Resolve's B&K 5128 Headphone Target - you can try the EQ's.....

[amirm]

Harman has a 5128, as far as I know (maybe things have changed) there were no plans to validate the 5128 vs their GRAS research so it would appear as that research may not (or may, hopefully in the future) be coming from them.

Sean's team got one of the first 5128s produced. In other words, they have had it for good number of years. An attempt was made to calibrate against it as measurements using it against the Harman target showed wildly different results for the same headphone. They took a shot at creating a target for 5128 (I think) but chose to keep it confidential from what I recall. Their new target lowered the variation to acceptable range. I linked to the thread where we discussed all of this earlier.

 

[Robbo99999]

Sean's team got one of the first 5128s produced. In other words, they have had it for good number of years. An attempt was made to calibrate against it as measurements using it against the Harman target showed wildly different results for the same headphone. They took a shot at creating a target for 5128 (I think) but chose to keep it confidential from what I recall. Their new target lowered the variation to acceptable range. I linked to the thread where we discussed all of this earlier.

(I think they tried doing a shortcut didn't they. They took something like 10 different headphone models, measured them on GRAS, then measured them on B&K, then created a conversion curve from GRAS to B&K, then applied that conversion curve to the Harman Target. They didn't ever start from scratch with the B&K though.)

 

[solderdude]

Sean's team got one of the first 5128s produced. In other words, they have had it for good number of years. An attempt was made to calibrate against it as measurements using it against the Harman target showed wildly different results for the same headphone. They took a shot at creating a target for 5128 (I think) but chose to keep it confidential from what I recall. Their new target lowered the variation to acceptable range. I linked to the thread where we discussed all of this earlier.

A recap with quotes from Dr. Olive:

The 5128 is most different from the IEC 711 couplers and there is a new standard to include it. Beside the HF differences which could be pinnae related is the low frequency response. Some of this could be related to the larger volume of the 5128 coupler. The other difference is how the headphone seals to the shape of the head/cheek which always creates a greater leak than the flat plate of the GRAS 45 CA. We see similar leaks with the KEMAR head compare to the GRAS45CA. The question is which fixture best simulates average headphone leakage measured on humans?

That difference in leakage is what seems to create the difference and thus will differ with headphones and can range from hardly a difference to a substantial difference in the 20-500Hz range.
The question indeed is... what is the more accurate seal representation and will that change per head and over time (pads forming to headshape/softening)
Fascinating but may just have to 'live with the differences'.
Remains an open question... driver/headphone performance in the lows thus will always remain an issue with 5128 as it can not be compensated for.
On FP I have also seen differences in seal caused by stiff pads or pads that do not have a uniform shape (height) or low clamping force or deliberately 'human contour' shaped pads.
So ... incorrect measurements are possible with both types of fixtures but is more likely to be 'closer to perfect for repeatable measurements' with FP instead of 'human shaped' manikin.
Manikin might be more representative for a lot of cases.
There seems to be no way around it other than to measure on both and compare (gauges seal issues in the process a bit).

2 years ago Dr. Olive wrote this:

One more thing: My intention of this presentation was not to disparage the B&K 5128, and I don't think this presentation did that.

We own a 5128 and I hope to use it more in our research. The problem up until now was our research was based on a different test fixtures, and we needed some way to be able to better interpret headphone measurements made on the 5128. That was the motivation of the research.

Has more research been done in the meantime or was it left at that or did something more substantial about usability for headphone measurement research come out of this ?


Regarding Stealth measurements... were the Stealths used the exact same ones ? Is it also possible that product variations might be in play ? Could it be positioning or averaging ? Afterall they are just 2 plots... how do we know they do not sound somewhat different (talking about the 3-6kHz range) and not the bass/lower mids.
We do know Stealths are very seal sensitive and especially between 4kHz and 12kHz are very position sensitive (20dB !).
2 single headphone plots should never be the reason to discard one or the other method but is intriguing as to what conditions were present. Without thorough investigation of just 1 headphone we don't really know anything for certain other than the conditions, smoothing and headphone are different.

I have that data. We made 5 reseats for each of the 20 headphones we measured using different amount of pressure on the cups. So a total of 20 headphones x 5 reseats x 3 pressures x 6 fixtures =1800 measurements

Would have been interesting if Dr. Olive would have had a Stealth in that 20 headphones pool and using those differences in a presentation as well. Now it is not known exactly what is the cause of the displayed plot differences could perhaps be fixture related only but more likely has a different cause.

