[castleofargh]

finally spent the time to read the article and there are also a few points I'm not entirely agreeing with. I mostly like what I read so it's more nitpicking than really thinking he's wrong. like saying that some electrical components cannot change behavior if checked at the same temperature. while for all intended purposes it's ok, extreme statements will always have something to prove them wrong at least a few times.
 
also I half agree with that:

 
The impedance measurements Tyll made (almost at the bottom) are more reliable IMO simply because they are measured in the electrical domain (signal path). The impedance has a direct relation to mechanical compliance and damping. I think Tyll did show that driver properties actually changed by these impedance measurements alone.
Of course it doesn’t say anything about the audibility of these small changes.

I started writing about my concerns and half understood theory, but an actual example is better. click on graph for readable size:

I just measured this right now for the cause. my hd650 with fairly new(a week old) pads on my head with earplugs for the measure. then the same impedance measurement with some 3year old pads, again on my head. and then one with the headphone lying on my desk and the cable however it fell when I put the headphone down.
it's limited to impedance, obviously no burn in the all thing took 10mn including repeated measurements for control and the headphone is several years old. yet it's not hard to get a change and even a shift in the peak frequency. so I obviously don't agree that Tyll's impedance measurement is more conclusive for changes in the driver over time.
 
for the sake of clarity, I did a few measurements and finished with another measure of the new pads to check if manipulation alone could have as much impact.I got a very tiny difference but for this scale, almost identical to the first measurement of new pad I did before changing them for old ones. just to say the variations on the graph seem significant and not simply the result of lack of precision in my measurement method.

 

[Cold Train]

I'm the kind of guy who always buy everything by pair from $10 to $100, most of my gaming headphones as well as they get in an ''unrepairable or repairs will cost as much as a new pair'' position quite often. When swapping from the old pair to a brand new one, there's a HUGE difference in terms of sound quality, I will have to use it at least 100 hours before matching the old pair.

 

[sonitus mirus]

I'm the kind of guy who always buy everything by pair from $10 to $100, most of my gaming headphones as well as they get in an ''unrepairable or repairs will cost as much as a new pair'' position quite often. When swapping from the old pair to a brand new one, there's a HUGE difference in terms of sound quality, I will have to use it at least 100 hours before matching the old pair.

Yeah, maybe.

This is what I can come up with from the top of my head. I'm sure there are probably several other scenarios that could contribute to your experiences.

1. Several models of headphones costing between $10-$100 have to burn in for at least 100 hours and undergo huge sonic changes during this period.
2. The manufacturing process for several models of headphones costing between $10-$100 have build tolerances that allow for huge variations in performance and the user gets accustomed to the sound signature after 100 hours of use.
3. Several models of headphones costing between $10-$100 wear out over time with regards to the ear pads, clamping tightness or other fit, and at 100 hours of use, it is possible to have a huge change in the sound signature.
4. The differences between several models of headphones costing between $10-$100 are nearly identical in sonic performance immediately after being unboxed and after 100 hours of use. Bias is responsible for any person believing they are hearing a huge difference between the new and old version.
5. Maybe a little of each: 1, 2, 3, 4

 

[Cold Train]

Yeah, maybe.

This is what I can come up with from the top of my head. I'm sure there are probably several other scenarios that could contribute to your experiences.

1. Several models of headphones costing between $10-$100 have to burn in for at least 100 hours and undergo huge sonic changes during this period.
2. The manufacturing process for several models of headphones costing between $10-$100 have build tolerances that allow for huge variations in performance and the user gets accustomed to the sound signature after 100 hours of use.
3. Several models of headphones costing between $10-$100 wear out over time with regards to the ear pads, clamping tightness or other fit, and at 100 hours of use, it is possible to have a huge change in the sound signature.
4. The differences between several models of headphones costing between $10-$100 are nearly identical in sonic performance immediately after being unboxed and after 100 hours of use. Bias is responsible for any person believing they are hearing a huge difference between the new and old version.
5. Maybe a little of each: 1, 2, 3, 4

I would choose the No1 but this actually applies to mostly every headphones, I also own headphones of up to $1700 (Audeze LCD X, Fostex TH900 and Ether C 1.1) the main difference with high end ones, would be a little improvement instead of a huge improvement (cf. Tyll Hertsens on page 1) , the technology does come into play, many people will deny that planar and balanced armature drivers requires break-in. The same applies to religions, everyone has an opinion and we shall be respectful.

