how to fix that problem
yeah...but you probably got all the volumes set the same...😉😉
twiddle some of them knobbie thingies and report back...😀
John L.
yeah...but you probably got all the volumes set the same...😉😉
twiddle some of them knobbie thingies and report back...😀
John L.
uuhhh
it's spelled e-x-p-e-c-t-a-t-i-o-n ... (bias,effect,error,mistake,etc.) pick your noun, if you dare...
well known to the research community...🙄😉
John L.
it's spelled e-x-p-e-c-t-a-t-i-o-n ... (bias,effect,error,mistake,etc.) pick your noun, if you dare...
well known to the research community...🙄😉
John L.
Agreed. And better, if the products didn't come with such baggage in the first place.
se
I will try to say something that is hopefully useful. I really don't find that double blind ABX type tests to be revealing, but tend to make EVERYTHING sound essentially the same. I have looked fairly hard at this problem.
First, I found that typical ABX testers were made with relays that I found 'opaque' even in open listening.
Second, I found that differences that were apparent when A or B were kept track of by the listener, could be heard, but these differences disappeared when ABX was invoked, even though I did not know, in advance what A or B really was.
Third, most ABX tests usually use audio reproduction equipment that most of us would openly reject just listening to it, so how can we make comparisons with it?
This worship of the 'null' is just like what we once we did with 'witches'. We tied them up and threw them in a lake. If they floated, they were witches, if they sank, then God would let them into Heaven, I presume. What a choice. I find the ABX test is just as arbitrary.
First, I found that typical ABX testers were made with relays that I found 'opaque' even in open listening.
Second, I found that differences that were apparent when A or B were kept track of by the listener, could be heard, but these differences disappeared when ABX was invoked, even though I did not know, in advance what A or B really was.
Third, most ABX tests usually use audio reproduction equipment that most of us would openly reject just listening to it, so how can we make comparisons with it?
This worship of the 'null' is just like what we once we did with 'witches'. We tied them up and threw them in a lake. If they floated, they were witches, if they sank, then God would let them into Heaven, I presume. What a choice. I find the ABX test is just as arbitrary.
Last edited:
Then what explains these:
ABX Double Blind Test Results: Tape Decks
ABX Double Blind Test Results: Level Mismatch
ABX Double Blind Tests: CD Players & DA Converters
ABX Double Blind Test Results: Tape Generations
ABX Double Blind Test Results: Phono Cartridges
Dan
edit: did anyone check this out? http://www.youtube.com/watch?v=BYTlN6wjcvQ
Last edited:
I done twiddling John. I report back now. I noticed that when I turned the input knob to the right it done get louder. The opposite happened when I done turn the knob the other way.
I do. I think everyone should move on to somewhere they can do something productive.
What I want to know is what shows up in an ABX test, except level, polarity (maybe) frequency response, and GROSS distortion? We have been getting these right for the last 35 years or more in all of my comparison tests. I'm not saying the everything that is made is that accurate, but my stuff certainly is, yet I hear differences in my own designs, from the original JC-2 (the prettiest of the bunch), the JC-80, the CTC Blowtorch, and the new Parasound JC-2. Just my imagination, I guess, except that doesn't make the differences go away, so I keep on making audio equipment.
I've been burned out, and had to start from scratch. I then TRIED to use Radio Shack interconnects for almost 1 year, but sooner or later, I got rid of them, and I put them in my lab as test interconnects. Unfortunately, accidently I measured the RS interconnects, all by themselves, and they showed up as having distortion of their own. What a concept, and what a hassle, because it meant that I could not trust these interconnects for serious distortion measurements. And so it goes, but of course, it is just bad measurement technique, and listening expectations.
I've been burned out, and had to start from scratch. I then TRIED to use Radio Shack interconnects for almost 1 year, but sooner or later, I got rid of them, and I put them in my lab as test interconnects. Unfortunately, accidently I measured the RS interconnects, all by themselves, and they showed up as having distortion of their own. What a concept, and what a hassle, because it meant that I could not trust these interconnects for serious distortion measurements. And so it goes, but of course, it is just bad measurement technique, and listening expectations.
