[sarge7359]

I think the only way to prove this is 1) have a 100% accurate way to test headphones 2) apply it before and after burn in.
 
Anybody got #1 all set?
 
No?  Well, get on it...

 

[ranfan]

I'm interested in hearing what a quantum mechanic theorist/scientist thinks about this
-
In any case, scientific measurement devices have what is called, measurement uncertainty.
 

Spoiler: q

https://www.wmo.int/pages/prog/gcos/documents/gruanmanuals/UK_NPL/mgpg11.pdf
 
Uncertainty of measurement is the doubt that exists about the result of any measurement. You might think that well-made rulers, clocks and thermometers should be trustworthy, and give the right answers. But for every measurement - even the most careful - there is always a margin of doubt. In everyday speech, this might be expressed as ‘give or take’ ... e.g. a stick might be two metres long ‘give or take a centimetre’.
..
Uncertainty is a quantification of the doubt about the measurement result.
 
Since there is always a margin of doubt about any measurement, we need to ask ‘How big is the margin?’ and ‘How bad is the doubt?’ Thus, two numbers are really needed in order to quantify. One is the width of the margin, or interval. The other is confidence level.

For example. We might say that the length of a certain stick measures 20 centimetres plus or minus 1 centimetre, at 95 percent confidence level. This result could be written:
 
20 cm ±1 cm, at a level of confidence of 95%.
 
The statement says that we are 95 percent sure that the stick is between 19 centimetres and 21 centimetres long.

 
So unless there is a measurement which can prove that an earphone after 'burn in' produces a different sound than before. And that the difference can be confidently said to be not within the standard deviation of the measuring device. And not affected by environment or any other factors. It is safe to assume that 'burn in' is not real.
-
An observation. Like all electronic devices, there is durability of use. I'm not sure about planned obsolesence, maybe an unplanned one, or a natural one. Eventually devices will break after some time/use. Now, as for the case of electronic sound producing devices. During this process, will there be a change in sound? And if there is, doesn't that coincide with many people' report of 'burn in' effect? I can't be sure at the moment.

 

[Intensecure]

I'm interested in hearing what a quantum mechanic theorist/scientist thinks about this

-

In any case, scientific measurement devices have what is called, measurement uncertainty.

So unless there is a measurement which can prove that an earphone after 'burn in' produces a different sound than before. And that the difference can be confidently said to be not within the standard deviation of the measuring device. And not affected by environment or any other factors. It is safe to assume that 'burn in' is not real.
-
An observation. Like all electronic devices, there is durability of use. I'm not sure about planned obsolesence, maybe an unplanned one, or a natural one. Eventually devices will break after some time/use. Now, as for the case of electronic sound producing devices. During this process, will there be a change in sound? And if there is, doesn't that coincide with many people' report of 'burn in' effect? I can't be sure at the moment.

The only time that I have ever xperienced anything that would correlate to that is an old pair of quite good earphones. After 5 years use, they would be what most would consider well burned in. As I liked them, I decided to get a second pair as back-up, and found to my horror that the new pair didn't sound like the old ones! I directly A-B tested them, and got second opinions, but the old ones were definitely more "mellow" less strident and the new ones were more like they had sounded when I originally bought them. Damn, I thought, since I do not believe in burn-in, yet these seemed to show it..
One month later, and you couldn't make this up, the old pair died. In their sleep, . The "softening" effect was probably less burn-in than burn-out as they finally reached end of life. Made me wonder, when I see these ludicrous tales of 1000 hour burnins to get the IEM to sound good, if they are not just pushing it to failure and thinking that it sounds better.
On a more general note, I'll probably be leaving Head-fi soon, because I honestly cannot take reading any more ridiculous reviews, threads and posts where people, who presumably had some level of education in their life, talk about earphones sounding so much better after 50, 100, 200+ hours of "burn-in" or how their cable change "opened out soundstage dramatically" or burning in their solid state device "doubled the bass response".
Quite frankly, the level of common sense, allied to a basic understanding of science, seems to be so lacking in such a high proportion of people who frequent these forums that it is actually concerning for the future of audio as a whole.
Ok, enough ranting for now, off to swap from silver to copper cable to bring out the warmth in my music.