 

[sonitus mirus]

I would choose the No1 but this actually applies to mostly every headphones, I also own headphones of up to $1700 (Audeze LCD X, Fostex TH900 and Ether C 1.1) the main difference with high end ones, would be a little improvement instead of a huge improvement (cf. Tyll Hertsens on page 1) , the technology does come into play, many people will deny that planar and balanced armature drivers requires break-in. The same applies to religions, everyone has an opinion and we shall be respectful.

Sounds good, thanks for sharing.

 

[castleofargh]

I would add, random accidents and daily handling. I'm one to try and deal with my gear as nicely as possible, even for putting the headphone on the table for 10 seconds to go grab a drink, I have a dedicated old mouse pad that's fairly thick and very flexible? it has a gummy feeling? I don't know the proper word for that ^_^. just so that I never hits a solid surface with a headphone even when they're not stored properly on a stand because I'm lazy. but despite all my care, fact is I did drop the hd650 I measured above, maybe 4 or 5 times in the last 3years. it's never from a full height and I have some thick carpet on the floor, but still crap happened. and if we start talking IEMs.... guilt is all I have thinking how badly I treat some.

but my little experience with having owned 2 pairs of BT headphones and a few pairs of cheap IEMs, they typically have an audible difference from the get go. the manufacturing accuracy isn't so great that I can't tell them apart when someone hands them to me with my eyes closed. so I would be careful about using audible differences between 2 pairs to draw conclusions on burn in. as @sonitus mirus put it, burn is but one of several possibilities. we cannot and shouldn't conclude anything about burn in without first having the testing means to control or remove the other possibilities. I know I sound like a broken record at this point, but burn in discussions seem to be a constant reminder that the process of getting conclusive evidence isn't what most people think it is. burn in is just the subject revealing the logical flaw in most testing methods brought up by members.

 

[jagwap]

I would choose the No1 but this actually applies to mostly every headphones, I also own headphones of up to $1700 (Audeze LCD X, Fostex TH900 and Ether C 1.1) the main difference with high end ones, would be a little improvement instead of a huge improvement (cf. Tyll Hertsens on page 1) , the technology does come into play, many people will deny that planar and balanced armature drivers requires break-in. The same applies to religions, everyone has an opinion and we shall be respectful.

I would add that cheap headphones get minimal testing, while high end headphones should get longer testing, and hopefully a long term (24 hours +?) test, sometimes unfortunately called "infant mortality" test, as most failures happen in the first few hours of use. This will remove the most significant burn in hours from the users experience.

 

[Cold Train]

I would add, random accidents and daily handling. I'm one to try and deal with my gear as nicely as possible, even for putting the headphone on the table for 10 seconds to go grab a drink, I have a dedicated old mouse pad that's fairly thick and very flexible? it has a gummy feeling? I don't know the proper word for that ^_^. just so that I never hits a solid surface with a headphone even when they're not stored properly on a stand because I'm lazy. but despite all my care, fact is I did drop the hd650 I measured above, maybe 4 or 5 times in the last 3years. it's never from a full height and I have some thick carpet on the floor, but still **** happened. and if we start talking IEMs.... guilt is all I have thinking how badly I treat some.

but my little experience with having owned 2 pairs of BT headphones and a few pairs of cheap IEMs, they typically have an audible difference from the get go. the manufacturing accuracy isn't so great that I can't tell them apart when someone hands them to me with my eyes closed. so I would be careful about using audible differences between 2 pairs to draw conclusions on burn in. as @sonitus mirus put it, burn is but one of several possibilities. we cannot and shouldn't conclude anything about burn in without first having the testing means to control or remove the other possibilities. I know I sound like a broken record at this point, but burn in discussions seem to be a constant reminder that the process of getting conclusive evidence isn't what most people think it is. burn in is just the subject revealing the logical flaw in most testing methods brought up by members.

