ehtcom Posted November 3, 2014 Share Posted November 3, 2014 Makes sense to me. I not sure what you can't get. Blind test, A and B sound different. Person prefers B, His system now sounds better to him. Personally, I thought A sounded better. 2 Link to comment Share on other sites More sharing options...
davewantsmoore Posted November 3, 2014 Share Posted November 3, 2014 Makes sense to me. I not sure what you can't get. +1 Personally, I thought A sounded better. Perhaps you are right .... perhaps an ABX will narrow down that person B isn't that clear on his preference. Of course, often it just doesn't matter.... horses for courses, and all that. Controlled testing isn't designed to "tell" people what they should like (in many ways quite the opposite). Link to comment Share on other sites More sharing options...
ehtcom Posted November 3, 2014 Share Posted November 3, 2014 Hi Dave. We had a little blind test at Bathurst GTG a few years ago. Using a DEQX system, we switched between corrected group delay, and non corrected. 6 ppl, 3 prefered non, 3 prefered corrected. 1 Link to comment Share on other sites More sharing options...
Mining Man Posted November 3, 2014 Share Posted November 3, 2014 ... I have seen pics and plans of tassie devils room, and yeah wow gone to some lengths with the room design and setup with the hexagonal room ? ... It's pentagon shaped, with a vaulted ceiling. Thus, there are no parallel surfaces in the room. Pretty amazing, really. B) Link to comment Share on other sites More sharing options...
Guest myrantz Posted November 3, 2014 Share Posted November 3, 2014 If that's ALL we have to go with.... then you cannot. You have only showed "difference". You would need to draw on some other results, or redefine your experiment so it could say something more about preference. So you have an experiment to confirm there is a difference, and now added another undefined experiment to confirm a preference... Can we get to this now please? "It certainly can yield a better system". On the other hand, if your experiment was constructed to determine if X amount of distortion was discernible .... then you might be able to see something particular about distortion in your results - helping you determine what level of distortion might matter (ie. it might help you build a better audio system) Did johnmath did this to his customers? And I love the way you change from certainly to might. Although I still can't get the actual definition of "better" from him, based on what is given to me so far, I can work out the probability of that might, to the nearest percent - 0%.. Link to comment Share on other sites More sharing options...
Satanica Posted November 3, 2014 Share Posted November 3, 2014 Hi Dave. We had a little blind test at Bathurst GTG a few years ago. Using a DEQX system, we switched between corrected group delay, and non corrected. 6 ppl, 3 prefered non, 3 prefered corrected. Obviously the difference between the two wasn't big enough to detect, thus the 50/50 Link to comment Share on other sites More sharing options...
Addicted to music Posted November 3, 2014 Share Posted November 3, 2014 Reminds me of digital systems, though you forgot nand and nor...I have some old books somewhere only if you dont understand that words "preference" and difference" are differnet words with different meaning It's called Boolean Algebra.... It was taught in Digital Electronics 1 1 Link to comment Share on other sites More sharing options...
ehtcom Posted November 3, 2014 Share Posted November 3, 2014 Obviously the difference between the two wasn't big enough to detect, thus the 50/50 No, we could all pick the difference. 3 prefered one setup, and 3 prefered the other. 1 Link to comment Share on other sites More sharing options...
Volunteer sir sanders zingmore Posted November 3, 2014 Volunteer Share Posted November 3, 2014 Until you can showed me otherwise, yes, it's magic. Or a better word - placebo. Let me give you an example as to why your method is ineffective.. Let's say you have own item A and is thinking about upgrading it to B... You put both in a ABX test, and you have identified they're different. During this evaluation you do not have prior knowledge which is which.. So that allegedly takes out all the haphazard conditions you mentioned.. So 'objectively', sight unseen, you have determined A is indeed different from B. You then decided on B (note that when you made the decision you have no idea it's B). Happy the method "works", you sell A to buy B. Years later, with more money in the bank, you decided B is not good enough, and you repeat the test with latest and greatest C. After the test you decided on C (and again you don't know it's C at the time) is better. Very happy you sell B and upgrade to C. Seasons pass, and after you completely lost your hair you decided again to upgrade. So now you compare C with D. Same thing happens and you pick D... But when the cover is removed, what you thought is D is in fact the old A you gave up so long ago. You may well argue the above scenario will never happen.. Yeah right.. :D Certainly leads to a better system? Myth busted Edit: Fixed up for clarity.. Heaps of bad grammer but hopefully you get the idea.. As an alternative to your strawman, I present..... A multiplicity of DOTM threads Link to comment Share on other sites More sharing options...
