| Home |
| Search |
| Today's Posts |
|
|
|
#2
|
|||
|
|||
magic in a box
wrote in message ups.com... wrote: "Robert Morein" wrote in message ... wrote in message oups.com... Robert Morein wrote: "paul packer" wrote in message ... On Tue, 25 Oct 2005 08:24:32 -0400, "Arny Krueger" wrote: You only ABX equipment where there is a reasonable controversy as to whether they sound different. But according to Arny and his crowd all the "properly designed" components are the same electronically: cables, preamps, amps, cdplayers, dacs so all of them must sound the same,when ABX tested. This would be an highly advanced concept for you Ludovic, its what's known as a truism. Nope. A truism is a self-evident truth. Which that isn't. All well-designed cables, preamps, amps, cd players, dacs sound the same because they have the mission of sounding the same. A mission? Hmmm, sounds important. So you're saying that in electronics, whatever is intended ultimately comes to be. All cables, preamps, amps, cd players, dacs that sound the same provide evidence that they are well-designed. If they don't sound the same, then they have failed their mission, and its fair to say that they are not well-designed. You mean sounds the same to you or to me? However, just sounding the same is not the only criteria for something being well-designed. The device must also meet the other conventional economic, usability and durability criteria. Correct. And those factors are not sufficiently taken into account. Loudspeakers (and phono cartridges?) "we all know" sound different. Phono cartridges and speakers should also sound the same. In the case of phono cartridges some sound the same, some sound different. Care to list which is which. :-) http://www.pcavtech.com/abx/abx_phca.htm So no need for ABX. As the sainted leader says it is a "poor choice"- no room for doubt.. Just observing what should be obvious. In other words the only components worthy of a listening test are those that "we know" will sound the same. Wrong again. There has to be a reasonble controversy. And then ABX settles the controversy by "proving" there's no difference. I wish you would list the occasions when ABX has demonstrated the difference between "non-faulty" and similarly measuring amplifiers. It shouldn't take long. I think a succinct description of the ABXer's position is: All devices which are known, by prejudice or folk belief, to sound the same, sound the same when tested by ABX. All devices which are known, by prejudice or folk belief, to sound different, sound different when tested by ABX. Thus, the ABX box is a mechanical device that contains prejudice and folk belief! No disagreement.Just a clarification; You say: " All devices which are known, by prejudice or folk belief, to sound different, sound different when tested by ABX." There's no experimental evidence they would sound different under ABXing.. Sean Olive's double=blind loudspeaker test demonstrated that some 80% of his panelists could not distinguish one loudspeaker from the other when asked to differentiate but the same people had no difficulty choosing the better ones when asked to express preference. And this was just double-blind. One can only imagine what would happen were it a full, undiluted ABX. The real reason why Arnie shouts: "No need. The difference is obvious" is his fear that under ABX it even the "obvious" would vanish into thin air. Ludovic Mirabel It appears the ABX mojo is a bargain basement spell. That must account for why Harman and B&O make such crappy equipment, since they both use ABX, among other blind protocols. __________________________________________________ ___ That must account for why Harman and B&O make such crappy equipment, since they both use ABX, among other blind protocols Try harder; there must be someone else in this world who according to your confidential sources is using ABX. I've already posted work done by the BBC and Nokia. Bang&Olafson also use them. Could you quote the reference to where they published the results of that "use"? You know of course that without such a checkable reference you're only gossiping- good enough for RAO I suppose but unworthy of a scientist like you. I'm not a scientist, nor do I claim to be. Why not check the JAES for their articles or write to Sean personally? And of course it would be about comparing components because that is what a lay audio forum is interested in. It might be or might not be since as has been mentioned some work is done on circuit changes to existing equipment to see if made any difference. And of course there would be a few postive results; the panels had no problem distinguishing the components from each other. Ludovic Mirabel Do your onw research. Write to Sean and express your concerns. I'm sure he can give you better leads than I can. Unless of course, you are as abrasive with him and as loose with the truth as you've been here. |