I wonder if the GRAS was not used for all the Harman research but only the 5128 how that research would have ended up with or that the efforts would have been halted or maybe even the pinna and coupler were used on a FP to counter the seal issues. What would it have looked like then. We will never know I reckon.

Good that some 'amateurs' are looking into this and are willing to share results and discuss openly.

 

[Deleted member 16543]

I have validated my measurements through listening tests across nearly 100 headphones. Every time I measure a headphone I go through a process of creating an EQ and seeing if the correction relative to target results in better sound. Often I do this testing blind. As long as your expectations are that the target is a guide, then you come out fine. Many people have tried my EQ with similar results, sans minor adjustments.

When you have this volume of testing and public scrutiny for whatever is being cooked up for 5128, then we can talk. Until then, I would stay in the corner and not take shots like this.

That was a joke. But the point is that your method is still not Harman level research, which is what you've been demanding for the 5128.
So now you seem to say (and I agree) that we can lower the research expectation bar, because you think your way of doing things is still good enough (which it is).
So why shouldn't Resolve and people like him, a priori, be allowed the benefit of the doubt of being able to come up with their own good enough research?
Remember, the 5128 being a more accurate tool would probably allow them even more wiggle room than you have in using a less than perfect methodology.. which is what you've done, and even Harman!
All research in this field has been conducted with less than precise methods, at some point. Why require perfection just for the 5128?
For what it's worth, I think that finding a target for the 5128 the way I said (using balanced near field speakers), correcting headphones according to that and see what the internet thinks is a better starting point than Harman.
The whole rating of uncorrected headphones is the only thing I would find complex to perform, but those guys are smart and I'm sure they'll figure it out.
So I would suggest you stay in the corner too, and give them the proper time to share results when they think they're ready.
I myself am excited to see what they can come up with.

But I also want to spell out that your decision of using the GRAS at this time is still the most sensible one, or at least the most reliable, as it provides the best possible information for your readers (at this time).
I think the real reason for contention between you and people like me and @Mad_Econimist and @MayaTlab (et all) is that we have an eye for the potential of reaching a point closer to the Truth, while you're keeping your eyes on the ball of providing the best possible right now.
Both approaches are valid, and yours is probably the best for the function of this site.
Just.. don't rule out that in the future there will be better research to correlate measurements and listeners preference.
It's almost guaranteed it will happen, if you think about it.

 

[amirm]

That was a joke. But the point is that your method is still not Harman level research, which is what you've been demanding for the 5128.

Looks like it wasn't a joke after all... My fixture is only slightly different than Harman's. 5128 has no resemblance to GRAS.

vs

All research in this field has been conducted with less than precise methods, at some point. Why require perfection just for the 5128?

Perfection? Research? What are you talking about? There is no such effort being applied to 5128. Someone has thrown out a target curve and is asking the Internet population to vote on it in a totally ad-hoc and random methodology. Come back when you do have research with a body of headphones, controlled tests, etc. and then you have something. Until then, it is just a lot of talk with no substance.

 

[amirm]

I think the real reason for contention between you and people like me and @Mad_Econimist and @MayaTlab (et all) is that we have an eye for the potential of reaching a point closer to the Truth, while you're keeping your eyes on the ball of providing the best possible right now.

Truth? Anyone who uses this term has no idea of the complexity of the problem. Remember, I had Mad_Economist create a target for me for the 5128 I evaluated resulting in nothing useful. I was all set to buy the 5128, and put my faith in his theories only to fall flat on my face for basic reason: that without a properly researched target, the 5128 is a tool looking for a problem to solve. Compounding the problem was difficulty of mounting either IEMs or headphones on it.

I read a long time ago that companies that make dog toys aim to please the dog owners, and not dogs. This is why they make them look like hotdogs, bacon, etc. which has no meaning to a dog but a lot to the owner in the shopping isle. The 5128 seems like that to some people it seems. It appeals to their nerdy side that this thing must solve this problem better. They have gone ahead and purchased (?) this expensive tool with no proof point of it.

So yes, you do have to deliver the goods when you are arguing with someone who has used 5128 and knows its limitations. Pleading to accept your declaration is not going to work.