Guest myrantz Posted November 3, 2014 Share Posted November 3, 2014 As an alternative to your strawman, I present..... A multiplicity of DOTM threads not entirely sure what dotm means. as for strawman, you reckon so? Link to comment Share on other sites More sharing options...
Satanica Posted November 3, 2014 Share Posted November 3, 2014 No, we could all pick the difference. Just so I have this clear, each of you (the 6) did all prove that? Link to comment Share on other sites More sharing options...
Guest Posted November 3, 2014 Share Posted November 3, 2014 (edited) I had a 6-channel comparator made by Niles Audio at the hi fi shop that allowed two home theatre AVRs to be compared. It is easy of course to send the same SPDIF signal to two units and fairly trivial to set them up the same way at the same levels. It wasn't a truly blind comparison, because someone in the room had to operate the switch, but at least that person was behind the subject doing the evaluation. There are quite substantial differences between AVRs and their ability to drive speakers at normal home levels. The differences were pretty obvious and I don't think I ever had a customer that didn't have an opinion that one brand was "better" or "more correct" than the other where two brands were compared side by side. Clearly lower distortion, less harshness, better dynamics, deeper bass extension are all easily observed. Without the comparator, by the time all of the speakers are switched over, it is hard to hold the picture of what one component sounded like, and you can't just flick back. Manufacturers and suppliers didn't like this evaluation setup much. There were $1000 AVRs that sounded better than $10,000 ones. Edited November 4, 2014 by Guest Link to comment Share on other sites More sharing options...
ehtcom Posted November 3, 2014 Share Posted November 3, 2014 Just so I have this clear, each of you (the 6) did all prove that? To the best of my knowledge, yes. Basically, I sat in the listening position, music was played, the different setups were digitally switched seemless during the music/songs. I indicated with my hand when I could hear a change. Once I could reliably pick the changes, I then indicated a preference. I had no prior knowledge as to what changes were happening, just sit and listen, tell us if you can hear any changes. 1 Link to comment Share on other sites More sharing options...
Volunteer sir sanders zingmore Posted November 3, 2014 Volunteer Share Posted November 3, 2014 (edited) not entirely sure what dotm means. as for strawman, you reckon so? Textbook example Edited November 3, 2014 by Sir Sanders Zingmore Link to comment Share on other sites More sharing options...
davewantsmoore Posted November 3, 2014 Share Posted November 3, 2014 (edited) So you have an experiment to confirm there is a difference, and now added another undefined experiment to confirm a preference... Can we get to this now please? "It certainly can yield a better system". Yes, that's right .... other results. There are plenty of them that we know about - and you're allowed to use them. We know that people like "low distortion" (this is a known preference) ..... we can now conduct an experiment to see what type of distortion, or how much distortion, is audible in our audio system. Don't you think that knowing how much distortion is audible is going to help you make a better (improved) audio system? And I love the way you change from certainly to might I use "might" for a VERY specific reason. Because we may do the experiment, and the results may NOT actually tell you anything which can lead to a solid conclusion. "certainly can" .... means "it is certainly possible" ..... you know.... if you do some controlled tests then it is possible that you might discover something which will allow you to make a better audio system. Then again, you might find nothing of interest.... depends on what you are testing. Although I still can't get the actual definition of "better" from him, based on what is given to me so far, I can work out the probability of that might, to the nearest percent - 0%.. Is it really this hard? What on earth do you think better means? You find out what people like .... you find out how much of it you need (or need to avoid) .... You now know things which will let you build a better audio system. Controlled testing certainly can help to make a better audio system Bacon can certainly be delicious Asking people questions can certainly lead to learning new things Edited November 3, 2014 by davewantsmoore Link to comment Share on other sites More sharing options...
