|   An open letter to a cable company | Warren Warren
Jun 7, 2002 2:24 PM | | Please, if anyone responsible for the design/sale/marketing of audiophile cables is reading this, I would be more than willing to help design and administer a double-blind test that would certainly meet naysayer protocols and yeah-sayer standards, and we could put the issue to rest one way or the other. I would like to use high quality components and compare the cheap black and red interconnects and zip cord to your brand of interconnects and speaker cable. I offer myself and my system towards the furthering of this goal. I know that some of you out there (i.e. audioengr) work in the aftermarket cable industry. Please contact me via this thread if you are interested in conducting a listening test with a population large enough to give an accurate statistical sample. |
| Is this an intelligence test? | sam9
Jun 7, 2002 3:04 PM | | The chances of betting a really bright cable maker to bite on this is rather slight. |
| Are you so confident... | Warren Warren
Jun 7, 2002 5:24 PM | | ...that there will not be a statistically significant result that you think the cable manufacturers already know this? I would bet you are wrong, but that's why I made the offer. |
| Are you so confident... | sam9
Jun 10, 2002 7:38 AM | | As Norm Stong writes this is an asymetrical proposition. Unless the cable mabufactures have already run such a test with results that are unequivably in their favor, they are better to lest this offer pass. In game theory, by decioning they are taking a Mini-max stategy -minimize the mazimum risk. |
| Is this an intelligence test? | Norm Strong
Jun 7, 2002 5:51 PM | | I'd be surprised if even one of the less brilliant cable makers bit.
What's the upside? Corroborating what your customers already think is the very best you can hope for.
And the downside? Disaster. You're out of business. |
| re: An open letter to a cable company | 1par
Jun 7, 2002 3:52 PM | | What area of the country are you in? If you're in the SF Bay Area I'd like to participate/assist. |
| Sorry, but... | Warren Warren
Jun 7, 2002 5:22 PM | | I was in SF last year. Probably the best city on earth to be a straight single male. Too bad I'm not single, as all those poor girls out there are definitely lacking guys that like them. Also, my 801's would be a little hard to transport to the west coast, since I'm in DC. If you find yourself out this way, look me up.
WW |
| Sorry, but... | mtrycrafts
Jun 7, 2002 5:45 PM | | Here is a link to a prior attempt by someone well known and it fizzled:
<http://www.avahifi.com/cables.htm>
However, if you have something in concrete, I am sure I can convince Tom Nousaine to attend either as an observer or to participate in ensuring that it is properly conducted. |
| re: An open letter to a cable company | eV
Jun 7, 2002 5:40 PM | | Warren,
I think it is cool that you want to try this out. But a couple things. How many participants do you plan on using? I believe a good sample is also dependant on quantity of listeners. One or two listeners doing many tests are not a good sample in my opinion. A good sample is a good quantity of listeners also. And please, before the test is run, everyone must have a hearing test. Otherwise, the test is void before it even starts, if you are trying to prove differences. Also document everything you do. There is no such thing as too much information when documenting a test. You just need to cover all the bases.
Do you plan on manually switching cables, or using a switch? If you plan on using a switch, ensure that all levels and FR are equal pre and post switch, and between channels used. It would probably be easier to manually switch.
I applaud your determination in this matter. But as I have said before, and we disagree on this, I have not been convinced that DBTs are sensitive enough for subtle difference tests. The similar color shade analogy is a reason I have this opinion.
Anyway, good luck with the test, and have a little fun if you can.
eV |
| re: An open letter to a cable company | mtrycrafts
Jun 7, 2002 5:51 PM | | b But as I have said before, and we disagree on this, I have not been convinced that DBTs are sensitive enough for subtle difference tests. The similar color shade analogy is a reason I have this opinion.
Not a good analogy at all, nor is it a very good test if one can identify subtle changes in color or how subtle one can identify. You would need several color chips, some same and some different and see if one can consistently identify a difference in thecorrect one, dbt, of course, without the tester knowing which is which. So no, your analogy to supply two chips to be know to be different is not a valid test, is it? And, depending on the subtility, some may not see it, regardless how good their color vision is.
As to audio DBT and subtle difference, you should really contact a credible research institute, such as the Canadian Communications Research Council or the National Research Council in Ottawa who uses DBT as a necessity to identify subtle differences. They should hav the weight behind them. Don't take my word, please check it out, for yourself. |
| National Research Council in Ottowa | Warren Warren
Jun 7, 2002 7:32 PM | | Why should I use a Canadian facility when they are not even smart enough to live where the weather isn't below freezing 9 months out of the year? LOL- Just a poke at MM and our other Cannuck freinds out there! ; ) |
| National Research Council in Ottowa | mtrycrafts
Jun 7, 2002 8:59 PM | | b Why should I use a Canadian facility when they are not even smart enough to live where the weather isn't below freezing 9 months out of the year?
Why work more than 9 month? :) I think they are smart :) |
| re: An open letter to a cable company | eV
Jun 7, 2002 9:40 PM | | You must not have read the post where we discussed this earlier.