 

[MayaTlab]

Hi, I had another look at this just now this morning, and I'm still finding it confusing. Instead, could you just summarise into a couple of sentences the points you're making, as an outline overview if you will. This way I can probably take what you say and then be able to look at all your detailed graphs with some kind of perspective. If you can, try to avoid jargon such as "error curves". I'm not a laymen, but try to explain it in laymen's terms (or "ASR laymen's terms"), (which should help more than just myself).....I'll then take a detailed look at all your information you've put forward. As I understand it, the current purpose of our conversation is you proposing that only certain headphones should be used on the 5128 during the Target Creation & Listening test process - for finetuning the target curve.

I can't make it short but I'll try a different approach.

TLDR : think about listeners' preferences and EQ profiles. Why would they change for headphones for which we know that the listeners will have a reasonably consistent experience between them and between fixtures and humans, just because we'd use another test fixture to measure these headphones with ?

The longer version :

I'll still use the data from the HBK conference as it's convenient.

Forget about targets and the 5128 for now. Let's imagine you have a listening panel, give them a pair of HD800, and ask them to EQ them until they sound good to them.

Now, this is fictional as it can't really happen in practice with this method, but let's imagine that the EQ they, on average, come up with, is this (in that example it brings the HD800 - still using the HBK conference data ! - to Harman. It could be any other result for the argument's sake - it's just that the whole point of Harman's years of research is that it's quite a safe bet that it would be close to Harman) :

This EQ profile is what this listening panel preferred. logically, these people's preferences, for the HD800, wouldn't suddenly, for some strange reason, change right after this pair has been measured on a 5128 afterwards, would it ? So, in a certain way, the target they'd preferred, when using a 5128, is basically how these HD800 + the EQ profile would measure on a 5128. That's it .

The question you'd then have is : but wait, that will only work for the HD800 ! And you'd be right for high frequencies for certain.

The good news is that quite a few large, open over-ears not only are quite consistent across individuals, but also tend to produce a fairly similar difference between test equipments and between test equipments and real humans, clearly so up to 1kHz, and if you accept quite larger tolerances, generally so up to a few kHz. This is the lovely aspect of headphone designs that are insensitive to coupling issues.

This is quite evident with the HBK's conference data where the HD650, HD800 and Utopia produced a similar difference :

In this data set I believe that there are only four large, open over-ears. The K701, interestingly, doesn't behave like the others. I've never seen for these the sort of tests I like to see performed to get an idea whether a pair of headphones is more or less sensitive to coupling issues, for now that remains unexplained to me, even more so when we know that it performs consistently across individuals.

You'll also see the same trend from Rtings' data when they compared their own HMS with the 5128 :

I've also highlighted some open over-ears that don't follow that trend as an illustration that this isn't because it's large and sold as "open" that they'll systematically align well.

Generally the story is the same in other comparative scenarios (rig to human, different pinnae, etc.) : not all but most large, open over-ears tend to produce the same difference up to 1kHz, and tend to follow the average trend line reasonably well up to a few kHz if you allow for more inexactitudes. More specifically I'd single out some of Sennheiser's open over-ears, Focal's open headphones, and some of Hifiman's planars.

This would even more so be the case if you throw headphones with a (effective !) feedback mechanism in the mix, such as the AirPods Max, which show very little difference in the range where the feedback operates between 711 and 5128 measurements :

In all three examples, the HBK conference, Rtings, or the AirPods Max, you can notice that below 1kHz the difference between the fixtures, using only coupling insensitive headphones, is either flat or a very small tilt. If you want to translate the Harman target to the 5128 in that range, then it's just, well, actually, somewhat "trivial", using a term that's already been thrown into that thread : the resulting 5218 target shouldn't deviate from the 711 target by more than a dB at worst, in the form of a tilt, if not stay nearly the same.

Coming up with any other target in that range would logically mean that the preferred EQ for coupling insensitive headphones would have to change, which would mean that the aforementioned listening panel would have suddenly changed their mind. That's quite unlikely.

So this is why I believe that, for certain up to 1kHz, and we can extend that up to a few kHz if we allow for larger tolerances, if we want to rely on Harman's listening tests to jump start what a target for the 5128 should look like, we need to make sure that the EQ profiles that would be necessary to bring coupling insensitive headphones to a preliminary 5128 target would be a decent match for what listening panels would come up with to EQ coupling insensitive headphones to their preference. Which, for now, is safer to bet will look like the Harman target on 711 rigs, at least until additional listening tests are performed.

