Dan Clark weighs in on burn-in

[gregorio]

Wicht sound is use for the test and how do they compare the transient ?

A number of sounds are used for testing, depending on what we are testing for. White or Pink Noise is used, band limited noise and multi-tone signals but probably the most commonly used are simple, single sine waves. We also use actual music recordings sometimes.

In the specific case of transients, we can use extreme test signals (such as a Dirac pulse for example) or again, actual sounds/music. We can compare transients with a variety of methods, for example by applying a Fourier Transform (see next paragraph) or using a null test.

I understand that there are harmonics and combination pretty complexe to analyse.

That depends on what you mean by “complex to analyse”. It’s complex in the sense that you can’t do the math “in your head” but it’s relatively simple in the sense that it’s accomplished with old, proven mathematics (discovered by Joseph Fourier 200 years ago) and is easy for computers/chips to calculate.

Is it really possible to mesure before and after burn in (or cable change or whatever)

If there is actually a difference then “yes” it really is possible to measure it and has been for many decades (for example, the CRT oscilloscope was invented in 1897, used in laboratory testing by the 1920’s and was in wide use for measuring audio performance during the 1960’s). Of course though, it’s not possible to measure a difference if there isn’t a difference!

and be able to find difference in the range of human ears capacity ?

Again, for many decades audio measurement devices have been capable of far more accuracy and far greater sensitivity than human hearing (by orders of magnitude) and within a range that far exceeds human hearing (again, by orders of magnitude), so if there were a difference, we would be able to find it.

G

 

[bigshot]

I read that the first millisecond of the notes are importants for the ears

Samples in a CD are 25 microseconds apart, so it should be perfectly capable of properly reproducing one millisecond. The importance of the initial transient depends on the instrument I suppose. A snare drum would have a quicker initial transient than a flute. CD quality sound can capture any of those sounds and reproduce them as well as the ear can hear them.

 

[Roland de Gilead]

A number of sounds are used for testing, depending on what we are testing for. White or Pink Noise is used, band limited noise and multi-tone signals but probably the most commonly used are simple, single sine waves. We also use actual music recordings sometimes.

In the specific case of transients, we can use extreme test signals (such as a Dirac pulse for example) or again, actual sounds/music. We can compare transients with a variety of methods, for example by applying a Fourier Transform (see next paragraph) or using a null test.

That depends on what you mean by “complex to analyse”. It’s complex in the sense that you can’t do the math “in your head” but it’s relatively simple in the sense that it’s accomplished with old, proven mathematics (discovered by Joseph Fourier 200 years ago) and is easy for computers/chips to calculate.

If there is actually a difference then “yes” it really is possible to measure it and has been for many decades (for example, the CRT oscilloscope was invented in 1897, used in laboratory testing by the 1920’s and was in wide use for measuring audio performance during the 1960’s). Of course though, it’s not possible to measure a difference if there isn’t a difference!

Again, for many decades audio measurement devices have been capable of far more accuracy and far greater sensitivity than human hearing (by orders of magnitude) and within a range that far exceeds human hearing (again, by orders of magnitude), so if there were a difference, we would be able to find it.

G

i understand that we have the answer that burn in and cables doesn’t change the sound by mesuring for decade. So why is there still a debat ?

 

[gregorio]

So why is there still a debat ?

Good question! In fact we don’t have this debate in the pro-audio world, telecoms world or other fields. The only place where this debate still exists is in the audiophile community and that’s because audiophiles are generally not well educated about signal transmission/transfer or other factual details and therefore are open to suggestion and easily fooled by false marketing, sponsored reviews and cherry picked testimonials. And there’s a very big incentive for manufacturers to do this particularly with cables, because they’re cheap to design, manufacture and to ship but audiophiles will happily pay hundreds or even thousands of dollars for them!

G

 

[bigshot]

What was it that P.T.Barnum said?

 

[jagwap]

Good question! In fact we don’t have this debate in the pro-audio world, telecoms world or other fields. The only place where this debate still exists is in the audiophile community and that’s because audiophiles are generally not well educated about signal transmission/transfer or other factual details and therefore are open to suggestion and easily fooled by false marketing, sponsored reviews and cherry picked testimonials. And there’s a very big incentive for manufacturers to do this particularly with cables, because they’re cheap to design, manufacture and to ship but audiophiles will happily pay hundreds or even thousands of dollars for them!

G

As usual you are ignoring the report (by Klipple, the top driver and speaker measuring equipment) I put forward before that burn in on speakers is measurable, and is recognised in the pro audio industry as a phenomenon that needs to be taken into account.

So the debate continues, because some people will not take all the evidence into account.

