Sunday, January 27, 2013

The effect of break-in: Vsonic VC02


Introduction


It is generally known that a new pair of headphones must go through a "warming up(aka break-in)" phase for certain period of time before it opens up its true sonic potential. Previously, both electroacoustic and psychoacoustic aspects of a 100 hours-long headphone break-in have been presented with SONY MDR-EX1000Sennheiser HD650, and Fostex T50RP; The physical transformation is evident, yet its degree is not of a night and day difference, as audiophiles normally describe it.



On Vsonic VC02's cable, a tag, which is written in Chinese, is attached by the manufacturer, openly claiming the dramatic sonic change. The manufacturer states 100 hours of break-in improves the sound quality of VC02.


Test methods

A brand-new Vsonic VC02 is to be broken-in for 100 hours straight with XLO's break-in sample (100 dB SPL @ peak), driven with an Objective 2 headphone amplifier. Once the headphone is *placed on the occluded ear simulator, its physical placement must remain untouched for the next 100 hours in order to prevent any placement-related deviations. Each 10 times averaged pre break-in & post break-in measurement data are to be compared, and should any type of change occur, they are to be reproduced back utilizing a binaural recording technique to be ABX-compared.



Objective Assessment

Impedance
Overall impedance is slightly decreased after the break-in. This is the similar behavior observed from previous break-in experiments.

Frequency response
The resonances in the mid frequency range shifted. Interestingly, this resonance shift has been reverted once the voicecoil cools down.

Phase response 

Harmonic distortion
Although its degree is rather minimal, the amount of lower frequency distortion has been lowered. Again, such distortion decrement has been seen from previous experiments as well.

Time-domain characteristics 

Impulse

CSD



ETC: pre-break in / post-break in

Group delay


Subjective Assessment

ATTENTION: In order to accurately reproduce these binaural recordings, listeners must use a flat diffuse-field equalized headphone, such as Etymotic Research ER-4B and STAX Lambda PRO with ED-1. The accuracy of the reproduced result can not be guaranteed otherwise. The reference sample is an excerpt from the free hi-resolution sample provided by 2L - the Nordic sound. Downloading below samples is highly recommended for critical listening experience.

Reference sample @ 24-bit / 96 kHz:

A brand-new Vsonic VC02 @ 96 kHz:

A 100 hour broken in Vsonic VC02 @ 96 kHz:



Conclusion

As previously confirmed by others, 1 2 3 4 5 6 , and by myself, 1 2 3 , the physical effect of break-in is quite evident. Moreover, the psychoacoustic aspect of Vsonic VC02's break-in has been demonstrated for ABX comparison so that listeners could verify the audibility of 100 hours of what-so-called 'warming-up'. 

52 comments:

  1. How many hours did you wait after the 100 hours break-in? You wrote the resonance shift has been reverted after the voice coils cooled down, the other stuff didn't?

    Anyway, I do not see an improvement (like many audiophiles claim) after break-in at all. The changes are so tiny, you'd be hard pressed to tell a difference even if you had two exactly identical headphones - one new and one broken-in.

    The big changes these audiophiles talk about seem to be imagined. Getting used to a sound signature causes 1000x bigger changes than thousands of hours of break-in.

    ReplyDelete
    Replies
    1. It usually takes ~4 hours before comparable results can roll in. FR came back for sure. Other parameters did too, but not as much. Above comparison is to demonstrate the most dramatic instances, so that the audibility of the break-in can be transparently heard.

      Delete
    2. So the measurements don't have much to do with break-in, but with how the drivers characteristics change after hours of stress and heating up the voice coil.

      There's no point in calling it burn- or break-in effects if the changes are not permanent.

      Because you could have also used square waves played at high SPL to heat up the voice coils within a few minutes and quickly made the measurements after that.

      I know you wanted to help with this post, and I appreciate your hard work, but this one was kinda counter-productive.

      Delete
    3. "FR came back for sure. Other parameters did too, but not as much."