|
#3
|
|||
|
|||
|
magic in a box
NYOB says: That must account for why Harman and B&O make such crappy equipment, since they both use ABX, among other blind protocols Try harder; there must be someone else in this world who according to your confidential sources is using ABX. I've already posted work done by the BBC and Nokia. Bang&Olafson also use them. Could you quote the reference to where they published the results of that "use"? You know of course that without such a checkable reference you're only gossiping- good enough for RAO I suppose but unworthy of a scientist like you. I'm not a scientist, nor do I claim to be. Why not check the JAES for their articles or write to Sean personally? And of course it would be about comparing components because that is what a lay audio forum is interested in. It might be or might not be since as has been mentioned some work is done on circuit changes to existing equipment to see if made any difference. And of course there would be a few postive results; the panels had no problem distinguishing the components from each other. Ludovic Mirabel Do your onw research. Write to Sean and express your concerns. I'm sure he can give you better leads than I can. Unless of course, you are as abrasive with him and as loose with the truth as you've been here. __________________________________________________ ___ The subject under discussion is: Quote ONE SINGLE published (any mag. or Journal) report of the use of ABX to comparing components for their music reproduction properties with a POSITIVE outcome ie the majority of panelists differentiated with statistical validity. ATTENTION: We're talking about ABX specifically NOT just DBT. NYOB says: "I've already posted work done by the BBC and Nokia. Bang&Olafson also use them" Which "them".? DBT is not ABX. ABX is one particular derivative of DBTs. Re BBC 1) You never, never, never posted any "work done by the BBC" mentioning ABX. I hate to call you an unscrupulous, inveterate liar.. All you have to do to nail me is to REPOST 2) you sent me on a wild goose chase in a 10 year catalogue of hundreds of articles about anything under the sun published by the BBC 3) none of them mentioning ABX. Re Nokia (A Finnish cellphone company) Quote a precise reference to ABX being used in audio component comparison by Nokia Re B&O: The only site I wasted time looking for and found discusses propagation of sound. No mention of DBT: no component comparison I asked: Could you quote the reference to where they published the results of that "use"? You know of course that without such a checkable reference you're only gossiping- good enough for RAO I suppose but unworthy of a scientist like you. NYOB answers: "I'm not a scientist, nor do I claim to be. Why not check the JAES for their articles or write to Sean personally?" I said: And of course it would be about comparing components because that is what a lay audio forum is interested in. His "answer": "It might be or might not be since as has been mentioned some work is done on circuit changes to existing equipment to see if made any difference." Yes and Newton investigated sound waves too. But this is about the ABX use in ... etc. etc. I said: And of course there would be a few postive results; the panels had no problem distinguishing the components from each other. Ludovic Mirabel NYOB:: "Do your onw research. Write to Sean and express your concerns. I'm sure he can give you better leads than I can. Unless of course, you are as abrasive with him and as loose with the truth as you've been here". Words fail me: You persist in showing an incredible combination of bloody cheek with mind-shaking stupidity. Or cold you believe that everyone else is an idiot? YOU say YOU have a method for showing differences between speakers. Have you ever heard what the word *evidence* means?. Of course you have no evidence, .because none exists.. But you have the audacity to send me to find your imaginary, evidence for you. And an incredible lack of courtesy to a busy man directing people to "Sean" ( we ARE familiar aren't we?) to ask him to search for it. What's more this is the second time you're doing this. NYOB - others lost patience with you before me - and no wonder. I said it before but this time I'll stick to it. Till you have something sensible- like real evidence- to say I'll just repost this as an answer to this kind of obnoxious silliness Ludovic Mirabel __________________________________________________ _______________________ |
|
#4
|
|||
|
|||
|
magic in a box