Edit: apologies, the previous reply by Rolland also mentioned burn in, so I thought you were including this subject, and also it is the original subject. Agreed. There is zero evidence cables need burn in I have seen or experienced in my career.

 

[bigshot]

I would imagine that burn in for speakers would affect large woofers the most. Is that correct? Does the amount of excursion correspond to the amount of burn in?

 

[castleofargh]

As usual you are ignoring the report (by Klipple, the top driver and speaker measuring equipment) I put forward before that burn in on speakers is measurable, and is recognised in the pro audio industry as a phenomenon that needs to be taken into account.

So the debate continues, because some people will not take all the evidence into account.

He was talking about cable burn in.

 

[sander99]

Cable burn in or cable burn out?

 

[Dogmatrix]

I would imagine that burn in for speakers would affect large woofers the most. Is that correct? Does the amount of excursion correspond to the amount of burn in?

That is not my experience, I would not call it burn in really since that carries associations with positive effects. What happens in practice is drivers change over time, all of them.
Magnitude was always greater in returns from hot humid environments, teardowns showed corrosion on coils, perishing surrounds and sometimes split spiders
I guess since the smaller drivers are doing more cycles things even out
We did not consider it a positive thing since the speakers were designed and tested around "new" drivers any change lowered performance
Even from the harshest environments 10 years was a reasonable average
I am talking about extremes since speakers did not come back until they had malfunctioned
We once had a pair of large array type come in from a pig farmer, he had been running them on his porch so he could hear the music in the pig shed.
Unfortunately he ran them so hard the resistors heated up enough to melt the caps and if not for the fire-retardant foam they would have ignited
When I measured the drivers, they had not changed so we just fitted new xovers

 

[bigshot]

That sounds like burn out, not burn in. I would assume that burning in would be like breaking in shoes to an optimal state, not degradation. If we're talking about speakers wearing out over many years, I agree that happens, and the only difference of opinion is the choice of the term "burn in" to describe it.

 

[Dogmatrix]

That sounds like burn out, not burn in. I would assume that burning in would be like breaking in shoes to an optimal state, not degradation. If we're talking about speakers wearing out over many years, I agree that happens, and the only difference of opinion is the choice of the term "burn in" to describe it.

No difference of opinion, they are separate things
It was before my time but the concept of burn in or breaking in had been explored, the engineer said the drivers were considered stable out of the box and changes brought by running them for the amounts of time practical in a factory were not significant. So, it was concluded there were no improvements in performance or quality sufficient to warrant the production time.
Speakers that came in for service due to accidental damage like the kids poking the cones generally showed no significant change in the surviving drivers, so a straight swap from stores was possible without compromising performance.
Our tolerance on the flagship was plus or minus 2db between 27hz and 20khz
So, I would say from my experience that burn in as a short-term improvement is a myth

 

[Roland de Gilead]

No difference of opinion, they are separate things
It was before my time but the concept of burn in or breaking in had been explored, the engineer said the drivers were considered stable out of the box and changes brought by running them for the amounts of time practical in a factory were not significant. So, it was concluded there were no improvements in performance or quality sufficient to warrant the production time.
Speakers that came in for service due to accidental damage like the kids poking the cones generally showed no significant change in the surviving drivers, so a straight swap from stores was possible without compromising performance.
Our tolerance on the flagship was plus or minus 2db between 27hz and 20khz
So, I would say from my experience that burn in as a short-term improvement is a myth

Thanks for your usefull explication. What i still want to understand is if 2db is audible ?

Twice i made mistake when i reconnect some cable (Phase or mistake in the cable) and i heard a really thin changement that upset me more and more until i realise that i have made a mistake. The difference were so subtil that no one in my home realised it. That kind of range seems so hard to distinguish that i was sure that it would be very hard to see in a graphic. Now i understand that all of this have been mesure for decade and that the modification eared is a human’s mistake by biais.

So, if i hear a difference in a burn in or a cable or whatever snake oil, it’s by mistake even if i think i fell it and think to be able to recognize most of the time the origin.

Hopefully, quantic physic doesn’t apply because it’s kind of weird (funny thing i wrote wired first instead of weird).

 

[bigshot]

2dB is audible with tones. In music, it depends on where in the response the 2dB lies, but it isn’t likely to make much of a difference in practice. 2dB manufacturing tolerances for speakers is VERY good.

Unless you level match and do a direct A/B comparison, it’s much more likely that any subtle differences you hear are due to perceptual error.

 

[Northern]

As usual you are ignoring the report (by Klipple, the top driver and speaker measuring equipment) I put forward before that burn in on speakers is measurable, and is recognised in the pro audio industry as a phenomenon that needs to be taken into account.

It's Klippel, not Klipple.