Still using the example that this listening panel's preferred EQ would just happen to coincide with what is necessary to bring the HD800 to Harman, then you can see how, comparing this preferred EQ to the delta between the HD800, measured on a 5128, and the three targets I used earlier, it illustrates that for two of these (HP.com and Harman's own HBK 5128 target), the EQ required to bring the HD800 in line with the target would necessarily imply a change in preferences. LMG's target, despite not originating from simple examination of the difference between 5128 and 711 fixtures, requires less adjustments for coupling insensitive headphones.

Regardless of my own personal preferences, which aren't anyway that different from what I believe the Harman target is meant to sound like, this is why I think that below 1kHz, as a preliminary 5128 target, LMG's might be the one I'd feel more comfortable with for now.

 

[Deleted member 16543]

You should re-read what you wrote. Standards follow research. So B&K should not be the standard until the research is done. You keep saying science and physics but I don't think your understanding of those two things is correct in this case. You keep saying physics shows this, science shows that, as if what you are saying is some deductively valid thing but it isn't.

This is a straw man if I ever saw one. I have stated that the previous standards were very accurate within the range that most people would match and that you would want to show that the B&K was significantly better with research. I also stated that measurements inside your own ear or a replica of your own ear would be valid for you only. Neither one of those is this statement.

Any curve would technically do this. A curve that is mostly perceptively flat would be the most efficient at this. Harman Target already achieves this, what stops it from working is that industry doesn't adopt the good standard. That's it. If what people listened on is the same tonality as what was mixed on, it would sound better. That is the whole basis for the Harman Target.

You are showing amateur level reasoning here. You beg the question if the improvement of the B&K is significantly better. This is the point that needs to be shown, yet you assume it. If you have this already in your assumptions, discussing with you will go nowhere.

And this is fine. But the proper procedure needs to be done. If it turns out to be better, that is good.

You have a poor definition of sufficient proof. Explain the physics if you are so sure you are correct.

Yes it does. If you and they claim it is better, you have to show it is better. You can't just say it. I believe it is Hitchens's razor that states what is claimed without evidence can be dismissed without evidence. If you show no evidence, it can be dismissed without it.

Because that is how claims work. They need to be supported.

But we are not talking about infinite uses, we are talking about using them to measure and EQ/tune headphones that aim at a target curve developed on that rig. One will be better than the other for this, and if they are basically the same, might as well stick what what we have because it will allow comparison of what is done later with what has already been done.

I still am not quire sure you know reasoning well. Please show the premises and conclusions using only the reasoning shown that you can prove the GRAS is better than itself. Make sure to use only deductive and valid arguments. I would be amazed to see it done. You can do it traditionally or with predicate calculus.

Again, needs to be shown before you can accept it. I'm not saying B&K is useless, just that it needs to be researched.

Except the rig Amir uses is compatible with the target and research. There can be some variations in measurements, but as long as you know what those are and don't take the parts that deviate as de-facto accurate, no issues.

I think Mad_Econimist touched on all relevant points (thank God! Thats was a long post I really didn't feel like writing).
But specifically for this part:

"You have a poor definition of sufficient proof. Explain the physics if you are so sure you are correct."

Here you go: https://gearspace.com/board/all-things-technical/1288581-binaural-theory.html

Read first and get back with questions later, maybe? Instead of assuming the things I'm saying that you don't understand or agree with are due to me "showing amateur level reasoning here", or I don't "know reasoning well".
It is possible, for example, that another way of reading the situation is you are talking to someone smarter than you are (although that might not be the case).
So maybe ask polite clarifying questions instead of insulting?
This first time it's a friendly warning from me. Next time I'll report you.

 

[MayaTlab]

Sean's team got one of the first 5128s produced. In other words, they have had it for good number of years. An attempt was made to calibrate against it as measurements using it against the Harman target showed wildly different results for the same headphone. They took a shot at creating a target for 5128 (I think) but chose to keep it confidential from what I recall. Their new target lowered the variation to acceptable range. I linked to the thread where we discussed all of this earlier.

It's easy to reconstruct, cf earlier posts of mine. I've seen the method criticised at a conceptual level, but even if you accept the latter, Harman didn't select the right sample of headphones for the job and the resulting curve doesn't make a lot of sense, at the very least below 1kHz.

