DrWho Posted January 20, 2011 Share Posted January 20, 2011 Why does DBT always have to lead to crappy systems? My personal experience has been quite the opposite actually. Quote Link to comment Share on other sites More sharing options...
DrWho Posted January 20, 2011 Share Posted January 20, 2011 His philosophy would lead everyone into having a generic $500 stereo full of corporate components that sell by the pound. So you're saying the $500 system is NOT crappy, but the only reason an audiophile won't like it is because it's not crazy/unique enough? Quote Link to comment Share on other sites More sharing options...
Deang Posted January 20, 2011 Share Posted January 20, 2011 Never mind. Quote Link to comment Share on other sites More sharing options...
whatever55 Posted January 20, 2011 Share Posted January 20, 2011 His philosophy would lead everyone into having a generic $500 stereo full of corporate components that sell by the pound. So you're saying the $500 system is NOT crappy, but the only reason an audiophile won't like it is because it's not crazy/unique enough? No ... it is because it is only $500..... now if you spent that for an audio grade, gold plated, hi freq. fuse..... you got sumpin' there. [:'(] Quote Link to comment Share on other sites More sharing options...
Don Richard Posted January 20, 2011 Share Posted January 20, 2011 The most important step in any testing method is proper interpretation of the data that is generated by the test(s). Our good friends at Bose ran many acoustic tests prior to the release of their flagship 901 speaker system. What they noticed was that there was a lot of reflected sound in what we hear everyday. I believe the number "70% reflected sound" was floated around at that time. Then they went wrong. Someone presumed from that data that the loudspeaker should be constructed in a manner that would augment the reflected sound since we are hearing so much of it anyways. This effort resulted in one of the worst sounding and worst reviewed loudspeakers ever made. Folks who understood sound avoided the 901 and the great unwashed bought tons of them, and a great commercial success was born. So, you see, it wasn't the tests that they ran that were flawed but rather the interpretation of the data. Bose made a lousy speaker not because they ran tests, but in spite of them. Quote Link to comment Share on other sites More sharing options...
SSnyder Posted January 20, 2011 Author Share Posted January 20, 2011 So you're saying the $500 system is NOT crappy, but the only reason an audiophile won't like it is because it's not crazy/unique enough? hat saying is his has GOLDEN EARS but somehow under the seveer mental annguish(sp?) of controled sciencetific test the same GOLDEN EARS fail misurabley you cant' convinse the ill-loggical no matter how hard you tried Quote Link to comment Share on other sites More sharing options...
SSnyder Posted January 20, 2011 Author Share Posted January 20, 2011 Never mind. why bother too post at all if this is all Quote Link to comment Share on other sites More sharing options...
SWL Posted January 20, 2011 Share Posted January 20, 2011 At least it was almost spelled correctly. [:^)] Quote Link to comment Share on other sites More sharing options...
djk Posted January 20, 2011 Share Posted January 20, 2011 Why are we even taking someone seriously that is unable to spell, has poor grammar, and does not bother with capital letters or punctuation? Quote Link to comment Share on other sites More sharing options...
russ69 Posted January 20, 2011 Share Posted January 20, 2011 beets mee Quote Link to comment Share on other sites More sharing options...
Dennie Posted January 20, 2011 Share Posted January 20, 2011 Why are we even taking someone serious that is unable to spell, has poor grammar, and does not bother with capital letters or punctuation? Who's taking him seriously? [*-)] I should also say, English may not be his first language. I don't understand why he isn't involved in the (or any) conversation? This is his thread. I'm not one to talk though, English is my first (and only) language and I'm terrible at it. But I keep trying! Dennie Quote Link to comment Share on other sites More sharing options...
NOSValves Posted January 20, 2011 Share Posted January 20, 2011 I'm not one to talk though, English is my first (and only) language and I'm terrible at it. But I keep trying! Dennie Hey now lets just get something straight here....I'm the king of butchering the English laguage! I've been here nearly ten years and have only improved ever so slightly..... I don't get called out on it much anymore but I credit that to other folks on this forum learning my craiglish... Quote Link to comment Share on other sites More sharing options...