 

[gregorio]

  [1] I agree that the human perception is prone to errors but that with sufficient training you can have certitudes about your perception.
[2] I mean precision athletes like bowmen or even gamers achieve insane precision without tools to mesure them all the time,
[3] or people copying paintings don't use colorimetric probes at all times,
[4] so why shouldn't I be able to notice that my IEM's have slightly more bass than before over a range of situations?

 
1. It doesn't really work like that in practise. Training doesn't improve one's hearing, it just improves our ability to dissect our perception and to focus our concentration on those dissected parts. Then there's the issue of perception vs reality, which can be improved to an extent but generally we don't want to improve it too far. We want to improve the accuracy and reliability of our ability to dissect our perception, not eliminate all the "flaws" of perception which make perception different to reality. A hypothetical audio engineer who did eliminate the "flaws" of perception, whose perception therefore matched reality, would be a terrible audio engineer because they wouldn't be able to perceive music and therefore wouldn't be able to apply changes to make it sound better. In practice, we generally have far more knowledge of our perception and of the differences between our perception and reality compared to the average person but we only have more "certitude" in some aspects.
 
2. Two issues here: Firstly, you're comparing apples and oranges. Hand/Eye coordination plus muscle memory is entirely different from hearing perception acuity and aural memory. Secondly, you seem to be confusing scale; "insane precision" relative to what? Certainly not relative to some tools, say an electron microscope for example, just relative to other less trained/skilled people. Same is true for music performance or audio engineering. Some of the "night and day" differences reported by some audiophiles go way beyond "insane precision" and into the equivalent scale of an electron microscope.
 
3. Again, there's different issues at play here, for example: Comparing and matching colour between two static images is far easier than comparing and matching colour between two constantly changing images or, I would think it would be extremely difficult to colour match without the use of tools if the copy was being created in a situation in which the variables which affect our perception of colour were significantly different (such as the lighting for instance).
 
4. Depending on how much more "slightly more" actually is and how that scales to audibility, depending on how the other variables which affect our perception vary, depending on hearing knowledge/acuity and depending on the speed of comparison relative to aural memory, then yes, you should be able to. In practise though, these variables rarely align to the point where there can be absolute certainty of an actual difference, not least because most audiophiles are ignorant of, or discount most of these variables and their "certitude" is often nothing more than a delusion. In the vast majority of (if not all) cases, it's the listener's perception which is undergoing noticeable "burn-in" rather than the equipment.
 
G

 

[sarge7359]

I recall an advert on the front page for a testing device/service available at Can Jam (SoCal?). Seems that's our chance to see how it works.

First you need a baseline - pick 3 headphones people rate as excellent and 3 that people rate as average. If at all possible, make all these headphones 'burned in' already. Maybe use demo units? Run the testing device on them. Find some commonality between the excellent versus the average. That proves the testing device and gives us measurement perceivable difference. (If the testing device gives same values for all six headphones then need a different testing device)

This is presuming the same DAC/amp setup, of course. Something reasonably nice.

Then pick the same headphones straight out of the packaging and test again. Compare versus the perceivable difference already baselined.

Now here's the biggest problem with the above plan -- I don't think burn-in matter for many headphones, but maybe it does for some. I do not have the experience to know if any of the headphones used are ones that change with burn-in.

But if you want to prove it, I think that's the change.

And the people who should really want to do this are the testing device people. If they can prove their testing device can pick out the better headphones then that's excellently marketing fodder for them.

 

[ranfan]

  2. Are you really saying that faeries, unicorns, mermaids, etc., are not "unreal", they're just unproven?

 
Quote:

  In (A), the only situation when you can say in absolute certainty there is no faeries, unicorns, mermaids, etc on earth

These mispelled 'faeries', not sure if intentional, bothers me more than it should

 

[castleofargh]

I recall an advert on the front page for a testing device/service available at Can Jam (SoCal?). Seems that's our chance to see how it works.