 

[MayaTlab]

I have stated that the previous standards were very accurate within the range that most people would match

Not for IEMs, across the entire spectrum. This is factual information we already know. Cf. earlier discussions here.
That doesn't mean that the 5128 doesn't have faults in that regard either !

I'm not expecting much out of the 5128 for over-ears (based on blocked vs open ear canal measurement tests I'm skeptical that the improved ear canal impedance will be a particularly significant factor). I also think it is unnecessary for that use case, particularly given that the main issue with OE is HPTF and sample variation, something for which the 5128 won't help, but neither do I think that the use of the 5128 is a huge problem nor that the target issue is insurmountable, particularly below a few kHz.

For IEMs though, while I'm still very curious about HPTF variance issues, there's already evidence that, warts and all, it can provide quite a bit of useful info. Again cf earlier.

 

[Deleted member 16543]

Looks like it wasn't a joke after all... My fixture is only slightly different than Harman's. 5128 has no resemblance to GRAS.

vs

We already showed that this slight difference is enough to create variations on the measurements on the same ballpark as the 5128, so just because a head looks different than a cylinder, this doesn't make your rig more similar to the one used in Harman's research. The real issue with measuring headphones is the shape of the pinna and ear canal, and the coupler.
But the head does play a role in finding a good target based on balanced speakers.

Perfection? Research? What are you talking about? There is no such effort being applied to 5128. Someone has thrown out a target curve and is asking the Internet population to vote on it in a totally ad-hoc and random methodology. Come back when you do have research with a body of headphones, controlled tests, etc. and then you have something. Until then, it is just a lot of talk with no substance.

Why isn't asking the internet to validate your results not the best approach, when you're trying to investigate population preferences?
As I said, the rating part of it would probably be a nightmare to hack, but just wait and see.

In the meantime, keep using your rig and its target. It seems to provide the most consistent results, and it's the safest bet so far.
But.. it might not be for long.

 

[Robbo99999]

I can't make it short but I'll try a different approach.

TLDR : think about listeners' preferences and EQ profiles. Why would they change for headphones for which we know that the listeners will have a reasonably consistent experience between them and between fixtures and humans, just because we'd use another test fixture to measure these headphones with ?

The longer version :

I'll still use the data from the HBK conference as it's convenient.

Forget about targets and the 5128 for now. Let's imagine you have a listening panel, give them a pair of HD800, and ask them to EQ them until they sound good to them.

Now, this is fictional as it can't really happen in practice with this method, but let's imagine that the EQ they, on average, come up with, is this (in that example it brings the HD800 - still using the HBK conference data ! - to Harman. It could be any other result for the argument's sake - it's just that the whole point of Harman's years of research is that it's quite a safe bet that it would be close to Harman) :

View attachment 282727

This EQ profile is what this listening panel preferred. logically, these people's preferences, for the HD800, wouldn't suddenly, for some strange reason, change right after this pair has been measured on a 5128 afterwards, would it ? So, in a certain way, the target they'd preferred, when using a 5128, is basically how these HD800 + the EQ profile would measure on a 5128. That's it .

The question you'd then have is : but wait, that will only work for the HD800 ! And you'd be right for high frequencies for certain.

The good news is that quite a few large, open over-ears not only are quite consistent across individuals, but also tend to produce a fairly similar difference between test equipments and between test equipments and real humans, clearly so up to 1kHz, and if you accept quite larger tolerances, generally so up to a few kHz. This is the lovely aspect of headphone designs that are insensitive to coupling issues.

This is quite evident with the HBK's conference data where the HD650, HD800 and Utopia produced a similar difference :

View attachment 282730
In this data set I believe that there are only four large, open over-ears. The K701, interestingly, doesn't behave like the others. I've never seen for these the sort of tests I like to see performed to get an idea whether a pair of headphones is more or less sensitive to coupling issues, for now that remains unexplained to me, even more so when we know that it performs consistently across individuals.

You'll also see the same trend from Rtings' data when they compared their own HMS with the 5128 :

View attachment 282737
I've also highlighted some open over-ears that don't follow that trend as an illustration that this isn't because it's large and sold as "open" that they'll systematically align well.

Generally the story is the same in other comparative scenarios (rig to human, different pinnae, etc.) : not all but most large, open over-ears tend to produce the same difference up to 1kHz, and tend to follow the average trend line reasonably well up to a few kHz if you allow for more inexactitudes. More specifically I'd single out some of Sennheiser's open over-ears, Focal's open headphones, and some of Hifiman's planars.