Dennie Posted January 20, 2011 Share Posted January 20, 2011 I'm not one to talk though, English is my first (and only) language and I'm terrible at it. But I keep trying! Dennie Hey now lets just get something straight here....I'm the king of butchering the English laguage! I've been here nearly ten years and have only improved ever so slightly..... I don't get called out on it much anymore but I credit that to other folks on this forum learning my craiglish... Sorry Craig, Steve(SSnyder) is winning this one! [:#]Dennie Quote Link to comment Share on other sites More sharing options...
NOSValves Posted January 21, 2011 Share Posted January 21, 2011 Well I can't agrue that one... Quote Link to comment Share on other sites More sharing options...
edwinr Posted January 21, 2011 Share Posted January 21, 2011 The Underlying Flaw in All AB-style Testing Every form of comparative listening test such as AB, AB/X and DBT suffers one catastrophic flaw that renders the results close to useless. The flaw exists in spite of the overwhelming scientific evidence confirming the flaw. What's wrong is the assumptions used about listening and perception. The model which anchors comparative testing is that human perception, and ultimately preference, is divided into discrete channels much like independent files or directories of some hard disk recorder in the brain. And as such, the test attempts to isolate the "sound channel" from all others by using a black box model of testing. Black box means the entire sound producing mechanism is unseen and unknown ("behind the curtain"), and only the output (sound) is isolated and fed to the subjects. This concept of the brain is wrong and is nothing at all related to how perception works. So the results of any such test have extremely limited use. DBT testing is a cornerstone of science, and is very useful. We get most of our medicine through that process. But you have to interpret the results in a meaningful way. Used in listening tests, the conclusions that are generally reported are of narrow use when it comes to building a home sound system. Perception is Selective First of all the brain is not a hard drive recorder for perceptions. All inputs to the brain must be heavily filtered, otherwise you would freeze up from overload. What you are aware of consciously, is perhaps one billionth of the total perception information being streamed into to the brain. This billionth portion is not always the same portion. It will vary depending on total circumstance, which includes all internal and external stimuli. If you get this sensory stream now "ABCDEFGHIJK" you might be conscious of just "CFK". If you get that stream later, your may be aware of "ADG". So, any attempt to form a consistent judgment about the input stream is skewed by this selectivity. Perception is Entangled The "brain as tape recorder" metaphor has to be ditched. It is no such thing, and doesn't in any way model a tape recorder. Nuerons are cells that are entangled not only with each other, but entangled with the external environment. It is NOT equivelent to a "bit" in digital terminology as many imagine. It is not per se a storage device. It is a massively complex intelligent machine by itself that is a microcosm of the entire human body. It is accepting thousands of signals and inputs and then making proteins and sending signals in response to inputs. Entanglement here, means that neurons are not divided up into discreet sound-neurons, sight-neurons, odor-neurons, touch-neurons. All of our senses "mix" because the neuron network is entangled to do just that. Furthermore, the neuron, like all cells, is accepting EXTERNAL signals from the envrionment, which cause changes to the neuron's output signalling and protein making. With this model, it is now clear that when you "listen" to a stereo at home the judgement you make about the "sound quality" is the result of an entangled process of perception that includes more than sound pressure at the ear. That process will include a multitude of inputs which are not sound. The floor pressing up on your feet, the slight draft in the room, the many odors, the lighting conditions, the ambient temperature, the ambient noise, your internal emotional state, your hydration level, your hunger level, pain level, and literally thousands more conditional inputs to the brain. There is no brain in which you can perfectly isolate discrete phenomena from all other phenomena. It simply doesn't function that way. (Yoga is the practice of attempting to do just that very thing.) "Sounds Better" - What Does That Mean? Sounds better is the qualitative value we use most commonly to engage in the hobby. We buy X because it sounds better. We do Y to the system because it sounds better. We hear things and proclaim, A sounds better than B. How do we arrive at the qualitative judgement we call "sounds better?" It can only happen through the entangled process of perception. We have no other tool in our human toolbox. We are incapable of shutting off all inputs except for the "sound channel," and then doing an internal string comparison of two sound streams. This is what the DBT testers attempt to do. We are incapable of that because we are not made in that fashion. There are no equivelents to recorders, oscilloscope and voltmeters in the brain. It is invalid to suggest that such homologues exist. Try this. Listen to your stereo for half an hour, and then take onboard some mood elevating substance like alcohol, pot, X or the drug of your choice. Nearly always the system will "sound better." Clearly your system didn't change, you changed your inputs to the neuron network. You added a new chemical stimulus to be integrated with your other perceptions in your brain. Your new judgement about