      Hopefully I don't have to post the post 4 hour analysis data as well, and compare with the above data. :D

      Delete
    4. Have you also done 10 impedance measurements? One after another or evenly distributed over the 4 hours? If the measurements are spread over 4 hours we should be able to see a trend in the impedance due to cooling down.

      Delete
  2. Another thing is auditory memory. You cannot compare the slightly changed signature after burn-in with the signature you heard a hundred hours ago. Yet, reviews regularly contain seemingly highly detailed information of what changed during burn-in. I call BS on that.

    ReplyDelete
    Replies
    1. The thing is, we have little changes, here, little changes there, little changes everywhere. Although it can be argued that most of these changes on their own won't make the difference, but what happens when they are added together? Will that make a large enough difference? The impedance response changes that range from 0 to .1 ohms that isn't a linear change (IEMs becoming a little more efficient) along with harmonic distortion across the frequency. There is the slightest change in frequency, it's tiny, but still adds in. You can also see changes in decay as well as some major things in group decay.

      Many of these are small in magnitude, but what's not to say together they don't form a large change together? We always seem to be dealing with the changes on singular properties (can you hear the frequency change? Can you hear the impedance change? Can you hear the phase change? Etc). However, the way a headphone sounds depends on all of these things together. So I ask the question, when we add the frequency, impedance, phase, decay, delay, etc changes together, can we hear it?

      I want to use a fingerprint as a loose analogy. If we can match 1 small portion of a print to a set of full ones, we still don't which print we are looking at. However, as we add more portions into it that match up, we can slowly begin to distinguish. Eventually, we hit a point where we know which fingerprint we are looking at. However, each of the individual points we matched won't tell us that, it's the sum of the smaller points.

      Similarly, we might not be able to hear the frequency differences (alone), or impedance, or decay. Add them together though, we might be able to, it might be bigger than you feel it is as well.

      Obviously I'm talking theory here, but there is reasoning behind it.

      Delete
    2. The OP talks about common sense in electroacoustics/psychoacoustics, but my friend, tinyman also talks about the practical issue arising when driving a headphone in real life, which totally makes sense.

      Then again, if the playback chain's quality is transparent(Hi-Fidelity) enough, such practical variables can be opted out. Still, I wonder how many regular listeners would have such accurate devices at their disposal? Are reviewers all day-dreaming, or are objective parameters rather fooling our perception?

      Delete
    3. @tinyman
      Problem is that they're not just small changes, they're ridiculously tiny, most non-existent in practical terms. CSD [which is merely shifted], impedance and resonant change were just due to the heating of the coil, they go back to normal after cooling down, thus no change in practical terms and even then, tiny. The THD and FR change is ridiculously small, even put together, the changes just don't add up to anything noteworthy even if crumbed together. Biases occur when breaking-in a IEM based on what is believed and auditory memory is horrible for direct comparisons. Agree with anonymous here.

      Delete
    4. @Inks, I need a source that these changes won't be audible since they are individually small... Small things added up get quite large mind you. Even small changes can have drastic results in many ways.

      Now, this isn't entirely audio, but I do want to use a math example as an analogy of how a small change in data can lead to extraordinary change in the final results. So, lets take the roots of two similar equations. First, x^2 - 4x + 2. There is one root, x = 2. Now, take the similarly close equation, x^2 - 4.0001x + 2, we wouldn't expect the roots to wander too far from 2. However, the solution to this equation show that the roots change drastically, x = .5858, 3.4144. In this second case, we changed the 4 by 0.0025% to create an approximated 70.7% (in both instances) change in the final result.

      Now, this math may have little to do with audio and perceived audibility. However, it is a good analogy to what I am trying to say. Don't rule out a set of tiny changes (set = 1 or more), unless you can show the sum won't make a difference.

      My question still stands Inks, does the addition of the FR, TDH, Decay, Group Delay changes cause an audible change in sound? This question was meant to be much more rhetorical as no one has enough information to answer this question (not even myself) as it hasn't been tested (or tested heavily enough) fully. You can speculate all you want, but it'll remain just that, a speculation.