This would even more so be the case if you throw headphones with a (effective !) feedback mechanism in the mix, such as the AirPods Max, which show very little difference in the range where the feedback operates between 711 and 5128 measurements :

View attachment 282739

In all three examples, the HBK conference, Rtings, or the AirPods Max, you can notice that below 1kHz the difference between the fixtures, using only coupling insensitive headphones, is either flat or a very small tilt. If you want to translate the Harman target to the 5128 in that range, then it's just, well, actually, somewhat "trivial", using a term that's already been thrown into that thread : the resulting 5218 target shouldn't deviate from the 711 target by more than a dB at worst, in the form of a tilt, if not stay nearly the same.

Coming up with any other target in that range would logically mean that the preferred EQ for coupling insensitive headphones would have to change, which would mean that the aforementioned listening panel would have suddenly changed their mind. That's quite unlikely.

So this is why I believe that, for certain up to 1kHz, and we can extend that up to a few kHz if we allow for larger tolerances, if we want to rely on Harman's listening tests to jump start what a target for the 5128 should look like, we need to make sure that the EQ profiles that would be necessary to bring coupling insensitive headphones to a preliminary 5128 target would be a decent match for what listening panels would come up with to EQ coupling insensitive headphones to their preference. Which, for now, is safer to bet will look like the Harman target on 711 rigs, at least until additional listening tests are performed.

Still using the example that this listening panel's preferred EQ would just happen to coincide with what is necessary to bring the HD800 to Harman, then you can see how, comparing this preferred EQ to the delta between the HD800, measured on a 5128, and the three targets I used earlier, it illustrates that for two of these (HP.com and Harman's own HBK 5128 target), the EQ required to bring the HD800 in line with the target would necessarily imply a change in preferences. LMG's target, despite not originating from simple examination of the difference between 5128 and 711 fixtures, requires less adjustments for coupling insensitive headphones.

View attachment 282728
Regardless of my own personal preferences, which aren't anyway that different from what I believe the Harman target is meant to sound like, this is why I think that below 1kHz, as a preliminary 5128 target, LMG's might be the one I'd feel more comfortable about for now.

Ok, so to encapsulate what you're saying, you're saying that large coupling insensitive headphones (HD800, HD650, Utopia) should be used to translate the Harman Headphone Curve into a B&K Harman Headphone Curve? You're saying that you'd EQ those 3 headphones to the Harman Headphone Curve (using existing GRAS measurements), and then you'd measure those same EQ'd headphones on B&K 5128, you'd then take an average of those 3 headphones on B&K 5128 and that would be your starting point for a B&K Harman Headphone Curve - am I understanding your intentions there?

 

[MayaTlab]

Ok, so to encapsulate what you're saying, you're saying that large coupling insensitive headphones (HD800, HD650, Utopia) should be used to translate the Harman Headphone Curve into a B&K Harman Headphone Curve? You're saying that you'd EQ those 3 headphones to the Harman Headphone Curve (using existing GRAS measurements), and then you'd measure those same EQ'd headphones on B&K 5128, you'd then take an average of those 3 headphones on B&K 5128 and that would be your starting point for a B&K Harman Headphone Curve - am I understanding your intentions there?

You could do that, but I'd rather simply use such headphones as a final check on whichever preliminary target you come up with, regardless of the methodology used to create it. As I said, LMG's target wasn't derived using the method you propose, yet it's so far the one closest to not requiring a significant change in EQ profiles, below 1kHz, for coupling insensitive headphones.

Above around 1kHz, I'd have some issues with using that method to be honest and would have a much looser appreciation of the "no change to EQ profile" rule, and above around 4-5kHz it definitely is the wrong approach.

 

[solderdude]

5128 and GRAS will never give the same results as they differ too much in some aspects.
It is a good thing to try and learn how big the delta is on average and can be in certain situations to gain more insight in how it can be used.

Both GRAS and 5128 are not conform reality, as in everyone's individual heads, ears and listening habits, and will always be an approximation that can say something about tonal character in 'ideal' conditions. Not how it will sound exactly to each and everyone.