wrote in message oups.com... NYOB says: That must account for why Harman and B&O make such crappy equipment, since they both use ABX, among other blind protocols Try harder; there must be someone else in this world who according to your confidential sources is using ABX. I've already posted work done by the BBC and Nokia. Bang&Olafson also use them. Could you quote the reference to where they published the results of that "use"? You know of course that without such a checkable reference you're only gossiping- good enough for RAO I suppose but unworthy of a scientist like you. I'm not a scientist, nor do I claim to be. Why not check the JAES for their articles or write to Sean personally? And of course it would be about comparing components because that is what a lay audio forum is interested in. It might be or might not be since as has been mentioned some work is done on circuit changes to existing equipment to see if made any difference. And of course there would be a few postive results; the panels had no problem distinguishing the components from each other. Ludovic Mirabel Do your onw research. Write to Sean and express your concerns. I'm sure he can give you better leads than I can. Unless of course, you are as abrasive with him and as loose with the truth as you've been here. __________________________________________________ ___ The subject under discussion is: Quote ONE SINGLE published (any mag. or Journal) report of the use of ABX to comparing components for their music reproduction properties with a POSITIVE outcome ie the majority of panelists differentiated with statistical validity. ATTENTION: We're talking about ABX specifically NOT just DBT. NYOB says: "I've already posted work done by the BBC and Nokia. Bang&Olafson also use them" Which "them".? DBT is not ABX. ABX is one particular derivative of DBTs. Re BBC 1) You never, never, never posted any "work done by the BBC" mentioning ABX. I hate to call you an unscrupulous, inveterate liar.. All you have to do to nail me is to REPOST 2) you sent me on a wild goose chase in a 10 year catalogue of hundreds of articles about anything under the sun published by the BBC 3) none of them mentioning ABX. Re Nokia (A Finnish cellphone company) Quote a precise reference to ABX being used in audio component comparison by Nokia Re B&O: The only site I wasted time looking for and found discusses propagation of sound. No mention of DBT: no component comparison I asked: Could you quote the reference to where they published the results of that "use"? You know of course that without such a checkable reference you're only gossiping- good enough for RAO I suppose but unworthy of a scientist like you. NYOB answers: "I'm not a scientist, nor do I claim to be. Why not check the JAES for their articles or write to Sean personally?" I said: And of course it would be about comparing components because that is what a lay audio forum is interested in. His "answer": "It might be or might not be since as has been mentioned some work is done on circuit changes to existing equipment to see if made any difference." Yes and Newton investigated sound waves too. But this is about the ABX use in ... etc. etc. I said: And of course there would be a few postive results; the panels had no problem distinguishing the components from each other. Ludovic Mirabel NYOB:: "Do your onw research. Write to Sean and express your concerns. I'm sure he can give you better leads than I can. Unless of course, you are as abrasive with him and as loose with the truth as you've been here". Words fail me: You persist in showing an incredible combination of bloody cheek with mind-shaking stupidity. Or cold you believe that everyone else is an idiot? YOU say YOU have a method for showing differences between speakers. Have you ever heard what the word *evidence* means?. Of course you have no evidence, .because none exists.. But you have the audacity to send me to find your imaginary, evidence for you. And an incredible lack of courtesy to a busy man directing people to "Sean" ( we ARE familiar aren't we?) to ask him to search for it. What's more this is the second time you're doing this. NYOB - others lost patience with you before me - and no wonder. I said it before but this time I'll stick to it. Till you have something sensible- like real evidence- to say I'll just repost this as an answer to this kind of obnoxious silliness Ludovic Mirabel Since evidence is meaningless to you, you post any ****ing thing you want. If you really cared to find any instead of making specious claims about how o such results exist, I might give a ****. You can lead a man to knowldege, but you can't make him think. __________________________________________________ _______________________ |
|
#5
|
|||
|
|||
|
magic in a box