the sound quality is a synthesis which now includes the drug inter-action. You have no choice, that's how it works. You would have no means of "ignoring the action of the drug" to decide about sound quality. That's an example of putting a very heavy thumb on the scales. We expect those results. But let's try a lighter touch. See if this rings a bell. You are looking at your system and see that after a year it has become dusty, disorganized, cables dangling here and there, dirt on everything - untidy. You get disgusted and clean it all up. You clean the equipment, the shelves, the glass. You tie up all the dangling wires and cables and make everything neat and tidy. You polish the speakers, clean the cathair off the grill and so on. Now, you sit down to listen and suddenly it "sounds better." Why? Because it looks like it should, and you can not isolate that "good feeling" from what you are hearing. Your neuron net won't let you do that. It works exactly the same as the mood altering drug. Inputs change perception. This isn't a hypothesis, it is a fact. And the range of inputs is far, far greater than we can usually know at any time. The body internally is generating thousands of signals about its own condition. The external environment is broadcasting thousand more into your cells. Your trillion cells are individual intelligences reacting to all these inputs snd signals. Then the brain integrates this into judgements as you demand it to do so. "Sounds better" is one such judgement that arises from all this input. So, in our normal actions of the hobby, what influences the "sounds better" judgement? Here are a few well understood motivations for preferences of consumer goods: Aesthetics Status of the brand Physical attributes like size, color, weight even odor Personal emotional baggage (e.g. Grandpa drove Buicks) Past experience with brand Expert opinions Peer pressure Utility (specs are better) Cost Now, that's a small portion of the real list. But just those alone form an amazing part of our judgement. What needs to be understood is that these are REAL inputs to our conclusion that "A sounds better than B." They are not parts that can be hacked off and isolated when we make real-world judgements. "Feeling better does make things better." That is not a hypothesis, it is a fact. In DBT testing, they try to hack-off as many of those "external biases" as possible in a hopeless attempt to find meaning to isolating your "sound channel" - your imagined tape recorder. People regularly fail such DBT testing because in truth, it isn't how their perception systems operate when they demonstrate preference. The black box model has no actual purpose outside the DBT test itself. It simply is unrelated to how people form judgements about audio gear. Even the testing design itself is physically flawed. The basis is this: Stimulus Stream A is given to the subject at time = T1. Then at time = T2 Stimulus Stream B is delivered to the subject, and at time = T3 the subject must decide if A was equal to B, or if A was not equal to B. You don't need any audio equipment at all to show how easily people fail this test. You can perform the test on everyone you meet. Do this: Recite the two streams aloud, one after the other, to any subject you choose, and ask them if they are the same or different. Stream A: Kentucky mary george iowa henry alaska lanny albert maine robert Stream B: Kentucky mary georgia henry iowa alaska manny albert maine robert Now, this is a very simple set of very short inputs. Nothing near as complicated at listening to a passage from Sgt. Pepper. This sample has a distortion figure of around 5% between A and B. That should be totally easy to detect. And yet, I think you'll record no better than coin toss odds of right answers if you ask enough subjects. The brain is simply not a recorder in the sense that such tests imply. A famous test of visual perception is the Simons-Chabris "gorilla film." In which subjects were told to watch a one minute film and count the passes made with a basketball between players. During the film, a woman dressed in gorilla suit walks into the center of the film, waves to the camera for 5-seconds and walks off. More than half the subjects who watch this film never see the gorilla! The brain is not a video recorder, nor is it an audio recorder. When subjects fail DBT testing are they simply failing to see the gorilla, or is there no gorilla to be seen? The problem is that people's real-world preferences of audio gear is generically reported as "sounding better," but in fact sound is but one attribute in the entire stack of contributing inputs to the preference. If all gear was DBT tested, and you bought the system of gear which tested best, would you think it "sounds better" than a system you selected by your own regular method? If they delivered that system and it turned out to be an $100 LG receiver with a $29 Samsung CD player, and speakers made by Bose - let's just say - would you happily ditch your Khorns and monster amps and such? A few people are going to be stubborn and say "yes of course", but I think they would be a small minority. DBT testing of audio gears can tell you a little bit about how perception works in the human brain: "Big news - people have trouble comparing input streams displaced by time!" It can not tell you anything about what audio gears you would enjoy having in your system. DrWho-- Here's the post in which I answered your question about generic stereos. I'm quoting this whole post because Mark (and I mean this most sincerely) you write really, really well. I think you've missed your true calling in life. Maybe you should write for an online magazine or something. Quote Link to comment Share on other sites More sharing options...