      Delete
    5. if you need a source that a change of less than 1db is small, then I think we can end the discussion there, plenty of established readings will confirm this. Or an increment in THD that is still far below the established audible treshold, no speculation there. The analogy doesn't work for this case.

      Delete
    6. Now, I'm not talking about this graph specifically, but a 1 dB bump in a wide enough spectrum in the right place can be audible. You really do need to stop making assumptions Inks, you have little to no evidence to what you say at this point, so yes, we are done. For some odd reason, you still believe that human hearing is both linear and stable, that may not be the case.

      Delete
    7. @tinyman: Frequency response is easily the most important characteristics of how a speaker sounds (assuming distortions are low enough, which they are in this case), and FR is the thing that reverted after the voice coil cooled down.

      Impulse response, CSD, group delay all are related to the FR, which I repeat, reverted to the initial curve.

      @udauda: Practical issues when using headphones in real-life cause much, much, much bigger changes. Just one example: placement.
      You already know this, but others maybe don't. If you take some measurements, re-insert the earphones and do another set of measurements you'll see big changes due to different placement.

      This is another reason why I don't trust subjective reviews. It's impossible to re-insert/place ear/headphones identically after a couple of hours of break-in, let alone remember the exact sound signature.

      I don't think reviewers are day-dreaming. They just don't seem to know anything about biases and how much the brain is involved in hearing.
      Simply listening a few minutes to a new headphone will cause some getting accustomed to the new sound signature. This alone will make the new headphone sound better on the second listen, regardless if they were broken-in a couple of hours or not.

      Delete
    8. @tinyman: "First, x^2 - 4x + 2. There is one root, x = 2. Now, take the similarly close equation, x^2 - 4.0001x + 2, we wouldn't expect the roots to wander too far from 2. However, the solution to this equation show that the roots change drastically, x = .5858, 3.4144. In this second case, we changed the 4 by 0.0025% to create an approximated 70.7% (in both instances) change in the final result."

      This is completely wrong.

      x^2 - 4x + 2 using x=2 we get: 4 - 8 + 2 = -2, not zero.

      The first root is 0.58578 and the first root in the second example is 0.58577.

      Delete
    9. You are right, guess I could've used some sleep. The roots for the equation should be x^2-4x+4 will be x=2. If we change the second number to 3.9999 we will induce imaginary numbers. In essence, I was using the instability of the quadratic equation as an analogy to whether or not changes in the measurements will ALWAYS be stable (what Inks seemed to imply). Methods to solve many problems are unstable in practice, how do we know this isn't.

      Delete
    10. So given impulse, CSD, and group delay, you can draw an accurate graph of the frequency every time? I'll agree that they are somewhat related, but won't agree that they are 1-1 transformations of each other.

      Delete
    11. Asked the question improperly. I wanted to ask them in both directions, not just one. So can you generate an accurate FR graph using the others, but can you generate the others from the given FR graph.

      The data above actually gives me enough information saying that if you can, the outcome is unstable. Small change in FR causes huge change in group delay.

      Delete
    12. @tinyman:
      Math: In the first case the imaginary part is zero, in the second it is close to zero.
      Unless you damage the driver during break-in you won't see any crazy changes.

      Relationships: No, all that is needed is the impulse response. They are not "somewhat related", the relationships are clearly defined.

      The CSD is a visualization method of the impulse response.

      The frequency response shows the magnitude of the frequency response gain and the phase response shows the frequency response phase shift - this information is also in the impulse response.
      There's a unique relationship between magnitude gain and phase in a minimum phase system, so yes, you can theoretically go both ways.

      Group delay is just the derivative of the phase response with respect to frequency.

      Besides sleep you may also need glasses (scnr), because the group delay is almost identical. There's a red (post break-in) line on top of a hardly visible green line (pre break-in).

      There's no instability. The differences are tiny, despite using drivers with heated up voice coils. I guess with cooled down voice coils the tiny differences would almost vanish (there's always a small measurement error).