I guess that what Amir is trying to say (and Dr. Olive as well) is that the GRAS is more suited for research and repeatability and may, in more individual cases, be 'better' in these aspects so might well be a better 'standard' than what the 5128 can do even with a final target.

That does not mean 5128 is worthless and a good target can not be made for it. There just will likely be a bit more variance in some headphones below several 100Hz.
This is probably best 'adjusted to taste/recording' anyway.

 

[Deleted member 16543]

Truth? Anyone who uses this term has no idea of the complexity of the problem.

I think I have a track record of proving exactly the opposite, by now.

Remember, I had Mad_Economist create a target for me for the 5128 I evaluated resulting in nothing useful. I was all set to buy the 5128, and put my faith in his theories only to fall flat on my face for basic reason: that without a properly researched target, the 5128 is a tool looking for a problem to solve.

Yeah. The problem being how do we match measurements to preferences more accurately (specifically for the headphones field).
The problem is real. And the 5128 is a more accurate tool to look for a solution (getting closer to the Truth was a figure of speech, obviously)

Compounding the problem was difficulty of mounting either IEMs or headphones on it.

That is indeed a huge problem. I can see why, on top of not having a valid target, this is good enough reason for someone like you to stick to what is known.

I read a long time ago that companies that make dog toys aim to please the dog owners, and not dogs. This is why they make them look like hotdogs, bacon, etc. which has no meaning to a dog but a lot to the owner in the shopping isle. The 5128 seems like that to some people it seems. It appeals to their nerdy side that this thing must solve this problem better. They have gone ahead and purchased (?) this expensive tool with no proof point of it.

Meh. This is a weak point. The hot dog shaped toys are used for dogs.
The head shaped measuring rig is meant for people, who indeed have a head shaped like that.
The proof point is just that. It is a more accurate tool to investigate what our eardrums sense, because it more closely matches the geometry we all have that's surrounding our eardrums.
It can better replicate the interactions that create the sound wave finally sensed by our eardrums, because the geometry of it matches our head to a higher degree.
Arguing this is like arguing that a ruler that shows mm has to be proven better than one that only shows cm.
If you're doing approximate work it isn't, from the practical point of view.
But in and of itself it is better (as in more accurate). There's no denying it.

So yes, you do have to deliver the goods when you are arguing with someone who has used 5128 and knows its limitations. Pleading to accept your declaration is not going to work.

As per above, your rig is "better" in terms of results (because it comes with a researched target. Well, actually, your rig is different than the one used for the research, so that's kind of questionable too, but let's not get into this too, unless you really want to).
The 5128 is better because it is a more accurate tool (but alas, almost unusable at the moment because of the lack of a validated target curve).

We're talking about 2 different things: practical value (GRAS) vs accuracy for further, more solid research (5128).

 

[Deleted member 16543]

5128 and GRAS will never give the same results as they differ too much in some aspects.
It is a good thing to try and learn how big the delta is on average and can be in certain situations to gain more insight in how it can be used.

Both GRAS and 5128 are not conform reality (as in everyone's individual heads, ears and listening habits, and will always be an approximation that can say something about tonal character in 'ideal' conditions. Not how it will sound exactly to each and everyone.

I guess that what Amir is trying to say (and Dr. Olive as well) is that the GRAS is more suited for research and repeatability and may, in more individual cases, be 'better' in these aspects so might well be a better 'standard' than what the 5128 can do even with a final target.

That does not mean 5128 is worthless and a good target can not be made for it. There just will likely be a bit more variance in some headphones below several 100Hz.
This is probably best 'adjusted to taste/recording' anyway.

If all these fitting issues are indeed real with the 5128, I can envision a world where the 5128 full rig is used to come up with a valid target when used as an anatomically accurate binaural microphone to measure well balanced speakers at the listening position, then mounting the ears on a rig that gives less problems with fitting, since headphones bypass the head anyway (unless you are measuring headphones acoustical cross talk. For that you would need a human head as well).

One field of potential use for the 5128 not many seem to have considered so far is investigation on how the perceived balance changes when using speakers (which have a lot of acoustical crosstalk) vs headphones (which have close to none, psychoacoustically speaking), and the adjustment curve to make the latter match the former.
Did Harman consider this? If so, how did they investigate it without a full, proper binaural microphone?
It's little things like these that make me say that the research in this field is still in its initial stage. And yes, reaserch is never really over, but it comes a time when improvements are few and far in between.
We're not there yet, and I see the 5128 simply as the better tool that will make us walk the next few steps.

