ueyteuor
1000+ Head-Fier
- Joined
- Apr 8, 2007
- Posts
- 1,379
- Likes
- 12
burn in is true
/thread
/thread
Originally Posted by dgbiker1 /img/forum/go_quote.gif He also talks about how 0s and 1s can't be corrupted in the signal stream. This is also completely wrong. It is less susceptible to line noise for sure, but you can lose 0s and 1s. In reality the 1 is a TTL signal, around 5V +/- some threshold. It is entirely possible for line noise to pass that threshold and send a false value. It is more robust than an analog signal though since it can handle some line noise without changing the actual signal, but it's not bulletproof as the author suggests. |
Originally Posted by vcoheda /img/forum/go_quote.gif buy 3 K701s (since that is said to be very prone to burn in). burn in 2 of them (or 1) and not the other. then have someone - someone who is familiar with the K701 sound - blindly listen to them and see if he or she can tell you which 2 sound the same and which different. repeat a few times. it wouldn't be conclusive, but it would be interesting. |
Originally Posted by Acoustic Chef /img/forum/go_quote.gif #1734 with ~9000 hours (broke). #17835 With ~ 700 hours. #15233 With ~10 hours. #1734 -> #17835 was night and day. But #1734 was also the old build. Still #17835 -> #15233 very differant. |
Originally Posted by ueyteuor /img/forum/go_quote.gif burn in is true /thread |
Originally Posted by amanieux /img/forum/go_quote.gif this is totally false, of course scientific interpretations of FACTS (FACTS=experimental results and not the interpretation you make of it) are changing in time with the constantly evolving threories but numbers measured in an experiment are FACTS that remains valid. if you measure the sound from headphones using a microphones you interpret it as you like but these numbers will not change depending on what most of us believe in. example : * fact is : we measure a difference of 0.1dB over 16Khz before and after burn in. * interpretation is : the best human ear cannot identify a difference under 0.3db over 16khz so we conclude burn in is busted. so if you want to put in question the interpreation it is up to you but you will have to bring evidence that you can hear a sound difference under 0.3db over 16khz. note : the number of 0.1 db and 0.3 db is pure invention, i don't have the real numbers, it is just to ilustrate my point. i am a computer scientist not an audio specialist so you guy must have these numbers and draw the conclusion by yourself if your a minimum honest. i have no interest of proving that burn in exist or not, i am just curious (as every scientist should be). |
Have you see how many variables must be accounted for even before you can try to scientifically prove/disprove the theory of burn in? If you can account for all them so far, we can start discussing how to collect data and what those data means.How are you planning to define the term 'headphone'? To totally disprove the effect of burn in, you must account for every kind of headphones in every style and shape (IEM, canalphone, dynamic open/close, electrostatic, 5.1 channels, what-ever) , than you must conduct experiment on every one of them, can you do that?
Second, how are you planning to define the term 'headphone SQ'? I will have no clue about this. What quality of sound must a headphone have? Is it the official measurement done by the manufacturer of the 'phone? Can we account for the variation within the same model in every single case? If we can, what scientific study has been done to verify our theory/assumption that such variation can be accounted for?
Than comes the big one: The 'noticeable change' by the 'user'.
How are you planning to define the term 'user'? Is it everyone that capable of using a headphone? How about the hearing capacity of them? Do we have to specify only the golden ear can be enlisted for trial? Certainly not. So I think we have to include a few thousands of user (probably more) from 5yrs-old to 85yrs-old in multiple classes which based not only on age, but also living environment, habit, biological inherent and heath? That's the only way I see that can be called a 'fair' trial.
Than how are you planning to define the term 'noticeable change' in hearing? I am sure you can try and source a dozen of scientific study done on the human hearing sensitivity in responds of sight change across hearing range, but I doubt you will find an absolute number that tell you 'such and such dB can be detected but such and such can not'. Well, good luck with that.
In the end, how are you planning to define the terms 'a certain amount time', and 'cumulative active usage'? How can you be certain that 100hrs of cumulative usage on one headphone equal the same effect on another? If you can, what is the scientific proof that 100hrs is better than 500hrs, or 1000 hrs?