      Delete
    13. I'm sorry, but I don't respond to insults Mr. Anonymous. Good day to you.

      Delete
    14. I'm sorry if I hurt your feelings, but you cannot offer criticism if you cannot even read a graph. Granted, the image is tiny and fuzzy..

      Delete
    15. No worries, not exactly hurt feelings, I just don't respond to insults (normally means we both need to calm down; EG something got out of hand). You are right about the graph though.

      Regarding your relations being clearly defined, I will agree that there are clearly defined relationships. However, I feel that the relationships aren't each other, reason why I say somewhat rather than completely. When I say somewhat, I mean they have clear overlap, but they aren't completely overlapping each other. Two different opinions about the subject, I can live with that.

      There are reasons why I feel that the FR graph isn't 100% related to the rest. This is just from my observations, but I've seen multiple impulse responses, as well as square waves point to the same FR graph. Even in this, the spectral decay isn't the same, which means a given FR for the bass points to two types of decay (two types of decay relate to the same FR). This, of course is just my opinion on it.

      Delete
    16. @ tinyman
      These graphs are the matter at hand, why are you changing the topic or graph to discuss? Show me the 1db difference in a controlled burn-in test for IEMs? I have little evidence to show that increments WAYYY below the audible treshold of THD and a change of about 0.5 or less dbs being hardly noteworthy even with these 2 together? I'll let the readers decide, common sense. Because of human error in seal and insertion depth, I believe many simply attribute these factors as burn-in. Even then the changes REVERTED. This is simply yet more proof that burn-in in IEMs is not something to consider and more proof that many reviewers out there let their bias and ego manufacturer these kind of myths. Have you even tried the test udauda posted? tell me there's a big difference there...

      Delete
    17. Now you bring up the point of time, that I'll respect. 1 dB over a set range of time may not be audible, and yes, should be below the threshold. As for me bringing up other graphs, it was in response to the statement that all the graphs are 100% related to each other. Anonymous made his statements sound like he can derive an exact FR graph from any other graph. My statement back is that he can't derive the entire graph, but definitely parts of it as well as others.

      As for your "proof" that it's the bias going to the myth and ego, I have to disagree. You are free to believe this, I won't though. I will wait until information comes out as what you are stating is untestable right now (ego and bias). Wait for more data to come, then make a decision is what I say.

      Delete
  3. I admire your work. Now I can believe the effect of break-in. However, it's to be noted that the audibility of difference between pre-break-in and post-break-in DO NOT imply the improvement of sound quality. I always worry about this confusion.

    Carrot

    ReplyDelete
    Replies
    1. Thanks for your kind words, Carrot! Absolutely. Although the sound quality may change due to break-in, but the effect may/may not be preferable.

      Delete
    2. Agree with both of you here. The data above shows objective ideas of break in. It doesn't answer whether or not the changes can even be heard (I don't feel it should either), the subjective realm. That said though, it's always been unknown whether or not break in will make a headphone better or worse.

      Delete
    3. It doesn't show break-in. It mainly shows the changes of a heated-up voice coil.(See the third reply to my first comment.

      Delete
    4. So using the data, there are two possibilities. The measurements were taken with cooled coils, or hot(ter) coils.

      If the impedance curves were taking after the coils cooled (which most likely means the rest were too). This would show that the impedance did indeed drop a little. It shows some sort of burn in. Still questionable whether or not its audible.

      But, if the measurements are taken with a heated voice coil, we have a few interesting bits of data with the impedance curve. The resistance goes up as the coil's temperature increases (in copper, silver, and aluminum). The data above, however, shows the resistance dropping in the impedance curves. Which means that there is something going on internally with the driver. When cooled, however, the resistance will drop further (cooler temperatures = lower resistance), thus showing burn in.

      In both cases, we see something has changed, physically, with some internal mechanisms. If we measured while cool, then we see an impedance drop. If we measured while hot, we see an impedance drop when it should have gone up.