On a personal note, I think that if it wasn't so obviously overpriced the controversy around the 5128 would be much less.
Buyer bias (which works both ways) would be greatly reduced if the rigs were similarly priced.
And no, this is not an invitation for GRAS to quadruple its price

Sorry, @solderdude. I started replying to your post and kind of went off on a tangent reasoning..
I guess your post inspired me.

 

[Robbo99999]

You could do that, but I'd rather simply use such headphones as a final check on whichever preliminary target you come up with, regardless of the methodology used to create it. As I said, LMG's target wasn't derived using the method you propose, yet it's so far the one closest to not requiring a significant change in EQ profiles, below 1kHz, for coupling insensitive headphones.

Above around 1kHz, I'd have some issues with using that method to be honest and would have a much looser appreciation of the "no change to EQ profile" rule, and above around 4-5kHz it definitely is the wrong approach.

Cool, so you're just saying you'd use those 3 headphones as a means of checking the 5128 Target Curve? Well I think it's logical to at least exclude the high positional / high person to person / high rig to rig variance headphones from the study during the listening testing & feedback.

 

[Robbo99999]

5128 and GRAS will never give the same results as they differ too much in some aspects.
It is a good thing to try and learn how big the delta is on average and can be in certain situations to gain more insight in how it can be used.

Both GRAS and 5128 are not conform reality (as in everyone's individual heads, ears and listening habits, and will always be an approximation that can say something about tonal character in 'ideal' conditions. Not how it will sound exactly to each and everyone.

I guess that what Amir is trying to say (and Dr. Olive as well) is that the GRAS is more suited for research and repeatability and may, in more individual cases, be 'better' in these aspects so might well be a better 'standard' than what the 5128 can do even with a final target.

That does not mean 5128 is worthless and a good target can not be made for it. There just will likely be a bit more variance in some headphones below several 100Hz.
This is probably best 'adjusted to taste/recording' anyway.

And you can probably get around some of those drawbacks (in relation to bass leakage) by using a flat cheek version of the 5128 - which you said Crinacle is gonna be doing.

 

[KiyPhi]

I think Mad_Econimist touched on all relevant points (thank God! Thats was a long post I really didn't feel like writing).
But specifically for this part:

"You have a poor definition of sufficient proof. Explain the physics if you are so sure you are correct."

Here you go: https://gearspace.com/board/all-things-technical/1288581-binaural-theory.html

Read first and get back with questions later, maybe? Instead of assuming the things I'm saying that you don't understand or agree with are due to me "showing amateur level reasoning here", or I don't "know reasoning well".
It is possible, for example, that another way of reading the situation is you are talking to someone smarter than you are (although that might not be the case).
So maybe ask polite clarifying questions instead of insulting?
This first time it's a friendly warning from me. Next time I'll report you.

Report me all you'd like. You linked me to a post about binaural audio. It didn't really touch on anything involving this discussion. None of that justifies the claims you've made about the B&K being better or the claims you've made about reasoning, physics, or the conclusions from the things we're saying. It didn't do anything other than waste my time giving you the benefit of the doubt that there might be something in there. You have shown that further discussion with you would be fruitless. Nothing more, nothing less.

 

[Deleted member 16543]

Report me all you'd like. You linked me to a post about binaural audio. It didn't really touch on anything involving this discussion. None of that justifies the claims you've made about the B&K being better or the claims you've made about reasoning, physics, or the conclusions from the things we're saying. It didn't do anything other than waste my time giving you the benefit of the doubt that there might be something in there. You have shown that further discussion with you would be fruitless. Nothing more, nothing less.

You asked for the physics. The physics explanation of why a better match between the geometry of the rig and that of a human head is necessary for better accuracy is plainly shown there.
The match ensures that the sensed pressure at the mic capsule is more consistent to the one sensed by our eardrums, IRRESPECTIVE of the source position. That's the whole reason to use an anatomically accurate rig. Different headphones are different sound sources. Given the location of the driver one can also say that they are distributed sources, rather than point sources. Therefore the shape of the pads, the volume, closed vs open, etc.. and of course the geometry of the head around the eardrum needs to be recreated as close as possible to reality to get a better measurement of what our eardrums sense in reality.

If you have specific questions I can try to help, but it looks like you just came here to throw punches, to be honest.