You can conjecture that the A and B sources must have to be very different for a well-conducted ABX to show positives, but you would be wrong. When levels are matched yet the sources not identical in some respect and a broad range of material is presented over time, a trained listener will indeed be able to render a positive result, I have seen it happen many times. Some people are so good at critical listening they achieve near perfect scores repeatedly while 99% of others do not. As a result of my involvement I have a few conclusions of my own:
1) For an ABX test to give more than null results, the listener must be trained, rested and relaxed and have time to audition as they see fit.
2) Other than for research purposes there is zero reason for a bias to be intentionally introduced such as the ability so see what one is listening to. This type of test is then a control for evaluating factors affecting the test, not a valid test of listener sensitivity.
3) Many adults have damaged or degraded hearing so the sensory apparatus is poor, yet they have not had it evaluated and believe their hearing is as good as anyone else.
4) Many audiophiles and recording engineers have a greatly inflated idea of their own critical listening abilities. Additionally most of the most consistent positive results we saw were by recording engineers who master highly regarded recordings...If these tests were so biased for null results, how did they do that? Perhaps it was the training and years of experience?
We concluded that being trained to listen for specific problems was the #1 factor affecting test results. For example, a Dolby engineer was perfect at picking out slight Dolby mistracking, a vinyl mastering engineer easily picked out slight inter-groove crosstalk the rest of us did not, etc... Once again, when pointed out the issue was easier for the rest of us to notice. I am sure you amp designers have developed the ability to hear minute amounts of crossover distortion which most of us would miss.
If one of you thinks there has to be a better way to conduct testing, please propose, conduct and report a new type of test, we would then have actual data to discuss. It is easy to sit back and critique the results of others tests, but it would be much more helpful to do tests yourself and let us know how they come out...like PMA is doing.
Cheers!
Howie
Thanks for the cite Pavel,
I am afraid you missed the claim that it puts you in the fifth row of Carnegie Hall. This may actually be believable! Fifth row center is not so good. Not centered is really not so good. Best seat in the house is front center balcony!
I can't do an ABX test of this, but just wandering around during a rehearsal makes this clear to my ears.
@KBK, Some very good points. But, if you are in a position to see all that you can probably also see saying it once won't change much, if anything. You could have a lot of work ahead if you if you actually want to have an effect.
EDIT: However, some of what you say seems a bit overly dramatic.
EDIT: However, some of what you say seems a bit overly dramatic.
Last edited:
Which bears exactly zero relationship to the discussion and you know it. An F1 driver is closer to a musician telling his guitar tech how he wants things set up as at the end of the day 99% is the simulations, wind tunnel testing etc and the last 1% is getting the balance the driver prefers. And there is a direct feedback, as the driver gets faster the closer that balance is to his ideal. If the driver says he 'feels' faster but the stopwatch says differently he will be told he's wrong.
I didn't look at the Youtube video yet but from your description above I feel there is a deeper understanding of perception needed. Our auditory perception works in two modes, bottom up & top down. Both modes are concurrent & active all the time, not one or the other active. Bottom up processing is when the brain processes the nerve impulses coming from the auditory pathway & top down mode is when the brain predicts what signals should next arrive along the auditory pathway based on patterns, models statistical averaging that is being continuously compared to the incoming signals. So it is using predictive modelling. As I said before auditory processing is a continual best fit process. As you say this reduces the workload & speeds up processing (biology always uses the most efficient means to achieve a result - due to limited resources).
I can give the research papers on this if anyone is interested.
One technique used in auditory research is called mismatch negativity (MMN) where it can be seen when the brain EEG (FMRi? can't remember) shows that an expected pattern of sound signals delivered in the auditory pathway. has deviated from the expected pattern
One of the reasons why we are so keenly aware of the attack of a sound - we categorize the sound based on this fast rising (usually millisecs) attack portion of the sound & this narrows our internal pattern search space.
Yes, this predictive system can get it wrong but I would suggest that it is not the "trick in hearing" - we can't avoid this prediction - it is a fundamental mode of operation of auditory perception (& I suspect other perception) that can't be negated
Last edited:
I don't agree with the bolded bit of your statement but perhaps the rest is correct - I don't know - I have't been involved in F1 driving but what you say seems to make sense
I mean it's the certain beauty of competitive sports -- the end measurement (time/finishing place) cuts through the fog of ambiguity so well.
That said, I don't think your analogy is quite perfect there, Bill. 1% is probably a bit of an exaggeration, but, yes, it's a data-driven sport.
Telemetry is such a huge thing in F1, but historically (and depending on the race series, now), driver feedback has been critical to suspension setup/etc. And there are some drivers (can't remember names, sorry) that were/are famous for their ability to enumerate the feel of the car and translate that into setup changes that corroborated extremely well with what the telemetry told them as well.
But it's certainly *not* screwing around, like so many examples of voicing here on diyA exemplify. (but certainly not all)
That said, I don't think your analogy is quite perfect there, Bill. 1% is probably a bit of an exaggeration, but, yes, it's a data-driven sport.
Telemetry is such a huge thing in F1, but historically (and depending on the race series, now), driver feedback has been critical to suspension setup/etc. And there are some drivers (can't remember names, sorry) that were/are famous for their ability to enumerate the feel of the car and translate that into setup changes that corroborated extremely well with what the telemetry told them as well.But it's certainly *not* screwing around, like so many examples of voicing here on diyA exemplify. (but certainly not all)
Last edited:
Sorry about last weekend, it all got a bit fractious for me. I'm not used to being selectively picked to pieces and reassembled in a form that I do not recognise as being me. It's horrible, and the only thing I can do is stand up to being bullied. Even so, I felt sullied by the experience. I know you people are clever and erudite, and I don't have an agenda of any kind. I am just interested in what is being discussed, but the way in which it is being discussed is, at times in what seems like a mosh-pit, and it is not very nice. So, please be nice to me, and especially so when I think out loud outside the box, because that's what we're here for, isn't it?
I warned you. This is a form of audio related punishment. It will keep you strong though if it doesn't kill you.
It takes one to know oneThat's for gpauk, before you get paranoid tapestry
Paranoid? Moi? - NEVER!
- Status
- Not open for further replies.