First you need a baseline - pick 3 headphones people rate as excellent and 3 that people rate as average. If at all possible, make all these headphones 'burned in' already. Maybe use demo units? Run the testing device on them. Find some commonality between the excellent versus the average. That proves the testing device and gives us measurement perceivable difference. (If the testing device gives same values for all six headphones then need a different testing device)

This is presuming the same DAC/amp setup, of course. Something reasonably nice.

Then pick the same headphones straight out of the packaging and test again. Compare versus the perceivable difference already baselined.

Now here's the biggest problem with the above plan -- I don't think burn-in matter for many headphones, but maybe it does for some. I do not have the experience to know if any of the headphones used are ones that change with burn-in.

But if you want to prove it, I think that's the change.

And the people who should really want to do this are the testing device people. If they can prove their testing device can pick out the better headphones then that's excellently marketing fodder for them.

this simply doesn't work. 2 pairs of the same model will measure and very possibly sound different from the get go. there are other problems, but this one is a deal killer.
you can measure your gear out of curiosity(like I do when I get new stuff), but proper testing to answer questions about burn in even limited to one headphone model, would require statistically significant headphone quantities. else the results are of little value. the manufacturing variations will often show as much or more differences than all the stuff accounted as burn in can do(including the huge impact from pads being pressed and altered by sweat over time).
manufacturers for sure have such data for their own gears. having some samples run for X hours with measurements at different intervals seems like something everybody with half a fiber of professionalism would do from time to time. at least when developing new gears. but we silly consumers only deserve marketing and urban legends. real data is often said to "confuse" us 

.

 

[sarge7359]

 
this simply doesn't work. 2 pairs of the same model will measure and very possibly sound different from the get go. there are other problems, but this one is a deal killer.
you can measure your gear out of curiosity(like I do when I get new stuff), but proper testing to answer questions about burn in even limited to one headphone model, would require statistically significant headphone quantities. else the results are of little value. the manufacturing variations will often show as much or more differences than all the stuff accounted as burn in can do(including the huge impact from pads being pressed and altered by sweat over time).
manufacturers for sure have such data for their own gears. having some samples run for X hours with measurements at different intervals seems like something everybody with half a fiber of professionalism would do from time to time. at least when developing new gears. but we silly consumers only deserve marketing and urban legends. real data is often said to "confuse" us 

.

IF you are saying that the differences between 2 pairs of the same model will vary more than the differences between different models, without burn-in even being a factor, then... what's the point of rating a headphone model at all?
I'm hoping that's not your position.  The whole point of the testing equipment is to make sure the manufacturing process produces consistent measurable results.  The whole point of manufacturing is to produce consistent results.  If the resulting product varies so widely, then the manufacturing process has failed.
 
IF the testing equipment can differentiate between 'excellent' and 'good' headphones then the same process could be applied to 2 pairs of new or 'burned in' headphones.  If there is a significant measurable difference between 2 random NEW pairs of the same model then that's huge news for fans.  If there is a difference is attributed to 'burn-in' then that is also testable, assuming the burn in process takes less time than availability of test equipment.
 
 

 

[sonitus mirus]

We simply need to have some idea of the manufacturing tolerances with any particular model before we could even begin to test for potential differences from burn in.  You also have to account for fit/placement and physical wear on anything that might impact the fit/placement.  The differences being measured on the few sites that post anything remotely useful are not showing drastic changes, and these could all be well within the manufacturer's tolerances or simply the result of squished foam in the ear pad or dried material.  This is assuming the goal is to see if the transducer is changing over time by any significant amount.

 

[castleofargh]

  IF you are saying that the differences between 2 pairs of the same model will vary more than the differences between different models, without burn-in even being a factor, then... what's the point of rating a headphone model at all?
I'm hoping that's not your position.  The whole point of the testing equipment is to make sure the manufacturing process produces consistent measurable results.  The whole point of manufacturing is to produce consistent results.  If the resulting product varies so widely, then the manufacturing process has failed.
 
IF the testing equipment can differentiate between 'excellent' and 'good' headphones then the same process could be applied to 2 pairs of new or 'burned in' headphones.  If there is a significant measurable difference between 2 random NEW pairs of the same model then that's huge news for fans.  If there is a difference is attributed to 'burn-in' then that is also testable, assuming the burn in process takes less time than availability of test equipment.
 
