Do you believe in the famous "burn-in"?

  • I will never believe in it
  • I believe only in Hifiman products
  • I believe only in planar drivers
  • I believe only in dinamic drivers
  • I always believe it exists.
  • Exists and can change headphone technicalities
  • It exists but is minimal and imperceptible.
  • It exists but is noticeable with more than 100 hours.
  • It exists but is noticeable with only 20 hours.
  • I will never know

0 voters

There is no “believe” here. All components change with age. Some capacitors have it stated in their datasheets how they only fully form in the first 24 hours of use.

Question is how noticeably burn-in affects sound.


Edit:
image
From here

3 Likes

I don’t really think the answer is as simple as the hard and fast rules as in the poll above. It really all depends

From a scientific perspective yes it clearly exists, but the question really shouldn’t be if it exists, but if it actually matters/is audible to the listener. On some products you might find nothing really changes overtime and that it doesn’t really matter, on others it might be more noticeable. So the argument really becomes if it’s something the listener should care about or not. Some mfg will even take time to burn in stuff at the factory, some will not, some will suggest you let something run in for a specific amount of time, others will not, really all depends on the thing at hand. Personally I’d give things time regardless, I’d just leave something on/playing for a few days just in case because imo that should get you far enough through most burn in periods if it is consequential. Burn in/break in can happen to all types of audio equipment, both with transducers and electronics, if it’s actually audible or significant will depend on the specific product at hand.

Something else to consider is if what you are hearing is actually burn in/break in or if it’s just acclimation to something new, aka brain burn in. I’d say in most cases this is more impactful than the mechanical/electrical burn in you might experience, but again all depends. IMO giving things time is just really beneficial overall regardless if you care about burn in/break in or not

5 Likes

Why put it to a question of faith? Surely someone has done some scientific assessment with some resultant data and measurements to support or reject the hypothesis.

1 Like

Presumably it’s also an issue of varying QA testing. If two manufacturers make clones of the same headphone (Sivga/Blon for instance) and one does a dozen hours of testing on each headphone while the other only tests a sample of each batch, the efficacy of burn in on the two headphones will likely be completely different to the end user even though the drivers are basically identical.

I’ll leave this here, maybe of use to some, presented without further comment.

(From Floyd Toole’s Sound Reprodcution Book)

2 Likes

I wonder why noone in the audio world looks at time domain behaviour, it is always frequency domain…

1 Like

By the way if this is in fact a “myth”, or in the realm of the difference being basically inaudible, then I would consider it one of the more harmless myths. The only cost is time, consumers aren’t being roped into spending money on snake oil.

2 Likes

Some of the statements and testing seems a bit ignorant and dismissive right away, so I’m not surprised they got to that result. Only measurement done was frequency response apparently, would have liked to see a proper suite of measurements to more accurately see what was going on, fr is only one piece of the puzzle. This really reads like “I haven’t seen any tests that I agree with, so I made one that I could agree with”. The comment about always improving performance also seems like a simple answer to me, it’s likely because a manufacturer isn’t going to design something to sound it’s best with brand new components, the design they come up with is likely made with components that have already gone through the process and have different values/characteristics/tolerances/etc than if they were new, so break in/burn in brings it closer to how it was designed to perform (if something was designed to sound best before anything was broken in that would be pretty dumb, and also I’ve seen people prefer things pre burn in more than post, more often than I’d thought I’d would, since it is all preference in the end). And things can sometimes change over time, drivers might wear out, some electronics will break down or change overtime, although that’s more wearing out than breaking in. A lot of things actually have to re break in if they sit for a long time, for woofers as an example sitting over a long time unused stiffens them again so even older speakers can re go through break in. Another example would be capacitors, as mentioned they will change when new in a forming process where the structure of the materials will change, and the chemistry of these capacitors sometimes makes it so after sitting unenergized it may need to go through that process once again

It’s really interesting that the concept of break in/burn in is so controversial in audio when it’s commonplace and accepted in other industries

Let’s ignore finding controlled testing examples for a second, why would this be so hotly debated enough to be considered controversial if it was easily proven or disproven buy a simple test? If it was really as open and shut as some make it out to be, would we be here having this discussion? Would there really be many thousands of threads and discussion around this topic? Would we see manufacturers have such wide differing stances in burn in and how they may accomplish that in their own facilities? etc, you get the point, it’s really not something that’s either easily proven or disproving as mattering, there’s merit to both sides, the truth is somewhere in-between. True actual scientific testing is extremely costly, intensive, and most importantly very time intensive, it is not as simple as taking basic measurements or creating a small single method single group listening test, there’s a whole lot more that goes into actual science, it’s quite a process, and you will rarely find anything in audio that’s truly scientifically proven, most of the time those budgets, energy, and resources will go toward something (unfortunately) more critical and beneficial/important to the larger public rather than putting all this time and resources into proving something for a small niche subset of a subset of a population so they can stop arguing in online forums lol

The real question at hand here is meaningful change though, so it’s really going to come down to the skill and the conditions of the listener, and the product at hand, and that’s typically going to lead to very mixed experiences every time. All I can really personally go off of are my own experiences, and with the amount of stuff I’ve tried and the amount of time I’ve put into audio, my results are… inconclusive, I have no idea lol, it all depends as I can see the argument for either side

1 Like

So, I’m not really interested in debating this, as I said, it’s a fairly harmless myth (if it’s a myth).

But to be clear, the myth isn’t whether things change, I think it is safe to say “things change”, it is whether those changes are audible, and whether certain claims about the soundstage “opening up” or whatever are reliable.