Oh man, am I still on ignore?? Surely not from everybody??
Has a time been set now?? Crikey, the only reason I have kept up to date with the thread is because (hallelujah) someone had the brass monkeys to put 'em on the line. NOW I know why Tom is often sore, he has bigger balls than most!
A new thread has started?? Can we have a link to it please??
JC (tho I think you too have me on ignore), you don't like DBTs. OK, as a statement of preference I fully accept that you think them useless.
(this applies to others) Can I ask just how then you could go about showing audibility?? It's a serious question BTW, not trying to ambush you or ridicule you.
:Leave aside any notion of 'I have no desire to prove to others' etc, how could we go about it to get proof as strong as 'Yes, you DO see that apple over there.'
Can it be done via different methods?? Would any part of your method require knowledge of the cable being tested??
Genuinely curious question, to you and others in the same 'dbt is unreliable/misleading' boat.
Nah, this whole knob twiddling thing is just too far fetched to be plausible
file with magic stones.
I think this is a misinterpretation of a DBT (if I understand you correctly). Its a person who is making the judgement about whether a difference can be heard - you can't "remove the person" from the test. The protocols of the DBT are designed to give that person the best chance of making a valid choice with known biases removed.
Well, its not "passing a test" as if there is a right or wrong answer.
We need to look at the claims we are seeking to test.
If the claim is "a certain % of the population can tell the difference between cable A and cable B" then you would need:
a) a valid DBT
b) a valid number of people drawn from the target population who are representative of the target population and who record a positive result in the DBT (allowing of course for the usual stat tests to make this judgement).
But the test being organised by SY is not making claims about the population as a whole. Just the question: "can a difference be detected between cable A and cable B". The person who is doing the listening claims they can (prior to the DBT). The DBT removes the biases and we'll see if the difference result still holds.b) a valid number of people drawn from the target population who are representative of the target population and who record a positive result in the DBT (allowing of course for the usual stat tests to make this judgement).
So what inferences can be drawn from the result? If there is no difference detected then:
a) the person heard a difference prior to the DBT but they are not able to once bias factors were removed. For this person, there is now no audible difference between the cables. (From a self knowledge point of view its a wonderful opportunity to gain insight into how bias factors can screw up our perceptions.)
b) the two cables in this particular instance are not audibly distinguishable by this person. If this person's hearing was representative of the population then you might be drawn to make some claim about the general population ("most people couldn't hear a difference between cable A and cable B") but that would be a long bow, and subject to a large error term.
b) the two cables in this particular instance are not audibly distinguishable by this person. If this person's hearing was representative of the population then you might be drawn to make some claim about the general population ("most people couldn't hear a difference between cable A and cable B") but that would be a long bow, and subject to a large error term.
If there is a difference detected then the claim "all non pathological cables that measure within the same audio tolerances for inaudibility sound the same and the differences are inaudible" is falsified. And it only takes one positive result to establish that via a well designed and implemented DBT.
Of course that claim is more far reaching then one which is just "there is no audible difference between these two particular cables". A positive result also of course falsifies that claim but its not a very enlightening result compared with the broader claim.
Overall its worth remembering that this is not the only DBT that's been done on cable/interconnects audibility. Also, audio engineering knowledge supports the claim that a priori we should not expect a difference. But the DBT in effect puts all that aside and sees whether there is a difference to be detected.
That may be true (and I agree) but that's not the point, is it. The point is whether someone can hear a difference between cables. It's immaterial how he or you or I expecrience that difference (if any) or music, come to think of it.
Nice distraction though 😉
jd
OK, let's put you out of your misery.🙂
AFAIK, no specific date has ever been set. There were a couple of delays due to changes in Tom's system. IIRC, the current agreement is that Tom will let SY know when he feels he's ready for the test, and then they'll arrange a date.