 
 

http://www.innerfidelity.com/images/SennheiserHD800Ssn01067.pdf
http://www.innerfidelity.com/images/SennheiserHD800Ssn01070.pdf
I'm taking what is most likely a reference of stability and consistent manufacturing. TOTL Senn.
maybe it's Tyll's measurement method involving several positions and averaging that creates most of the differences here, but go ask anybody who has done some headphone or IEM measurements. I think consumers are indeed mostly unaware, else they would never waste time on how to lower -120db jitter, or on the merits of 24bit files. 2 drivers are rarely identical even on the same pair. I consider that we can't examine that on Tyll's measurements because each ear is measured by a different coupler, but I often measure stuff with 2 to 5db variations somewhere in the response between left and right with the gears I get my hands on, and that while doing my very best to measure until I get the closest result.  I do measurements trying to get a reliable placement to change global imbalance if there is one(and to measure sensitivity), then I try to see if I can align the response by moving stuff around a little. and after that extra effort, I'm still often in a 2 to 5 db variation somewhere in the audible range.
 
the er4sr or er4xr would probably be a good reference of stability for IEMs, like the hd800 for headphones. but all the same, you can go look for the some pics of the FR provided on the certificate by some users on the er4 topic, or just ask for a few, and you'll be able to see how variations from pair to pair aren't insignificant(and the Ety guys recognize it openly).
my pair of er4sr has the best channel matching out of all the IEMs I've measured. and it still has almost 1db over a few hundred hz in the midrange.
 
and all the same for headphone we would need to measure without the pads(unless we count the fancy "burn in" to be mostly pad crushing), and same for IEMs, for good consistency, it's better to measure without tips. overall it's a real bother TBH, I've tried like twice, one IEM one headphone, and the changes made me feel like I had absolutely wasted my time. again unless we count the pads, then yes burn in is real, burn in is almost always audible with enough time. pad definitely change over time, altering the seal and bringing the driver closer. 

 

[Russin]

  http://rinchoi.blogspot.fr/2012/04/effect-of-break-in-sony-mdr-ex1000.html
http://rinchoi.blogspot.fr/2012/04/introduction-it-is-generally-known-that.html
http://rinchoi.blogspot.fr/2013/01/the-effect-of-break-in-vsonic-vc02.html
http://rinchoi.blogspot.fr/2014/02/the-effect-of-break-in-creative-aurvana.html

 
Wow, talk about dramatic - you'd think if there were 10db differences along certain parts of the spectrum, that not only would this be well-accepted, but manufacturers might even mention it, no?
 
It really makes you think about the fickleness of measurement on the whole and how subjective the hobby can feel at times despite all of the "objective" instrumentation at our disposal.

 

[sarge7359]

The Rin Choi articles are fascinating around this subject. There are differences measured but for some the differences are tiny. Differences in 'pad crushing' would be more dramatic and I recall Rin mentioning the same in one of the articles.

http://rinchoi.blogspot.fr/2012/04/introduction-it-is-generally-known-that.html -- conclusion is that pad crushing is most noticeable with some small differences in 'warm up' with Sennheiser HD650.

http://rinchoi.blogspot.fr/2014/02/the-effect-of-break-in-creative-aurvana.html -- conclusion says effect is evident. The energy time curve and group delay effects do seem significant.

http://rinchoi.blogspot.fr/2012/04/effect-of-break-in-sony-mdr-ex1000.html -- conclusion says effect is there BUT it isn't night and day. "Still, a radical sonic metamorphosing quality, of which audiophiles usually refer to, is nowhere to be seen." The graphs all appear very similar.

http://rinchoi.blogspot.fr/2013/01/the-effect-of-break-in-vsonic-vc02.html -- manufacturer says there is a difference, conclusion says it is noticeable. IS THIS PROOF FOR SOME HEADPHONES? (Frankly I'm not seeing a huge difference, but hard to tell with the measurement graphs) ... also interestingly this is the article with the most comments, many of them questioning the testing process. Looks like same testing process used before, but nobody questions that.