What would seem also true to me, is that our subjective impressions are often unreliable, we can get accustomed to different sound signatures over time, and we can think we are perceiving something more or less when we listen with some extra focus, or a different volume etc.

Then why make the posts in the first place? If you are going to comment something that asks a question or provokes a response, then would it not be reasonable to assume it would start some debate or discourse? This isn’t meant to be rude, just curious

I would agree

There’s a ton of things that go into it for sure, although those are the impressions that really matter imo, since that’s how something is actually being used in the end. Regardless of how things look on paper, they aren’t going to be used on an audio analyzer (if you buy audio gear so you can watch it draw lines on a pc, you might be doing this hobby wrong lol), so I’d personally not dismiss subjective opinion even if it is technically less reliable, it could end up mattering more, who knows. Ideally a mix of both is nice, use that measured data to support the listening, not vice versa. I obviously err on the side of actual listening to come to my conclusions as I prefer to evaluate things in practice rather than theory, but either way can produce results in some fashion, I’m pretty sure that with the recent monoblocks I bought the designer didn’t even listen to the amp when creating it and relied solely on measurements to design it, and only heard them for the first time when setting them up to demo at a show lol (hey but it worked out for him, they sound great)

Well I thought my statement regarding the harmlessness of the myth would clarify my position. Why did I post in the first place? To add some relevant information to the discussion. The previous post was pertinent to the conversation, and not really a statement on my position, rather the presented opinion of a well regarded expert. My last reply was just further clarification. I honestly don’t have a strong opinion on this, nor much motivation to argue about it.

More broadly (outside of the scope of this topic), I do think giving so much weight/respect to subjective impressions leads to confusion, and unrpredictable outcomes. I spoke in another thread about the difficulty an inexperienced consumer can have when trying to understand how to optimally spend their money on audio equipment. Should they buy this headphone or that headphone, does it need some amp or dac, should they spend more on an amp or dac, should a specific headphone be paired with some special type of amp or dac, does it need special cables, should I not listen to it for the first 200 hours while it burns in with pink noise, should i be lisening to 24bit flac files, or is vinyl better etc etc?

In the context of all of the above, “burn in” as I said is a relatively harmless concept, hence my low motivation to argue about it.

PS you seem to be editing your posts multiple time while I compose a response, it’s hard to keep up with you :wink:

Gotcha, I just read the posts and somewhat more took it as a posed question rather than just information pertaining to the topic (although it’s also not unreasonable to assume any posted information would be commented on as well)

It really sucks, the only way to truly know is experience, they will need to jump in on something and go from there to really know what’s right for them, it will only come from experience, just an unfortunate aspect of this hobby, there’s no easy answers and no clear cut ways to go without direct experience

Fair enough, in the end I don’t really feel strongly either, regardless of burn in or no burn in, spending more time with one’s gear and getting better acclimated to things is beneficial lol

Sorry sorry it’s a bad habit lol, just like using too many tla in my posts as well

Or do I even fucking care :ballot_box_with_check: plug in your new set/amp/dac, listen to it for a week or two and then decide if you like your purchase :man_shrugging:

4 Likes

Since this is venturing off-topic, I “hide” the following away.

Off-topic

Purely on the technical side here, “Soundstage”, as in how big the reproduced space is, is an interesting subject to measure.
For insanely large space(s), you would see Soundstage as Red-Shift on a spectrum analyzer when compared to the input signal. Operating word there is compared to! A single peak means nothing since it lacks context, a wide peak means someone needs to tune their filters for the input signal better.

Realistically, Soundstage is a time-domain phenomenon (meaning the horizontal axis has t or seconds on it). As such, it would be hard to measure using FR-graphs (which are frequency domain).

An experiment I personally would like to attempt is to measure decay-time of an input pulse. This time, in theory, should be longer for a wide-stage headphone and shorter for a closed-in model. Problems on measurement side aside, this would only be skewed by resonances in the headphone.

I have seen people “volume match” using a hardwarestore check-out special multimeter that guesses the voltage more than actually measureing it.
To make this test somewhat meaningful, one would need to hook the headphone to the amp and measure voltage in the loaded state. Problems here (ignoring the additional load presented by the meter) come from bandwidth (most hand-held meters only go up to 1kHz for a sinewave) and from the uneven impedance the headphones present at various frequencies. (Example HD650 below)
image
Source
So this basically nails 1 headphone at 1 frequency down, leaving everything else to uncontrollably pivot about…

The more useful (and even more difficult) method would be to play various frequencies through the headphone/speaker and take accurate sound measurements. Problem here is the price of an accurate dB-meter (and measurement mic for it) and the need to get environment noise down so the resulting curve represents the headphone, not the nearby construction site.

1 Like

Easiest solution to the problem here lol

2 Likes

I appreciate your comments, but as you noted, off-topic, so I won’t pollute the thread further.

Personally, I don’t question or focus on it too much.

If something doesn’t sound fascinating out of the box, I will leave it to burn-in but will not do it while the IEM/headphone is on my head.

I strongly believe that psychological burn-in is a thing and what has many people going nuts.

I am more of a break-in type of guy, but whenever I do this I am not listening to the device.

If something sounds good out of the box, I judge that product out of the box.

2 Likes

Probably also product related.
If the product has been factory tested / “burned in” so will 100% perform as it should, before leaving the factory.
Consumer does not have to worry about a thing. Plug & Play.

Some products just leaves the factory and hope for the best with a % error margin it will work at all, including those that consumer does not even notice.

2 Likes

I just want to know what you think. I know that they are simple and short answers but I want to reach more people without putting so many technicalities… Besides, that’s what it’s all about, not putting 7-8 hypotheses to choose from.
Hifiman in all its products says that about 150 hours to get the best experience.