Islander Posted January 21, 2011 Share Posted January 21, 2011 I'm quoting this whole post because Mark (and I mean this most sincerely) you write really, really well. I think you've missed your true calling in life. Maybe you should write for an online magazine or something. I agree that Mark writes really well. That was very informative and easy to follow. [Y] Quote Link to comment Share on other sites More sharing options...
DrWho Posted January 21, 2011 Share Posted January 21, 2011 You say DBT doesn't work because of the "resolution" and "error factor" of human perception, or whatever you want to call it...but that limitation always exists, and refinement is all about learning to hone in one's perceptions and filter out distractions. Heck, if anything, your examples and talk about perception limitations would actually suggest removing as many variables as possible that aren't audibly related....like removing your sight, for example.... As far as the cheap gear on Bose vs good gear on Klipsch, you'd have to be deaf not to hear a difference in a DBT. I'll go a step further and mention that I have no problem blindly differentiating between amplifiers, just to list one example that many people often fail blind tests with. Why would I be interested in the justifications of one that can't hear the differences? I'm not disagreeing that other factors are impacting one's enjoyment, but there's a difference between "sounding better" and "more enjoyable". If you don't have the capacity to refine your listening, then you're just prescribing to a socially constructed image of refinement without actually experiencing the refinement itself. You see this all the time in the car world where people buy and talk about stuff because they're ascribing towards some culturally promised experience, but in all reality their ignorance mixed with a desire for speed has them doing constant burn-outs and power slides, which is one of the slowest ways around a track. Just compare the popularity of racing videos on YouTube...the guys demonstrating crazy talent are found less popular by the masses. If the masses rode in a car driven with talent versus the same car with a driver power slidkng the whole time, the masses are gonna think the talented driver was slower...but do we go by their perception of speed or the actual track time itself? The same analogy applies to audio. One type of tire might be faster in the power slide, but at the end of the day there are 10 seconds left on the table with serious speaker and room limitations.... Quote Link to comment Share on other sites More sharing options...
Don Richard Posted January 21, 2011 Share Posted January 21, 2011 DBT will show conclusively that the sense of "hearing" alone, is not how people make decisions about what they think "sounds best" in their stereo. Audio professionals such as recording engineers and real musicians are very good at focusing on the task at hand. Their ability to listen into the music or listen into the mix far exceeds the abilities of the general population. Musicians will record take after take until they are satisfied with the results. Mixdown engineers will work on a mix for hours or days getting reverb tails just right, things the general public will never even get a chance to hear on their Ipods. Listening evaluations done by professionals are more likely to have results that coincide with an audiophile's preferences than evaluations using the general population as test subjects. The general population is too easily distracted by non-audio events to give meaningful results in audio testing. Quote Link to comment Share on other sites More sharing options...
SSnyder Posted January 21, 2011 Author Share Posted January 21, 2011 Musicians will record take after take until they are satisfied with the results. never met one audio-phil musican... ... ... ever >>>>. . . . onley care about geting all a the notes corect Quote Link to comment Share on other sites More sharing options...
Don Richard Posted January 21, 2011 Share Posted January 21, 2011 Musicians will record take after take until they are satisfied with the results. never met one audio-phil musican... ... ... ever >>>>. . . . onley care about geting all a the notes corect Barbara Streisand is famous for recording as many as 65 takes of one song and incorporating as many as 30 snippets of these tracks into the final mix using the flying switch capabilities of a modern recording studio. She may not qualify as an audiophile but she certainly does as a perfectionist. Guitarists will generally make multiple tracks of their part and the producer will then pick the track they like the best. And so on for all of the performers. . onley care about geting all a the notes corect I specifically said "real musicians" earlier in this thread, not the doped up clowns who can barely play. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.