If a person could not tell the difference between two known subtle difference colors side by side, then there would be no use in continuing. And yes, a vision test would be required prior to this test. The vision tests I take every year, do include color difference tests. So hopefully there would be no color blind people in this test :).
I just used this analogy to make a point that even with two known subtle difference colors are distinguishable side by side, viewing them separately, could be a guessing game. Naturally, the closer the shades are, the more difficult. Try it sometime. I have. It's harder than you would think it would be. Keep score if you want.
What we are talking about in DBTs for cables is that it is not known if there are differences or not (which I believe is the correct attitude for going into this test). So, since we are speaking POSSIBLE subtle differences, that would make this test even more difficult than known differences.
I do not intend anyone to take what I say as fact. These are just my thoughts, maybe something to consider. Or not.
eV |
| re: An open letter to a cable company | mtrycrafts
Jun 7, 2002 10:14 PM | | b And yes, a vision test would be required prior to this test
Ah, a color test. Tells you if you are color blind or not, or deficient in some colors, not your abaility for JNDs in color.
b So hopefully there would be no color blind people in this test :).
Yes, of course, but that would not telly you their thresholds to differentiate. Only population testing what is the JND.
b I just used this analogy to make a point that even with two known subtle difference colors are distinguishable side by side, viewing them separately, could be a guessing game.
OK. Then, this is a convincing sounter for long term listening for differences as one would not compare side by side but one for a while then the other for a while. That is a very convincing analogy, for the clear thinkers. We agree it needs to be side by side and scan between the two, rapid switching.
b Naturally, the closer the shades are, the more difficult.
Oh, but that is where the subtle differences are, right?
b What we are talking about in DBTs for cables is that it is not known if there are differences or not (which I believe is the correct attitude for going into this test).
Yes it is. but that would work in your color comparison as well, not knowing.
b So, since we are speaking POSSIBLE subtle differences, that would make this test even more difficult than known differences.
Well, that applies in comparing in a color match. You don't really know if there is one untill you test it.
What you described before is good to identify JNDs in color.
b These are just my thoughts, maybe something to consider. Or not.
Yes, it would make sense as discussed above, to me :) |
| by the way | mtrycrafts
Jun 7, 2002 5:54 PM | | If DBT listening cannot identify subtle differences, what do you believe can? Something must be able to do this. |
| Spectrum analysis... | one50
Jun 7, 2002 8:31 PM | | Run some tones and test patterns blah blah...measure the stuff and see what you get.
But on the other hand....
b who cares?
If zip cord works for you so be it...if silver plated solid core hyper-litz works for me so be it...My money...
So many na-sayers so little time.
One50 |
| Spectrum analysis... | mtrycrafts
Jun 7, 2002 9:01 PM | | Oh, that is very easy to do, run the specs and measure. Been done, nothing new:) Too bad human hearing cannot compete with measured data, hearing is not that sensivive and evolution in the industrial age makes it even less. |
| One50--Spectrum analysis... | Abex
Jun 8, 2002 7:10 PM | | You know,that is exactly what I think also!Whatever works for your ears is what really counts.Do not know why people come here to fight over what is right or wrong.For some BOSE are the best there is and the whole shooting match of better products after that is taboo!
Just this past week a guy told me his neighbor had him over and he could not see anything that was different from his 1980's Pioneer reciever.In my learning I have been able to distinguish differences in gear and wires only because I started at a young age being taught then listening for myself! Not by what other's say!If I did not hear a diff then why lie about it to myself.Who would I be fooling!
I think spec's in alot of respects are not the final word in distinguishing diffs either!Amp specs are a good example! |
| One50--Spectrum analysis... | mtrycrafts
Jun 9, 2002 4:14 PM | | b Do not know why people come here to fight over what is right or wrong.
No, just over what may be imagined. After all, if one has a preference, nothing to question.
b I have been able to distinguish differences in gear and wires only because I started at a young age being taught then listening for myself!
How do you account for bias? You do, don't you? Better.
b If I did not hear a diff then why lie about it to myself.Who would I be fooling!
It is not a matter of lying, or not, it is a matter of perceptions and whether they are real or just imagined. Yes, imagination, perceiving what is not there is a very real possibility in audio.
b I think spec's in alot of respects are not the final word in distinguishing diffs either!
Of course not since most of the time the specs are well below audibility, so it is not possible to differentiate betwee some specs.
b Amp specs are a good example!
How so? |
| by the way | eV
Jun 7, 2002 9:54 PM | | I have said before, I am not convinced that there is a test right now that can. Something may be able to, but someone? A lot of variables involved when dealing with humans.
eV |
| by the way | mtrycrafts
Jun 7, 2002 10:16 PM | | b I have said before, I am not convinced that there is a test right now that can.
What would be convincing? I get the impression that not even the CRC or NRC would do for you. Don't think anything would convince you, right? |
| by the way | eV
Jun 8, 2002 10:01 PM | | I have visited those sites.