      I do want to thank you for the post, if you didn't post it, I would never have rechecked the impedance curves to see they dropped.

      Sources: http://www.stereophile.com/reference/1106hot/index.html

      Delete
    5. You're confusing DC resistance with impedance. DC resistance increases with heat, of course, but the curves above don't go down to DC (0 Hz).

      There are a couple of factors that explain the impedance dropping with a heated up driver. The heat is not only in the voice coil, it spreads to everything that is connected/around it. The diaphragm is less stiff when heated (that's bad), reducing the movement of air. The heat also weakens the magnetic force.

      If done cold (after everything mentioned cooled down properly) I'd say that either udauda got the curves mixed up or that the diaphragm was damaged.

      Anyway, udauda writes "Overall impedance is slightly decreased after the break-in. This is the similar behavior observed from previous break-in experiments."
      I cannot observe this at all. Only his t50rp break-in measurements show a clearly lower impedance. Maybe he did that one immediately after stopping the break-in.

      Delete
    6. @Anonymous: I think I mentioned I measure 10 times, and average it. That accounts all the possibilities from the moment I stop break in to the post-4 hours point.

      Delete
    7. Regardless of DC or AC, heat should still increase the resistance. The equations for both DC resistance and AC impedance is V/I, it should still go down.

      As for everything connected around it, think for a second, what is this everything? It's the wiring really. Copper in this case (or silver). both have increased resistance with temperature. Weakening magnetic force is expected as electrical force in essence is magnetic (they are intertwined in one another.

      You do make a valid point about the driver diaphram being less stiff, but this shouldn't come up in an impedance reading. It should come up in the frequency response though if it reduces the movement of air, it would also reduce the pressure exerted by the air (pressure measured in dB).

      As for Rin switching the before and after graphs. The statements he's made about seeing it in the past makes me feel this claim doesn't have enough merit. It is possible he mixed up this graph, the T50s, and all the others, but unlikely. As Rin stated, he measures 10 times and averages it (thanks Rin for that information BTW, very useful).

      I still say that there is something weird going on with something electrical. Could be cabling, drivers, insulation loss? IDK, but something is going on :p We just don't know what. We don't even know if it'd be audible ;)

      Delete
    8. I doubt the cables would heat up. Maybe the coil wire between the cable and the voice coil heats up similarly to the voice coil itself, but that piece of wire is usually very short. So that doesn't seem to matter.

      On loudspeaker drivers you see this big rubbery suspension around the diaphragm. The diaphragm and suspension in headphone drivers is often made out of polymer material. If you move this back and forth constantly (break-in) there's a build up of stress. Polymers can relieve this stress, but until they do the impedance certainly does change. If Rin uses a proper headphone amp with low output impedance there won't be big changes in the SPL. And who knows, maybe he isn't using a fixed output level but configures the volume control to match a certain measured SPL.

      The voice coil is surrounding the permanent magnet. Permanent magnets don't like heat. Again, this also changes the impedance.

      The 10 averages are certainly a good thing, but even more care (documentation) should be taken when each measurement is made in relation to when the break-in stopped.

      Delete
    9. We actually didn't see many changes in SPL (graph) here. So the headphone amp's output impedance would lower something that is already very small :p

      If the heat did affect the magnet, it would also weaken it, right? Wouldn't that change SPL output (weaker magnet would result in softer sound?). That would show up in the FR graph. You are right though, if the magnet was effected, it should show up in impedance recordings as well.

      I do agree, information regarding how long after break in recordings were made would be great ;) Other information you mentioned as well (how he configured the output level and volumes).

      Delete
  4. I think the changes are small enough to say that every time you insert the IEMs into your ear, the slight pressure & insertion depth differences might as well account for the change in data. (or difference from burn in when perceived subjectively)

    If the IEMs inserted pre-burn in was measured at a relatively high canal pressure, and that pressure slowly leaked over time of "burn-in", maybe that could account for the measured changes?