In the comments of the vsonic article somebody mentions: "The German loudspeaker manufacturer Nubert made a couple of (blind) listening tests with cables. They constructed the experiment very carefully (transparent switching device, level matching etc.). Listeners failed to distinguish the cables they tested.

Then they increased the level of one cable by +1 dB. Now listeners were able to identify (statistically significant) the "better" (louder) cable, even if they tested two identical cables."
^^THIS is maybe the real effect for most headphones on burn-in. In general with all the measurements you see a reduction in impedance after burn-in, so therefore louder yeh?

 

[Russin]

Then they increased the level of one cable by +1 dB. Now listeners were able to identify (statistically significant) the "better" (louder) cable, even if they tested two identical cables."
^^THIS is maybe the real effect for most headphones on burn-in. In general with all the measurements you see a reduction in impedance after burn-in, so therefore louder yeh?

 
Super compelling if true - especially if most people are using discrete volume controls (like the volume up/down button on a phone or the output volume on a Mac). Is it easy to show that impedance is legitimately reduced over time in a pair of headphones?

 

[castleofargh]

well, blind tests are recommended to be done within 0.1db matching. it doesn't mean that more is always obvious, but it has the potential to mess up with whatever else we're trying to test. and has been demonstrated to do so many times. you have a point about setting up the level on certain gears. I kind of enjoy testing DAPs, and aside from the nightmare that is aligning the track on 2 devices, volume matching can be hard to do as a good deal of devices will usually have like 0.5db or 1db increments at best.
 
about impedance changing over time, it's not hard to measure the headphone's impedance as long as everything else stays the same(including the measurement gear). in my case, I've lost the first little box of mess I had done for that purpose, then I used a blob of crocodile plugs for a short time, and now I've finally soldered something that hopefully I'll keep around for years. but right now I can't really take measurements from my V1 and use them to check possible changes with V3. I'll have to wait.
on rare occasions, I've had more variations from just changing how the cable is laid down on my desk than what is showed on Rin's blog. and talking about cables, I'm a black belt in cable abuse, should I consider the possible changes in the cable as burn in or as me ruining it? same for the headphone, the few times I dropped it, if that creates a measurable change, is it burn in?

 

[Pimmsley]

On a more general note, I'll probably be leaving Head-fi soon, because I honestly cannot take reading any more ridiculous reviews, threads and posts where people, who presumably had some level of education in their life, talk about earphones sounding so much better after 50, 100, 200+ hours of "burn-in" or how their cable change "opened out soundstage dramatically" or burning in their solid state device "doubled the bass response".
Quite frankly, the level of common sense, allied to a basic understanding of science, seems to be so lacking in such a high proportion of people who frequent these forums that it is actually concerning for the future of audio as a whole.
Ok, enough ranting for now, off to swap from silver to copper cable to bring out the warmth in my music.

You made some excellent points... And I feel the same way... I haven't been around for at least 5 years, came back for some research/advice and found what appears to be the worst of disinformation, conclusion jumping and just plain false advice...
The worst though is the almost religious behaviours of false idol worship and mob mentality leading to bullying and arrogance of stunning proportion...
I have worked in the pro audio industry for 20+ years and the debates over cable materials is hilarious... When I showed some arguments to a fellow engineer he almost crapped his pants with disbelief... Just wish we thought to start a boutique cable company years ago instead of working my ass off...
I get angry reading almost any random thread reading things that should not exist knowing what we know about physics and electronics...
Where do people get the confidence to believe their own BS or blindly regurgitate others falsities ???
I'm almost convinced the silver (better tops) and gold (warmer) argument is simply an affect of the emotion the colour of the metal imparts on the listener...
I wonder if there are any blind head-fiers that have been brave enough to chime in and pass on the wonders of how amazing and complicated and connected to the mind hearing really is, and how large a part psychoacoustics play in every minute of conscious listening...we had to study this as part of my audio engineering studies in 1993.

Reading your post and this section of the forum gives me some solice that all is not lost... but the main sections seem full of confused and very wrong absolutes..