The NRC from what I can tell is into Loudspeaker testing. I have no doubt that DBTs are a good tool for differences in Speakers.
The CRC is a pretty interesting site. Too bad they just deal with codecs. But they did have some interesting statements on the site.
"many situations where a measure of the perceived difference is required, an experiment with human listeners is impractical. Perceval, a computational model of an average listener, was created to predict task-specific, human responses."
I assume you read the Perceval section. The use of models is employed. This is models of the an average ear. Since ear canals and other physical characteristics vary so much person to person, I would guess there is ongoing improvements on this model.
"For example, listeners vary in their ability to hear sounds below a threshold energy that is frequency-dependent. Also, listeners vary in their ability to hear high frequency sounds. Young listeners are often able to hear frequencies as high as 17-18 kHz, while older listeners often are insensitive to frequencies beyond 8 kHz. Further, some elements of a sound that are physically present in a signal may not be perceived because of masking by other elements adjacent in time and frequency. Such masking also shows considerable variability across listeners."
This is one of the variables that concerns me with listening tests, and the very reason there should be hearing tests prior to a DBT for audible differences.
"These lossy compression algorithms take into account knowledge of human auditory perception, and typically achieve a reduced bit rate by ignoring audio information that is not likely to be heard by most listeners."
Most listeners? This one leaves you kind of hanging.
Anyway, as I said. Interesting sites. I do not think you will find any arguement here on DBTs and speakers. The CRC seems to be a step above anything I have seen. Is there anything on that site that would convince me of cable differences? Not yet. They just deal with codecs. But it does look like they have a world class facility. Maybe in time.
"Don't think anything would convince you, right?"
Wrong.
eV |
| by the way | mtrycrafts
Jun 9, 2002 4:20 PM | | You still didn't present that DBT cannot detec small differences. Perhpas after all these years their models are sufficient. Certainly was not used from day one as the basic data had to be arrived from solid data. Now, it is sufficient to use.
And, if it is acceptable in testing speakers, why do you dismiss it when you compare two components using a speaker? |
| Maybe not. | HTguy
Jun 8, 2002 8:36 PM | | The isn't currently a time machine. Subtle is a gray term. Obviously subtle means larger to you than it does to me. I believe science does not presently have an audio test that does not rely on memory. Like eV was discussing the visual tests, I can hold to shades of blue with subtle differences side by side in the same time frame. Maybe I see the difference, maybe I don't. Depends on how good my eyes are. Your DBT however, relies on one 'viewing' at a time, with time lapse in between. Thereby relying on human memory.
Where you are mistaken, is you belief that 'something must be able to do this'. As of now, time travel is not capable, nor is an audio test that does not rely on human memory. Until you can provide such a test, these 'subtle' differences will neither be proven nor disproven.
The Canadian reference you provide, uses the DBT to test speakers. I don't believe anyone here will argue that speakers make a larger difference than other componants. Hence, you need to prove that your test, which relies on memory, is sensitive enough to show differences between amps, cd players, preamps, etc. The difference between us is you blame the human ear (memory), I fault the test. It's the best that science can do, for now. I have faith that science will continue to evolve, and someday we may have a test that will actually test what we hear, and not what we remember hearing.
Until then, this argument will continue, not unlike a dog chasing his tail. You'll convince the yeasayers that your tool (DBT) is valid the same day we convince the naysayers that their tool (DBT) is invalid.
Around and around we go, do-se-do. |
| Maybe not. | mtrycrafts
Jun 9, 2002 4:44 PM | | b I believe science does not presently have an audio test that does not rely on memory
And your sighted listening doesn't rely on memory? LOL Ut is all about memory.
b The isn't currently a time machine.
So? You think one day perhaps? The evidence today is such that it is not possible. But, if the evidence is such to change ones mind, certainly that will happen.
That is exactely what happened in 1912 with Alfred Wegener and continental drift. The evidence was just not there, then. Not to say that audio is anywhere cutting edge today as was his theory then.
b Like eV was discussing the visual tests, I can hold to shades of blue with subtle differences side by side in the same time frame.
You think it is the same time fram. You scan between the two shades. Time is involved on a smaller scale.
b Your DBT however, relies on one 'viewing' at a time, with time lapse in between. Thereby relying on human memory.
No different from sighted listening, right? Yet, that is the golden rule for the yeasayers, no?
b these 'subtle' differences will neither be proven nor disproven.
You only need to prove that it can be differentiated. There is a sibtle difference as it can be well measured. That is a no brainer.
However, one only has to read all the claims for such an ability to hear subtle differences, right? You are no exception with those claims. So, you must have a test in mind to be able to do it, right? What is it? Or, just speculating on those differences?
And, you have not established that the only way to differentiate subtle differences is when memory is out of the equation. Silly, rally.
b The Canadian reference you provide, uses the DBT to test speakers.