    --
    I recall Tyll's article at Innerfidelity suggested that IMD improved with break-in, but the site's not working at the moment.

    http://www.innerfidelity.com/content/measurement-and-audibility-headphone-break
    I think this is the article.

    ReplyDelete
    Replies
    1. The site works, just not too well on Chrome right now :p Works fine on Firefox, Safari, and Mobile Safari, just not Chrome ;)

      You also have to remember, we have a huge amounts of variables here. We don't know what the cause of change is, we just know there is some change somewhere.

      As you've stated, insertion depth and pressure. These would be human error, we can sort of account for that using multiple measurements pre-burn in and multiple post to get an average (IDK if this is what Rin does). If it is, the pressure and insertion should cancel itself out in the average eventually after enough trials are done.

      But let's go back to the idea that we don't know what breaks in. I'm not sure if anyone notices this, but as tips are used more, they deform in shape. The deformation actually causes the tip to countour to your ear more. So you are right that pressure might go away as time goes. Could this have a similar effect as pads breaking in on a pair of full sized headphones with time, the headband stretching out? It may, it may not.

      As the raw data shows, we see changes here and there, small frequency response changes (if any), distortion changes as well as impedance. Whether or not this is audible is questionable, and up for debate. A second thing that also comes up for debate is what causes this change.

      Many people want to really point at one variable and say "it's this". Some say it's frequency graph only, others say pressure and positioning only, etc. I say everything impacts in some way or another. It goes deeper than that single variable you want to single out. We have a huge multi-variable set of sets here with complex relationships to each other. We are just chipping away at the surface right now, there is much more that lies below that is to be found. If it was found, we might get an answer... Might.

      Delete
    2. All of these non driver break-in related variables and the documented loudspeaker burn-in experiments make headphone burn-in, like audiophiles claim, highly implausible. The driver's are tiny, the diaphragm hardly moves at sane SPLs. Even if we found out that the driver's characteristics changed (after cooling down), they'd have to be tiny and would be worls, no universes, apart from what reviewers claim.

      Delete
    3. You make very true claims regarding the loudspeaker to IEM, Anonymous. I still can't believe the last claim though, I do respect your opinion, but as of right now, evidence doesn't suggest either way.

      At the end of the day, a dB is still a dB, and referenced from the same general area (your ear) as it is in the measurements. You just won't feel the bass in your chest with an IEM ;)

      Delete
    4. All the measurements (evidence) I've seen show that the differences are tiny. They're even smaller if the driver gets some rest after break-in.

      Delete
  5. Thing is, reviewers hardly make claims such as. "I feel there's been a slightly change in a frequency region or in transients". No, they claim things such as, "bass got tighter and more refined, treble is smoother, etc." The data is FARRR from these claims. Do the test udauda posted, tell me it makes a difference

    ReplyDelete
    Replies
    1. I don't have the Etymotic ER4 to test with unfortunately. I can test with others, and probably will. As for tighter bass, all you really do is have to reduce some loudness in the upper-bass, maybe lower bass. I still have no clue what they mean by more refined :p Smoother treble could mean that a small spike they heard in the treble went down a little, distortion went down or something like that.

      I do agree though, Inks, that many of these claims can be overstated at times. It could deal with the idea that a gain can also be felt as a loss in other areas. For example, they can say that the clarity and detailing came out a little more while the bass and treble got more controlled. If you focus on the individual areas, they aren't lying if all that happened was a midrange boost (midbass boost may have cause the bass to sound tightened or treble softer).

      At times though, I do agree Inks, we see these statements, but the graphs return no data to show that what they said happened, happened. At that point, the question becomes, why? How can we reproduce what they are hearing (thing with a theory is, it needs to be reproducible, or it remains inconclusive)?

      I will do those tests when I have time Inks. Maybe tonight or tomorrow :)

      Delete
    2. Just use the flattest IEM you have.