Originally Posted by amanieux /img/forum/go_quote.gif because my 2experiments are not significant , a valid experiment must be done mutiple time and in multiple configurations to bring enough material in order to start doing interpretations and conclusions. i was just asking people to put here all their experiences to collect data. if your read me carefully i wrote " from my 2 experiences , (unless my new headphones had already been burned in at the factory (audio technica, grado)), my conclusion is : burn-in is busted ". and it is because i am aware that 2 experiences is not significant that i proposed us to regoup all our experiences so we can draw a FINAL CONCLUSION. a final conclusion which i have not made yet because i read evidences from people that had positive experiences of burn-in. ( especially interested by the post that said that more than one person noted differences from a headphone to headphone comparison beteween an old and a new headphone of the same model but again this is not conclusive because the difference could have pre-existed out of the factory, this experiment must be done with more than 2 headphones like 10 people doing blind test from 10 new headphones and 10 burned-in headphones of the same model). |
Originally Posted by ClieOS /img/forum/go_quote.gif Honestly speaking (and no offense), your understanding of science is not impressing. We can have a discussion of the fundamental theory of science and how a proper scientific experiment should be carried out, but it will probably take a few weeks, so here is the short/simple version: 1) You observe the nature (or any setting related to your interest) and find an observation/claim/theory that hold a particular interest to you. In this case, the theory/claim of headphone burn-in, as defined as the phenomenon of noticeable change in headphone SQ to the user after a certain amount of time in cumulative active usage. In layman term: 'My headphone sounds better after XXXhrs of usage! 2) You define a theory which you like to challenge (prove/disprove): I believe your is: "There is no headphone burn in". In the world of science, it will be 'There is NO noticeable change of headphone SQ to the user after a certain amount of time in cumulative active usage' 3) You define an opposite theory, which is '"There is headphone burn in" or 'There is noticeable change of headphone SQ to the user after a certain amount of time in cumulative active usage' 4) You define the parameter of which your theory has based on. In your case, you need to define these term: 'noticeable change', 'headphone', 'headphone SQ', 'user', 'a certain amount time', and 'cumulative active usage'. I can keep writing, but here we already encountered a few fundamental fault of such an experiment: Have you see how many variables must be accounted for even before you can try to scientifically prove/disprove the theory of burn in? If you can account for all them so far, we can start discussing how to collect data and what those data means. Here is a preview on data collection: You can make a thousand measurement on the same pair of headphone with the most sensitive device there is, and I can absolutely sure that you will not get the same result repeated a thousand times. In fact, you will be lucky if you get 2/3 of your result close to what you wanted. We talk about the 'significant' of the data we collected while we discarded that we believe to be mathematically insignificant to our theory (Be warned however, insignificant is NOT equal to irrelevant). The most basic rule of being a Scientists is, we don't deal with 'absolute', no even when collecting data. Even the best experiment you can make up will not cover every single being you are testing, therefore any data or conclusion from any experiment contents not the 'fact', but merely the 'suggestion of the fact'. For example, an experiment is testing the color of a certain species of fly in a specific region in UK. After the experiment which involving catching 2/3 of the total population of such species of fly, the scientist observes that all of them are blue in color, therefore they draw the conclusion that such species of fly in such region in UK is blue in color, in 98% of certainty. So why didn't they say it in 100% certainty? Cause they can not, and will never be sure that there will not be an orange colored fly living in the deep forest of the same region, and it is impossible for them to catch every one of such fly (well, unless God is helping them, but lets leave God out of this). Does data actually tell us 'Fact'? I am very much doubt so. Well, i was thinking of replying that^, but now it is very much redundant after my reply above. Anyway, if you have a lots of money to burn, I am sure there are scientists out there (me included) ready to do a scientific research for you. No desired result guaranteed though. |
Originally Posted by amanieux /img/forum/go_quote.gif it just take one head fi member that buys a new k701 with a microphone and an audio freeware editor. maybee it wont satisfy you but with just this unique result i would be happy. |
Originally Posted by Acoustic Chef /img/forum/go_quote.gif #1734 with ~9000 hours (broke). #17835 With ~ 700 hours. #15233 With ~10 hours. #1734 -> #17835 was night and day. But #1734 was also the old build. Still #17835 -> #15233 very differant. |
Originally Posted by vcoheda /img/forum/go_quote.gif buy 3 K701s (since that is said to be very prone to burn in). burn in 2 of them (or 1) and not the other. then have someone - someone who is familiar with the K701 sound - blindly listen to them and see if he or she can tell you which 2 sound the same and which different. repeat a few times. it wouldn't be conclusive, but it would be interesting. |
...the resulting amplitude response graphs indicated that an end user would likely encounter larger system-to-system amplitude response differences (~1.04 dB Spl) owing to normal driver variances than would be encountered breaking in raw drivers... Speaker Break In: Fact or Fiction? - page 2 — Audioholics Home Theater Reviews and News |
Originally Posted by ClieOS /img/forum/go_quote.gif If that is so, and since you are the one who are most interested in proving/unproving burn-in, why not do it yourself? After all, there isn't much a reason to ask another member to spend the money and time so you can satisfy your curiosity, right? |
Originally Posted by amanieux /img/forum/go_quote.gif am i the only one being curious about burn in reality being measured ? |
Originally Posted by amanieux /img/forum/go_quote.gif why not ask if someone already made this simple test ? isn't what forums are made for ? sharing information ? by the way i had my response in a previous post : Speaker Break In: Fact or Fiction? — Audioholics Home Theater Reviews and News |