CRC uses it for codec development.
b I don't believe anyone here will argue that speakers make a larger difference than other componants.
You are as confused as eV is. DBT is used to eliminate you biases, nothing more, nothing less. Period. End of story. YOur hearing ability determines how subtle you can hear. DBT does not hinder this. If that is what you believe, you are plain wrong.
b . Hence, you need to prove that your test, which relies on memory, is sensitive enough to show differences between amps, cd players, preamps, etc.
Your sighted listening doesn't rely on memory? Really? What is your proof that you can hear subtle differences using sighted listening? ZEro.
YOu need to get away from that argument. Doesn't hold water.
DBT doesn't hinder anything, it controls for your biases. Incontestable.
b The difference between us is you blame the human ear (memory), I fault the test.
I blame ones limited hearing ability, nothuing more. You cannot accept that fact. Nothing wrong with the test being DBT as it does not hinder anything, it cannot. It limits nothing except your biases.
b and someday we may have a test that will actually test what we hear, and not what we remember hearing.
Oh, please, you are confused. Science does test what you can hear, very well. Be sure to apply this to your claimes based in sighted listening.
b You'll convince the yeasayers that your tool (DBT) is valid the same day we convince the naysayers that their tool (DBT) is invalid.
Sorry, DBT is incontrovertible. You can try in a feeble manner, but it is a fact, especially for small differences as that is where it is needed the most, and used. You just cannot see the forest from the trees. |
| Once again, you missed the point entirely. | HTguy
Jun 9, 2002 7:33 PM | | I made no claim on sighted listening. You claimed DBT is capable on measuring the subtleist differences. It is you that has the burden of proof here. Show me where your DBT tests have positive results on amps with different distortion specs, or cd players with different S/N ratio. The differences obviously exist, you claim that we all cannot hear them because your test says so. I say we can hear them, but the differences are beyond the limit of the test, which does rely on memory. Is a properly set up DBT more sensitive than sighted listening? Yes, I'm not argueing that point, it was not even brought up. However, one more time for the hearing impaired, your DBT only proves that differences DO exist, it CANNOT prove that differences DO NOT exist. If you wish to claim all amps, cd players, preamps, cables sound the same, you have the burden of providing a more sensitive test. I don't claim that everything sounds the same. Prove your test can show the difference between .1% THD and .05% THD. Otherwise, it seems as you put all your faith in a test which hasn't been properly validated to demonstrate subtle differences.
Sorry, DBT is only incontrovertible in your head. You believe what you want to believe. DBT may give you stronger circumstantial evidence, but in no way proves beyond a shadow of a doubt. Is it a helpful tool for identifying differences in speakers? You bet. Does it prove two preamps sound identical? Absolutely not.
What you have is like a scale that weighs in .1 KGs. It is fine when weighing yourself, but if you're trying to tell the difference between a spider and a fly, it won't be such a helpful tool. You can't just be blind enough to say that the spider and the fly are the same, since the scale says so.
Once again we chase our tails, round and round, can you not feel the futility? It is you that cannot see the forest for the trees. It's funny how you believe that it is a fact that your DBT is especially for small differences as that is where it is needed most. LOL. Yes, it's great at measuring the smallest differences within the limitations of the test. Surely the test has limitations, no scientist in his right mind would claim his test has no limitations. Only an absolute fool would believe his test is perfect. Is it better than sighted listening? Yes. Is it perfect? LOL. |
| Once again, you missed the point entirely. | Pat D
Jun 10, 2002 8:25 PM | | You have missed a number of points, and I really don't want to try to cover them all.
DBT is a method which can be employed for auditioning. It is not, per se, a measurement. One can correlate the results of an audition with measurements, and that would be how JNDs are established.
It is possible to measure differences much smaller than the established JNDs. And, this is what a small difference is in this context. If the differences is less than the JNDs, or should be smaller than the JNDs, then it is quite legitimate to ask how people establish that the differences is audible and do determine whether their methods are reliable.
http://users.htdconnect.com/~djcarlst/abx_crit.htm
You make a logical error by conflating DBT as a method with a paricular audition using that method. Obviously, in an audition, nothing can be determined that is beyond the hearing capabilities of the participants. Using DBT methodology cannot improve their hearing. I should also point out that there are other conditions, notably level matching, which are needed to obtain good results.
No one here claimes that all CDPs, preamps, amps, even cables, sound the same. It would be possible to make some of them less accurate than others, and it would be possible to make them so the differences are great enough to be audible. I could only speculate why you and some others attribute claimes to us that we do not make.
As well, one can absolutely prove even positive differences in an audition, as the results are only statistical, not absolutes. |
| Once again, you missed the point entirely. | mtrycrafts
Jun 10, 2002 11:28 PM | | b You claimed DBT is capable on measuring the subtleist differences
No, you cannot comprehend what has been stated.
First, your hearing is the limiting factor, period, regardless how small or large the differences are.