Guest myrantz Posted November 3, 2014 Share Posted November 3, 2014 Manufactureres and suppliers didn't like this evaluation setup much. There were $1000 AVRs that sounded better than $10,000 ones. I used my haphazard method and ended up with the same results too But of coz my better is likely different from your better - mine is $9000 better ... Link to comment Share on other sites More sharing options...
Volunteer sir sanders zingmore Posted November 3, 2014 Volunteer Share Posted November 3, 2014 Without wading through the whole thread, can someone explain to me what is better about evaluating components when you *do* know which is which? Link to comment Share on other sites More sharing options...
Tony M Posted November 3, 2014 Share Posted November 3, 2014 Without wading through the whole thread, can someone explain to me what is better about evaluating components when you *do* know which is which? It can avoid a lot of embarrassment. Link to comment Share on other sites More sharing options...
Guest myrantz Posted November 3, 2014 Share Posted November 3, 2014 Without wading through the whole thread, can someone explain to me what is better about evaluating components when you *do* know which is which? I can try and explain, but I doubt I can do it without rambling.. What you can do instead is rely on this so called science. i.e. perform a ABX or blind test with A = evaluating components when you do know; and B = evaluating components when you don't. Once you have confirm a difference.. Do that unknown test Dave mentioned to know preference. Then somehow by magic you'd know what is better ... Link to comment Share on other sites More sharing options...
Satanica Posted November 3, 2014 Share Posted November 3, 2014 To the best of my knowledge, yes. Basically, I sat in the listening position, music was played, the different setups were digitally switched seemless during the music/songs. I indicated with my hand when I could hear a change. Once I could reliably pick the changes, I then indicated a preference. I had no prior knowledge as to what changes were happening, just sit and listen, tell us if you can hear any changes. No, no, no, I'm asking if you and all the others could reliably demonstrate that you could perceive a difference before your expermient of preference. Sorry, but it really sounds like you and all the others did not do this first, it wasn't actually even attempted. And if you did not (and I don't think you did) then your "preference" experiment yielded the same result as tossing a coin. Link to comment Share on other sites More sharing options...
Volunteer sir sanders zingmore Posted November 3, 2014 Volunteer Share Posted November 3, 2014 I can try and explain, but I doubt I can do it without rambling.. What you can do instead is rely on this so called science. i.e. perform a ABX or blind test with A = evaluating components when you do know; and B = evaluating components when you don't. Once you have confirm a difference.. Do that unknown test Dave mentioned to know preference. Then somehow by magic you'd know what is better ... I can't really understand what you are saying but I can't see anything you are saying that answers my question Link to comment Share on other sites More sharing options...
ehtcom Posted November 3, 2014 Share Posted November 3, 2014 To the best of my knowledge, yes. Basically, I sat in the listening position, music was played, the different setups were digitally switched seemless during the music/songs. I indicated with my hand when I could hear a change. Once I could reliably pick the changes, I then indicated a preference. I had no prior knowledge as to what changes were happening, just sit and listen, tell us if you can hear any changes. No, no, no, I'm asking if you and all the others could reliably demonstrate that you could perceive a difference before your expermient of preference. Sorry, but it really sounds like you and all the others did not do this first, it wasn't actually even attempted. And if you did not (and I don't think you did) then your "preference" experiment yielded the same result as tossing a coin. Your answer is in my quote. Link to comment Share on other sites More sharing options...
Satanica Posted November 4, 2014 Share Posted November 4, 2014 Your answer is in my quote. You're beating around the bush so yes you have answered my question; an ABX test was NOT performed and passed. All good and good on you for giving it a go but what you guys did was just lacking the precision required to draw any kind of conclusion. Link to comment Share on other sites More sharing options...
ehtcom Posted November 4, 2014 Share Posted November 4, 2014 You're beating around the bush so yes you have answered my question; an ABX test was NOT performed and passed. All good and good on you for giving it a go but what you guys did was just lacking the precision required to draw any kind of conclusion. Where did I state it was an ABX test? Link to comment Share on other sites More sharing options...
Satanica Posted November 4, 2014 Share Posted November 4, 2014 Where did I state it was an ABX test? Exactly Link to comment Share on other sites More sharing options...
Recommended Posts