[
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2007, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]
58.179.100.90
In Reply to: RE: Blame the message, not the messenger! posted by KlausR. on December 23, 2007 at 02:51:37
You continue to point out the limitations of many of the experiments undertaken:
"In 1989 Olive & Toole used a single loudspeaker and a single reflection (experiments under anechoic conditions). They further used speech, not music.
In 1995/1996 Bech went a step further by investigating a complex sound field with multiple reflections. His conclusion was that for noise and speech only the floor reflection is strong enough that it contributes individually to the spatial aspects of the sound field. Also Bech used only a single loudspeaker."
You point out that music wasn't used, that only a single loudspeaker was used. You quote no studies of music and two or more speakers yet you say "Scientific evidence says, or strongly suggests, that reflections are not a problem". You criticise our reliance on personal experience with treatment in our rooms, listening to music reproduced on 2 or more channels, as being purely anecdotal and not convincing because it's not obtained under controlled conditions but you're convinced by "evidence" obtained under controlled conditions using source material that wasn't music and which relied on a single speaker. That's not overly compelling evidence because it says nothing about music and multi-channel playback.
You take research results, draw conclusions from them, and say this is the way the world is. That's not science. Science doesn't determine what will happen, it makes predictions and then attempts to confirm them. It doesn't make predictions and say "this is the way things are—if you hear something else you have to be wrong" yet that is, in effect, what you are telling us. You're saying that our observations are useless and invalid because they fail to support what non-comparable test data suggests. You criticise the "lack of controlled conditions" but, if there is a problem with our anecdotal reports, it's not that the conditions were uncontrolled but rather that the data gathering is uncontrolled. Test conditions certainly need to be specifiable and repeatable but what really is critical is the data gathering from the test which needs to be not only repeatable but also valid for the purposes for which it is to be used.
When I did a research project 16 years ago in my post graduate studies, I was interested in confirming or disconfirming an eye testing recommendation for which I could find no experimental evidence. That recommendation was contained in a paper co-authored by a professor of optometry and a professor of opthalmology. My study tended to confirm the recommendation and I subsequently presented my research at a conference chaired by the professor of optometry who had co-authored the paper containing the recommendation. We spoke after my presentation and I asked him where he and his co-author got their data. He said there was none and that their recommendation was simply based on personal experience. Mere anecdotal evidence but my study provided support for that recommendation.
If I learned one thing from that experience, it was that quite often the research that we would like to have available for our information simply does not exist and yet, despite that, people can and do reach conclusions on the basis of their own personal and quite anecdotal evidence and those conclusions can be subsequently confirmed. Anecdotal evidence need not be unsound but it certainly does need to be tested at some stage if at all possible and the way it needs to be tested is with tests that are conducted under conditions relevant to the issue. You can't test evidence about what happens with music and multi-channel reproduction by relying on evidence about speech and noise and single channel reproduction.
So what we have is a situation where people like Ethan and I are making claims based on our experience of music and multi-channel reproduction that isn't as rigorously obtained and presented as your data, but your data is about other sources of sound and single channel reproduction and there's no evidence to show that your data is appropriate to the circumstances we're talking about. There are problems with the data on both sides but at least we're drawing conclusions and making recommendations about the same sort of situation in which our experience was gathered. On the other hand, you're drawing conclusions and making recommendations about situations that aren't the same sort of situation as that in which your data was gathered.
If there is one big strength in the scientific method, it's that it relies on relevant data that meets certain standards. Here we have one side with relevant data that doesn't meet the standards and an opposing side with data that meets the standards but isn't precisely relevant. Neither side can totally claim the high ground in the debate but, if we're going to make recommendations about acoustic treatment for music and multi-channel reproduction I personally would rather rely on data obtained using music and two or more channels, especially if there is anecdotal data from a number of different people and that data is tending to be consistent which I believe is the case here.
You may prefer to rely on data that doesn't relate to music and multi-channel reproduction because the data is more rigorous and that's your choice but it reminds me of an interesting teaching story about the man who drops his keys in the street. A stranger comes along, sees him searching for something in the area under a street light, and asks what he is looking for. The man replies that he dropped his keys and is looking for them. The stranger asks where he dropped them and the man points to the shadows on the other side of the street. The stranger asks the man why he's looking on this side of the street if he dropped them on the other side and is told "Because the light is better here". The data may certainly provide more light on your side of the street but the answer we're looking for is in the darkness on the other side of the street where Ethan, I and many others are looking. The answer will eventually be found on our side of the street. There's no guarantee that our current conclusions are right but at least we're looking in the right spot to find what we're looking for.
David Aiken
Follow Ups:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2007, Audio Asylum, all rights reserved ]