Second, Only a DBT is valid in determining audible differences, especially the ones at the thresholds of hearing, indisputable.
b Show me where your DBT tests have positive results on amps with different distortion specs,
I have posted the citations. You have not bothered to read any of them.
Second. DBT is defacto protocol. It is a bias control protocol, nothing more. Get it through your thick head.
b The differences obviously exist,
What is obvious? The measured ddifferences? Yes, that is. Audible differences? If that is your claim, you have the burden to prove, nothing is obvious to anyone, except a few uninformed ones. Got it? Sighted listening cannot prove anything. Got it?
b you claim that we all cannot hear them because your test says so.
The evidence says it, not me. But, be my guest and present the evidence that I am wrong. Please do so promptly, life is short.
b I say we can hear them, but the differences are beyond the limit of the test,
LOL. Are you serious? Did you read wha you wrote? It is beyond the testing methods but you can hear it. Please tell us how you determined you can hear it? Sighted listening? ROTFLMAO. Get it through your head, sighted listening cannot prove anything, unreliable protocol. Simpel, defacto, incontestable.
b which does rely on memory.
You are trying my patience. Your listening doesn't rely on memory? LOL, give me a break.
b DO exist, it CANNOT prove that differences DO NOT exist. If you wish to claim all amps, cd players, preamps, cables sound the same, you have the burden of providing a more sensitive test.
You are turning into an i...
You have no evidence that they sound different, do you? Your claiming it is futile, not evidence. Your hearing is limited. Components have reached beyond your hearing. Got it? You keep claiming super human hearing, yet offer no evidence.
b I don't claim that everything sounds the same.
No, you claim they all sound different. Prove it. YOU HAVE NO EVIDENCE.
b Prove your test can show the difference between .1% THD and .05% THD.
It cannot, you i.... it is beyond your heaRING LIMIT. Educate yourself. Don't embarrass yourself. You really think you can hear such small distortions? Prove it.
b Otherwise, it seems as you put all your faith in a test which hasn't been properly validated to demonstrate subtle differences.
If the differences are below anyones hearing, nothing will show you can hear it. Got it yet? You are pissing up the wrong tree. Get educated first.Learn something for a change.
b Sorry, DBT is only incontrovertible in your head.
Your ignorance keeps amazing me with every post you make.
b Is it a helpful tool for identifying differences in speakers?
You are an idiot. It helps in all aspects of audio. Nothing else can. Talk to the pros, not the ignorants.
b Does it prove two preamps sound identical? Absolutely not.
There you go making up stories as you go. No audible difference is different. Can you comprehend that? Don't answer, you cannot.
b It's funny how you believe that it is a fact that your DBT is especially for small differences as that is where it is needed most.
No, what is so funny is your ignorance, refuse to find out for your self, listening to uninformed dupes. Talk to the ones in the know. You are at AA, talk to JJ, A world renowned pro. No, you are too far off the deep end to comprehend.
b Is it better than sighted listening? Yes.
Then please tell me how you arrive at audible differences, if not through DBT? Don't answer, you are ill informed to do so. |
| It's good to see that....... | HTguy
Jun 11, 2002 4:03 AM | | ....you find as much humor in my posts as I find in yours. I'm always ROTFLMAO reading your material. You should stop with the name calling though, as it violates board policy, and only serves to make you look more like an a$$. No more chasing the tail for me today son. LOL. |
| re: An open letter to a cable company | sam9
Jun 10, 2002 7:51 AM | | Color shading is probably not the best example. I work for a company involved in products where optical characteristics are important. On serious problem is that visual inspection of the product is not very reliable. Two samples comparred in one circulstance that seem the same can appear different if put side by side on use. On the otherhand two camples which are different when measured spsectrographicly and show significant variations often are indisdinguishable in the intended application.
The only relable approach is instrumentation coupled with "pass"/"no pass" tests limits. The "Pass"/"no pass" tests are application specific and determined by DBT procedures. |
| "determined by DBT procedures." | mtrycrafts
Jun 10, 2002 11:51 PM | | You think they will comprehend this? Believe this? Difficult to talk to the wall, the brain dead. |
| With all due respect... | JoeW
Jun 8, 2002 5:51 PM | | ...I think you underestimate the scale of your proposal.
What's a significant sample size? 100? 1000?
What determines a qualified listener? Any old shmuck who wants to play?
Aside from listening 'ability', how much listener literacy is required? This is a tricky question, but important for unassailable results. How much stock can you assign to someone who cannot recognize a bassoon? A Les Paul? A Fender Rhodes?
The time spent to establish the final and qualified test subjects cannot be ignored.
If you assume the 'best' 10% of listeners, you have to test 10 to get one. Add 'literacy' and that number jumps expodentially.
Lets ignore literacy, for now. Assume your preliminary test takes one hour.
At that rate, to get a sample size of 1000, you have to invest 10,000 test hours, just to arrive at your test subjects. That is pretty close to 5 man years. Thats for one test administrator. Add an assistant, or someone to ensure 'DBT' and that figure doubles. It adds up. If you employ qualified professionals, you're already to an easy 1/2 million. If you insist on literacy with your test subjects, multiply from anywhere from 4 to 10.