I don't remember a separate thread being started for Tom's DBT. However there was an attempt to organize an "audibility of cable burn-in" DBT in another thread, with Andre as the guinea pig. Perhaps that's the one Steve was thinking of?
Whatever happened to that one anyway? A lot of organization went into it, Andre was ready and waiting for the cables to be sent to him, and then ... nothing happened.😕
Man, I LOVE you haha. Thanks for that.
I am aware that Tom put it off recently (new cdp was it not?) but the new talk of a date had me wondering.
If andre does something, are you close enough to be part of it??
andre, fill us in on the goss (for those of us who have not seen that other thread)
Dunno. It didn't need local help though anyway, since the cables look the same before and after burn-in. The idea was to send cables marked "A" and "B" and see if he could tell the difference. If so then more would be sent to see if he could accurately identify which were burned in.
The thread's here. I had high hopes of something actually happening there but it just fizzled out.
My
: If cable differences are audible, I'd expect it to be easier to tell the difference between two different cables (as in Tom's test), than between two similar cables with and without burn-in.Actually, cable burn-in takes the normal "subjectivist" vs "objectivist" argument to another level.
The standard "objectivist" response to component burn-in is that the sound of the component in question does not change, but that it takes a while for the listener to adapt to the different sound of the new component.
However that argument can't be used if the component in question is a cable, without acknowledging that the new cable sounds different to the old.
Pass the popcorn. 🙂
Dunno. It didn't need local help though anyway, since the cables look the same before and after burn-in. The idea was to send cables marked "A" and "B" and see if he could tell the difference. If so then more would be sent to see if he could accurately identify which were burned in.
The thread's here. I had high hopes of something actually happening there but it just fizzled out.
My
Actually, cable burn-in takes the normal "subjectivist" vs "objectivist" argument to another level.
The standard "objectivist" response to component burn-in is that the sound of the component in question does not change, but that it takes a while for the listener to adapt to the different sound of the new component.
However that argument can't be used if the component in question is a cable, without acknowledging that the new cable sounds different to the old.
Pass the popcorn. 🙂
Cable burn-in takes time and it happens(if it happens)gradually,so the level of change cannot be easily be known or estimated.On the other hand directionality differences are there(if they are there)always no matter if the cable has burn-in or not,therefore they may be easier to check.
stating the obvious... why?
so...let me get this straight... doing really bad poorly designed experiments with known defective equipment using lousy sources produced useless results; QED: DBT/ABX/tests in general have no validity...... is that what you're saying?
Thanx for the insight... don't quit your day job, tho
Give the man a sucker, his wisdom is revelatory!!
John L.
so...let me get this straight... doing really bad poorly designed experiments with known defective equipment using lousy sources produced useless results; QED: DBT/ABX/tests in general have no validity...... is that what you're saying?
Thanx for the insight... don't quit your day job, tho
Give the man a sucker, his wisdom is revelatory!!
John L.
Last edited:
I'm no statistician, but I thought that a single result is not statistically relevant - i.e, it could just be luck!
Exactly. And no amount of navel gazing, armchair philosophizing, excuse-making, or lengthy discourse on irrelevancies can change that. Not ONE person has been able to demonstrate that they can hear these magic differences. And unlike the cowards who dance the superior dance but won't put their beliefs to the test, Tom has stepped up.
He's delayed the test start, but both in public and in private, he is sincerely trying to get ready- acclimating himself to his new system and practicing. He's NOT making excuses or stalling, IMO. Whenever he's ready, and I think it will be soon, we'll do the test.
As I've said before (but need to say again before all of the misquoting and downright lying begins afresh), a null result proves nothing, other than one more bone on the pile. It will do nothing to convince the faith based, any more so than turning up yet another rock and not finding a leprechaun (thanks for the analogy!) will convince the leprechaun believers that leprechauns are mythical. On the other hand, a falsification of the null hypothesis IS significant, and should cause any honest skeptic to reconsider.
- Status
- Not open for further replies.