Originally Posted by CC Lemon
The article you posted actually kind of supports the idea that break-in exists. Not definitively, but it shows more support than doubt.
Here's the article ac500 was referring to: http://www.innerfidelity.com/content/evidence-headphone-break
By no means a perfect test, but it's somewhat compelling for further investigation.
Here's the subjective test they did: http://www.innerfidelity.com/content/testing-audibility-break-effects
In theory and from the little evidence we have, there is reason to believe that break-in is possible. There is a proposed cause and some results that support the idea. I haven't really seen any proposed explanations against the idea that the mechanical movement will be affected with use over time and no evidence that suggests that there is no difference over time.
Arguing semantics is missing the point. He used the term placebo incorrectly. The meaning should have been understood. There are likely psychological effects that also contribute to changed perceptions of the sound over time. As I explained, placebo is still a possible contributing factor.
Here's the thing, my argument is never stating 'Breaking does not exist', rather that 'any evidence is still very much inconsistent to draw conclusive summation'. Which is why I found it a bit analogous to the HDMI discussion. Although the test itself and the nature and design/engineering behind speakers and headphones tends to lead to stronger presumption of break-in properties/hypotheses, it is still unporven without a shadow of a doubt.
Personally I don't really fall into any camp. When I buy headphones I do treat them to the Break-In process as much as I can, but to be honest with you, I have yet to been able to conclude audible change; or at least not the one described as per Break-In audible change. Clearly some manufacturers suggest break in period, but some others don't. Also, it is commonly known that it is very unhealthy to run speakers or headphones at blasting volumes right off the bat, rather gradually.. Not sure how much this aspect supports break in, but it certainly hints of flexibility within the driver itself and adaptation to abuse and usage.
As for subjective blind tests, even if done properly, with Double-Blind standards, ruling out Placebo effect, conducting multiple experiments via Between Subjects design and Within designs and applying every possible balancing method like Latin Square and such... We are still dealing with technology that apparently produces inconsistent performance within the said products. As already established, even with the same model from the same make (high end), the Sound and FR still fluctuates. Even if you get 2, 3, or even 10 headphones that are measured the same fresh out of the box, there is still no telling how they would respond with usage and how they will differ along the way. Yes, one can conduct intermittent measuring every so and so hours just like in the cited link, but what about the hours in between? That is still not measured. And even if you use the 'Best Guess' Method (within a marginal range), remember that anything within that marginal range could also be significant and meaningful in terms of sound and how one may hear those differences.
As for the Placebo, I wasn't arguing semantics. I was just responding to the comment ac500 made in regards that people should stop denying being under or affected by Placebo.
Yes, we all know Placebo exists and real, but regardless of that fact, it is still a mental state of mind, and naturally some people are more prone to it, while others are simply not. I'm not suggesting immunity of some sort, but while Placebo effect could be easily achieved and proved, realistically it doesn't affect people that quickly and that drastically; at least not to the levels ac500 seems to suggest.
Please remember, again, that for Placebo effect to occur, deception of the mental state of mind has to occur, and majority of the time, by the researcher.
Also, please don't confuse Demand Characteristics with Placebo, which sometimes is the case with experiments and reasearch.
Edited by Shotor102 - 3/20/12 at 11:16am