Thats what you'd have to do, before you could even begin your tests.
After that, you'd still surely face questions and criticism of your methodology, as (naysayer opinion notwithstanding) its simply not a defined 'science'
Me? I'd love to see it happen. But, I do not believe it would settle a single question. It would simply advance the 'science' of listening tests, which sorely need it. |
| With all due respect... | mtrycrafts
Jun 9, 2002 4:52 PM | | You seem to be able to hear differences, right? How was that confirmed? Did you have proper training? Have a hearing test to establish your JNDs? Thresholds? IN a soundbooth? How about in a room?
Are you a qualified listener? Yet you keep making wild claims, right?
b How much stock can you assign to someone who cannot recognize a bassoon? A Les Paul? A Fender Rhodes?
Oh, yes, you could in a DBT, right?
b If you assume the 'best' 10% of listeners,
Yet you have no problems hearing differences, right? Are you in that 10%? Who says?
b Thats what you'd have to do, before you could even begin your tests.
Nope, just your uninformed opinion..
b After that, you'd still surely face questions and criticism of your methodology,
Read some JOurnals. Cannot get away without it.
b But, I do not believe it would settle a single question. It would simply advance the 'science' of listening tests, which sorely need it.
Yes, you may not like the answers. The science of listening test are advanced already. You are just looking in the wrong direction. |
| With all due respect... | pctower
Jun 10, 2002 12:50 AM | | Hi mytr:
Thought I'd jump in here just to say hello to you and the rest. I tried posting a note to you several days ago, but for some reason it didn't post.
With respect to the issue at hand, I pretty much have to agree with you. My experiencce of cable differences certainly has to rely upon memory if those experiences are real, so I don't think DBTs are invalid in this context simply because they do rely upon memory.
I do think the Leventhal issue must be taken into consideration, and that may require more than 10 or 12 trials, but certainly not 1000 or even 100.
As for training listeners, as I was looking through the H-K site, they seem to have a procedure for training listeners that they believe works.
As for me, I'm taking it easy on the Cape. My son graduated from Havard Business School last Thursday and is getting ready to climb Mt. Kilamenjaro, but as for me my ambition doesn't stretch much beyond deciding where to get the best lobster tonight.
I mentioned in tht earlier post that I saw that kiosk at the mall in Boston that sells Nordost cables. I have to admit, seeing a kiosk in a busy mall trying to sell high end audio cables was pretty silly.
Talk to you all later. |
| With all due respect... | TomN
Jun 10, 2002 6:49 AM | | Hmm out on the cape, I'm asumming you mean Cape Cod or Cape Ann
Are you a wash a shore or a native :)
Tom N. |
| With all due respect... | pctower
Jun 10, 2002 7:28 AM | | Cape Cod and I don't think I'd even garner the respect of a wash a shore, seeing that I'm a desert rat from Phoenix. Talked to home yesterday and I think it was 112 in Phoenix. Gee, this is really tough. |
| With all due respect... | mtrycrafts
Jun 10, 2002 11:42 PM | | Ah, good to have you back.
As you can read,many are confused and try to dismiss DBT listening because it relies on memory. LOL. All listening and comparison relies on memory. DBT is only a bias control protocol, not a measuring tool.
b I do think the Leventhal issue must be taken into consideration, and that may require more than 10 or 12 trials, but certainly not 1000 or even 100.
That is altogether a different issue, isn;t it. Has nothing to do with the validity of using DBT in differentiating small differences, nothing. It has something to do with how well you trust the statistical data, its fairness to one side ot the other, nothing more. No one is preventing anyone from doing more trials. More is better. Never said otherwise.
b As for training listeners, as I was looking through the H-K site, they seem to have a procedure for training listeners that they believe works.
Yes, of course. They need reliable results, professional outcomes, repeatable, reliable. But, by that token then, none of the audiophile claims here would meet that requirements then :)But I won't hold that against anyone:)
b As for me, I'm taking it easy on the Cape.
Great. BE careful of sharks. Anoithe attack today, bit a small boy who thought he was just pushed, 100+ stiches :(
b My son graduated from Havard Business School last Thursday and is getting ready to climb Mt. Kilamenjaro,
WOW. Climbers didn't do so well in the last few weeks in the states. Make sure he doesn't take chances. Congadulations in order for your family:)
b but as for me my ambition doesn't stretch much beyond deciding where to get the best lobster tonight.
That is what I would do too:) |
| With all due respect... | pctower
Jun 13, 2002 2:45 AM | | b As you can read,many are confused and try to dismiss DBT listening because it relies on memory. LOL. All listening and comparison relies on memory. DBT is only a bias control protocol, not a measuring tool.
I must confess, I don't understand why so many yeasayers keep fighting you on this. Eliminating the possibility of unwanted bias in any decision seems to me a goal that is beyond argument. Oh well.