Totally agree with your statement, however the undeligned sentence applies only to lower/mid tier headphones. That would be a shame to hear such discrepancies on supposed '' hand matched drivers'' on $1500+ flagships

 

[castleofargh]

as I said somewhere(maybe here? ^_^) for the most part when they pay special attention it is indeed to get matched drivers. meaning they make sure that left and right driver are within whatever margin they decided to define(FR difference within 1 or 2dB inside a predetermined frequency range). but that doesn't mean you won't get 2 pairs of the same model with 2 or 3 db variations between them. that would be another measurement standard and I'm not sure who cares for that or which margin the manufacturer uses.
so the assumption that 2 new pairs will sound the same could do with some objective verification before moving on to get one "burned in".

 

[abm0]

when they pay special attention it is indeed to get matched drivers.

Sometimes they match them so strictly they give you a pair with two left cups. ) Sorry, still haven't stopped laughing about that "overmatched" 400i that popped up on Reddit.

 

[dotrunghieu]

I'm new here and have not read most of the posts above.

Few weeks ago I got the Grado PS1000e. listened a couple hours and then burned it over 200 continuously hours. I could not tell the difference between before and after burning. It is the same as I could not tell the difference between $5 and $1000 interconnection cables (or speaker wire as long as large gauge sizes. had friends who spent several thousand dollars for cables and claimed that the sound was significantly improved but I did not heard the difference).

Obviously, I do not have golden ears as many people.

HN

I own $700 after market headphone cables. But I use stock cable most the time because stock longer and lighter than my after market cable.

Why I buy after market cables. Because I thought I can tell the differences.


But I can tell a brand new headphone difference with 100 hours burn in headphone for sure. Not even headphones, I can tell differents any type devices like laptop speakers, sound alarm devices, door bell systems, that what they are brand new or have many hours burn in sound.

 

[EtherealPoetry]

I've always assumed that it was me adjusting to headphones, not the other way around.

 

[frodeni]

as I said somewhere(maybe here? ^_^) for the most part when they pay special attention it is indeed to get matched drivers. meaning they make sure that left and right driver are within whatever margin they decided to define(FR difference within 1 or 2dB inside a predetermined frequency range). but that doesn't mean you won't get 2 pairs of the same model with 2 or 3 db variations between them. that would be another measurement standard and I'm not sure who cares for that or which margin the manufacturer uses.
so the assumption that 2 new pairs will sound the same could do with some objective verification before moving on to get one "burned in".

This assumes that the cans can be objectively measured, which is what is contested. That has serious implications. If you want to test any break-in, just hand out new and burned out sets, and use qualitative research methods.

This is naturally not about burn-in or not. Both the measurement tests and user listening experience is at question, as is clearly shown in the OP. If it is possible to target out a sonic trait, that consistently beats out the measuring device, that would be great. Not to ditch the machine, but hopefully to improve it, or to to understand what went wrong, as to use it more correctly. The worst thing that can happen, is if people beat the machine, but no quantification of why is found. If no vitrifaction of any audio able traits are found, and that continues to be the case, then we got a problem. Particularly if test are rigged to generate true positives, as to quantify what the positive might be.

If this is ever understood, then it might be useful to run a larger scale test, using the specific sonic traits. That would typically be used to support the original finding, or to test along with improved measuring techniques.

If psychology turns up as a major player, as it very well might do, an obvious choice would be to try to include that field as well.

Personally I do not get the rally going around these topics. I simply do not get why people spend time testing a headset for burn-in as reported in the OP. Amplitude readings and THD is not in correlation with peoples experience, and proving that over and over, is good for what really? I also struggle with those who ignore a ton of really solid research, and just dismisses it like it is useless. There clearly is something major that is not understood, and to be honest, doing what is needed to uncover that, is not where a lot of people is going. It actually hurt us badly, as a community.

 

[castleofargh]

man! be more like that and you'll have me as your number 1 fan.

 

[Headzone]

I have my Sony's burning in atm . going for 200hrs on this pair. besides they're going for sale soon. I also have put my AKG pads in a pad flattener machine. (two pieces of wood and a 20lbs pc.) I call that "break in".