      Yup, but those bass changes are mostly due to seal IMO, I have yet to see it in a burn-in test. I have no clue to what's more refined either, just quoting burn-in believers. Again, I haven't seen a resonant spike get reduced either, the VC02's resonant spike remained intact, likely other tests will prove this. You are assuming some small changes that may be notable, like a 1db change, but tests haven't shown such changes, it has shown ridiculously small difference that even happened to revert back. Human error in seal, insertion depth and overall bias is what changes an IEM.

      Delete
    3. (also in response to your reply just above)

      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.

      The interesting thing is that the listeners said that the "better" cable sounded more "dynamic", "clearer", "smoother" ...

      Does that sound familiar?

      Delete
    4. @Inks: I don't see the treble spike going down either anywhere. Even the decay isn't substantial enough to do it. So I have no clue where they are getting that from. If anything, the spike should be more apparent seeing that the bass decay quickens.

      As for the 1 dB changes, it's audible if done instantly (EG, in EQ). Anonymous' data seems to show they can hear 1 dB as well, but probably not over a time span (a point you made earlier).

      @Anonymous: Thanks for the information, appreciate it. I've never been able to hear cable differences, but Rin does have some graphs of the Heir Audio with different cables. It does look to change sound if I remember correctly (I think he tested the Mangus as well as another no-name cable).

      As for their descriptors of the "improved" cable, it's plausible if the 1 dB wasn't uniform that they heard a hint more clarity (especially in the area where our ears are sensitive). My question to that is, what areas ended being boosted by 1 dB, did it effect decay in any way?

      I will admit this though, it does sound familiar ;)

      Delete
    5. @Inks (again, sorry :(): I will have to wait until I get home this weekend to test the tracks then :) I should have time though during the weekend (I'll test with my UE 900, HF5, and PFE022 with grey filters).

      Delete
    6. The 1 dB boost was over the whole frequency range, no EQ, just a tad higher volume. (If the boost is done over a smaller range of frequencies it becomes a lot harder up to impossible to hear a difference.)

      The main point is that no-one said: "oh, with this cable the system just sounds a bit louder". They all used their favorite terms for trying to describe what the "better" cable sounds like.

      Delete
    7. LOL, now we're talking something going on creating the anomaly. Could be in their head, but at times people seem to think louder is better.

      Delete
  6. Hey Rin,

    I've wanted to say this for a little while, but never really figured out where to say it, so I'll do it here. That said, it is off-topic. Well this is actually two things.

    Firstly, I do want to thank you for all the measurements you are taking, from everything to burn in data to just headphone measurements. They are very useful and helpful for lots of people. They also uncover some mysterious things ;) So thanks for that.

    The second is more of a suggestion (layout-wise). I don't know if the blogging software you have supports something similar to a "more" link, where it cuts the blog post short with a "read more" link to see the whole thing. It might make the root of your blog a little less congested and easier to navigate through the articles. Just a suggestion.

    -Tinyman

    ReplyDelete
  7. You can't compare the cable difference of loudspeaker and BA IEMs, completely different. Cables in BA IEMs, no doubt change sound but consider the BA's nature of impedance changes and impedance differences between cables.

    ReplyDelete
    Replies
    1. Ah, gotcha'. Thanks for the clarifications Inks, it does make sense.

      Delete
    2. Some speakers definitely do have a very peaky (ranging from below 4 to far beyond 20 ohms) impedance curves. The difference is that some stock in-ear cables are very thin because they cannot be thick and heavy. This means higher resistance, which in turn means higher output impedance. This causes small changes in the frequency response.

      Delete
  8. Confirms what I believe: no burn-in physically, though many Head-Fiers would disagree. Rin, are you going to attempt to modify it to further better the response or show the effect of different ear sleeves for this IEM? Or is this second part as far it goes for the analysis of the VC02?

    ReplyDelete
    Replies
    1. Well, I have all the data for sleeve analysis for sure. So that will be up sooner or later. After that, I can't be sure whether I can come back to VC02 or not due to other things on my to-do list. BTW I am doing a funky modification on MH1- STAY TUNED...

      Delete