wrote in message oups.com... NYOB says: That must account for why Harman and B&O make such crappy equipment, since they both use ABX, among other blind protocols Try harder; there must be someone else in this world who according to your confidential sources is using ABX. I've already posted work done by the BBC and Nokia. Bang&Olafson also use them. Could you quote the reference to where they published the results of that "use"? You know of course that without such a checkable reference you're only gossiping- good enough for RAO I suppose but unworthy of a scientist like you. Has it dawned on you that perhaps they don't publish the results of their prodcut research? I suggested before that you write to Sean Olive personally and if you ask him nicely, he may just be willing to help you on your quest. |
|
#6
|
|||
|
|||
|
magic in a box
wrote in message ups.com... And of course there would be a few postive results; the panels had no problem distinguishing the components from each other. Ludovic Mirabel Just remembered; you tried it all once before but it turned out.. you know how. And I'm sure you would not want me to remind you in detail. ABX is typically used when the research question is whether or not listeners can reliably discriminate between two audio products. Typical uses are listening tests on amplifiers, CD-DVD players, cables or high quality audio CODECS (high bit rate AAC, AC-3,etc) I frankly don't do many of these types of comparisons so I seldom use ABX.. instead I use multiple comparisons and measure preference and other sound quality attributes (timbre, spatial, distortion) I have used ABX for testing power amplifiers, where the measured objective and subjective differences were very small. I often include a preference measure with the ABX test in the event that if listeners can reliably identify X, I would also get a measure of which amplifier listeners preferred, and the magnitude of the preference. I have never published any experiments where I used ABX, since you often get a null result which is not very interesting to publish unless you like ****ing off audiophiles who spent $1,000 on their 1 foot interconnects. Since most of my testing and research focuses on loudspeakers and room interactions, the audible differences are sufficiently large that I do not require the use of ABX. My colleague Todd Welti [2] has used a similar triangle protocol (ABC ) to judge audibility of mono versus stereo subwoofers. The use of ABX has been recently reported in AES preprints [1] for tests of SACD (DSD) versus DVD-Audio (PCM )where they found no reliable audible differences in recordings encoded in DSD versus PCM. ABX was also used by Bill Martens and colleagues [3] at McGill to investigate audibility of MONO VERSUS STEREO bass and effects of subwoofer position and lowpass cut-off. ABX is not the only method for determining reliability of audible differences, and certainly not the most efficient. Since it is a 2-AFC method (ie there are 2 possible choices) there is 50% chance of guessing correctly . That means you need to run a reasonability large number of trials to establish a reliable audible difference at a 0.05 significant level. The Triangle Method (ABC) requires listeners to indicate which of the three sounds (ABC) is different. Here there is only a 1/3 chance of guessing correctly, meaning fewer trials are required to establish the same chance probability. This is why Todd Welti chose that method. There could be an argument that 2-AFC is an easier task for the listener than a 3-AFC but there is no real evidence to support it. Another 2-AFC method is the ABC (with hidden reference) that is a ITU-R standard (BS-R 1116.1) that was used for testing high quality codecs. Here A is identified as the reference (uncompressed audio) and B or C is the hidden reference (same as A) and the other is the compressed audio. The method requires listeners to rate B and C using the 5-point impairment scale. There are tons of papers in the AES that describe CODEC test results using this method. For lower quality CODECS (or anything where the audible differences are not in question) there is a new method ITU_R BS1534 (also known as MUSHRA or Multiple Stimulus with Hidden Reference and Anchors) that I use and recommend. It allows listeners compare multiple products at once which has been shown to produce more reliable and discriminating results. For more info I recommend this listening test Tutorial that the AES Technical Committee on Perception and Subjective Evaluation of Audio Signals gave at two recent AES Conventions. The presentations from that tutorial can be downloaded here http://www.aes.org/technical/documen....cfm?docID=168 Cheers, Sean Olive, Manager Subjective Evaluation R&D Group, Harman International 8500 Balboa Blvd. Northridge, CA, 91329 |
Hybrid Mode