As for the number of trials, I would think for my own personal purposes if I couldn't correctly identify 9 out of 10 times, then the difference isn't significant enough to warrent further personal investigation. I don't think I would have the patience for much more than 10 trials.
When I mention Leventhal I'm simply saying that in true scientific testing, his observations should be taken into consideration.
I know what you mean about this climbing bit. I'm not particularly happy about my son's lack of judgment, but I do have to respect his willingness to try just about anything. |
| With all due respect... | mtrycrafts
Jun 14, 2002 12:16 AM | | b I must confess, I don't understand why so many yeasayers keep fighting you on this.
Confession is good for the soul :)
Perhaps, once they accept this, there is nothing further but to have data under such conditions that may be embarrassing :)
b . I don't think I would have the patience for much more than 10 trials.
That is why you don't see more than 15-20 trials in published papers:)
b When I mention Leventhal I'm simply saying that in true scientific testing, his observations should be taken into consideration.
OK. You finally convinced me here:) Although, for such an undertaking, I would not have disagreed with the amount of trials over 10 :) 100 would have been fine with me:) as I would not have been one of the participants :)
b but I do have to respect his willingness to try just about anything.
Yes, as difficult as it may be for such an undertaking. |
| by the way | mtrycrafts
Jun 10, 2002 11:44 PM | | b but certainly not 1000 or even 100
How many trials would you like to see? |
| Sample Size | Warren Warren
Jun 13, 2002 6:01 PM | | I believe that a sample of 20 individuals with 5 tests each is more than adequate, assuming that they are all individuals who have had their hearing tested withing the last year and were proven to have average to above average hearing. After all, the test is only a yes/no result. Obviously the test is not designed for those with below average hearing, and ideally, we would raise the bar to include only those whose hearing tested in the top 10% of the population, but this would not only be cost-prohibitive, but also be time intensive. |
| Sample Size | mtrycrafts
Jun 14, 2002 12:08 AM | | b with 5 tests each is more than adequate,
Don't even think this. Automatic disqualification, insufficient number of trials by each person. You need 5 correct response.
10 trials is the absolute minimum with 15 or 20 preferable.
This is what pc was so upset and the Leventhal paper writes abou8t, insufficient trials by a subject, not the number of subjects |
| The Devil is in the details <nt> | JoeW
Jun 16, 2002 10:06 PM | | |
| With all due respect... | FLZapped
Jun 12, 2002 7:20 AM | | "What's a significant sample size? 100? 1000?"
I would say 100 is a good sample size.
"How much stock can you assign to someone who cannot recognize a bassoon? A Les Paul? A Fender Rhodes?"
This is inmaterial. You're looking for a listener to detect a difference when two devices are switched back and forth. Either they hear it, or they don't, and mark it on their score card.
"At that rate, to get a sample size of 1000, you have to invest 10,000 test hours, just to arrive at your test subjects. That is pretty close to 5 man years. Thats for one test administrator. Add an assistant, or someone to ensure 'DBT' and that figure doubles. It adds up. If you employ qualified professionals, you're already to an easy 1/2 million. If you insist on literacy with your test subjects, multiply from anywhere from 4 to 10."
Your numbers are way off. The majority of your time will be spent on collecting the equipment and finding a facility and then getting it all set-up correctly.
The next largest challenge might be finding enough volunteers and doing a basic screening.
Each test will only take about 5 minutes, add 5 minutes to swap people around and you can do 6 sets in each hour. The room should be able to seat 10 listeners at time, which means in an hour, you will have 60 samples. So figure about 2 hours per comparision. For 4 products, you have 6 sets to run, so you're looking at 12 hours.(If your room seats 20, you can cut the time in half.)
"After that, you'd still surely face questions and criticism of your methodology, as (naysayer opinion notwithstanding) its simply not a defined 'science'"
Agreed, that's called peer review.
"Me? I'd love to see it happen. But, I do not believe it would settle a single question. "
Likewise, but I disagree with your second statement, otherwise there is no point in trying.
-Bruce |
| With all due respect... | JoeW
Jun 13, 2002 10:54 PM | | >"How much stock can you assign to someone who cannot recognize a bassoon? A Les Paul? A Fender Rhodes?"
>This is inmaterial. You're looking for a listener to detect a difference when two devices are switched back and forth. Either they hear it, or they don't, and mark it on their score card.
I would agree with you, perhaps, on a significantly larger sample size. But, I cant take this as a valid scientific endeavor with a sample size of 1000. 100? Well, it might have a value when someone saddly discovers they are out of toilet paper.
A 5 minute screening? This is what makes DBT worthy of ridicule. DBT may have a value is the future, when their advocates stop demanding tests skewed to achieve a null. |
| With all due respect... | mtrycrafts
Jun 14, 2002 12:19 AM | | b DBT may have a value is the future, when their advocates stop demanding tests skewed to achieve a null.
DBT is nothing but a bias control protocol. Keep that in mind and try to understand that. Being DBT has nothing to do with null results. That may be the outcome of a test, or not.
Be my guest undertaking one of your approved rigor. |
| You have already disqualified yourself<nt> | JoeW
Jun 16, 2002 9:57 PM | | |
| What could possibly | audioengr
Jun 11, 2002 11:30 AM | | be in it for me?
I am confident that a fair, impartial and well-designed experiment would result in my favor.
However, I would be out thousands of dollars in product for an extended period of time. And I am also not confident that you could pull-off such an experiment. I dont know the quality of your equipment. Even if the result was favorable towards my cables, what benefit would I gain from this? An article in Stereophile?, a paper for AES - NOT LIKELY. Even if I sold a few cables as a result, I believe that it would still not be worth the hassle. And finally, after all of the flack that I have taken on this forum from you and the other naysayers, I'm not very motivated to cooperate with the likes of you.
My money is better spent on sending review samples to real reviewers. |
| I don't really think WW considers himself a naysayer. | markw
Jun 11, 2002 6:58 PM | | Are you as clear on all your facts as you are on this one? |
| What could possibly | mtrycrafts
Jun 11, 2002 10:47 PM | | b I am confident that a fair, impartial and well-designed experiment would result in my favor.
Of course you would be. If not, you are cheating your costumers today. However, there is zero evidence to conclude that your wire would prevail.
b However, I would be out thousands of dollars in product for an extended period of time.
How so? Is that what your cables cost for a couple of pairs? What a rip off.
b I dont know the quality of your equipment.
Why don't you identify the necessary equipment that would pass muster by you? After all, do all your customers who report success with your cables have the same quality of system as you do? LOL. If not, their claims are bogus before getting out of the gate. LOL. How do you know your customers can hear?
b I'm not very motivated to cooperate with the likes of you.
bla, bla. Don't worry, your cables would not prevail.
b My money is better spent on sending review samples to real reviewers.
LOL What a joke. Of course they are not biased and control for it in their reviews, right? LOL. |
| What could possibly | FLZapped
Jun 12, 2002 7:38 AM | | "I am confident that a fair, impartial and well-designed experiment would result in my favor."
" And finally, after all of the flack that I have taken on this forum from you and the other naysayers, I'm not very motivated to cooperate with the likes of you."
Well, aren't these two intesting dipolar statements.
"What could possibly be in it for me?"
Ahhh, the bottom line. If you indeed are correct in your assertions, then your employer would have bragging rights - they could corner the market with all the ads that would proclaim their product as king of the hill.....
Of course, if they don't, then yes, they'd be in deep doo-doo....and isn't that what the real fear is?
That's what the whole industry is afraid of, that in an impartial test, their products wouldn't shine as supreme over generics.
"However, I would be out thousands of dollars in product for an extended period of time."
Of course, with funding samples could be purchased, making the products put out by your employer a participant anyway.
-Bruce |
| after all of the flack that I have taken on this forum from you | Warren Warren
Jun 13, 2002 5:51 PM | | Have I ever given you or your company flack? I don't think so. I actually am in favor of buying good quality interconnects and speaker wire. I am against "audiophile" power cords, but even then, I haven't really given anybody flack about them. And, YES, selling a few cables should be the final result you are looking for. After all, when it is proven true that there IS a difference between interconnects and cables, you would have both legs to stand on, not just the marketing and hype. Also, how would you be out "thousands of dollars in product" for one set of interconnects and one pair of speaker cables. Unless, of course, your products are EXTREMELY overpriced. If you don't want to cooperate with me, that's too bad, as I really had hoped to have a positive response from SOMEBODY out there. Maybe the naysayers are right with their arguments. I guess we're not going to find out with the likes of you working at cable companies. |
| Bad WW! | JoeW
Jun 17, 2002 1:36 AM | | You're behaving like the monkeyboy. How do you hope to propel a discusion, when you insist all agree with your final analysis?
As Diy'r, I'm not a patron of audioengr's.
Still, he brings a lot more to the table than you do.
Sorry dewd, your passion for Star Wars does not establish a level of expertise for anything beyond Star Wars.
If you had any respect for hard work, you'd respect the people who did it |
| re: An open letter to a cable company | Jameson Davenport Briggs VII
Jun 12, 2002 7:22 AM | | We have done such tests here at the lab in the compound and have found enormous differences between cables. Indeed such tests are ongoing. Presently we are working with the Nordost Valhalla. A great cable, but sensitive to improper breaking in. our results are double blnd with expert audiophiles . We use the Wireworld Direct/Comparator. the results are definitive. Let's move on.
Further, we have found large differences between the SAME cables that have been properly aged and broken in than those which have not.
Jameson Davenport Briggs, Northern Maine. |
| Can I hitch a ride? | MonstrousMike
Jun 12, 2002 9:48 AM | | What star cluster in Northern Maine in? Are there any M-class planets? I'm getting tired of this solar system and could use a fresh start. |
| |
