Was anybody at the Sean Olive talk at CanJam?

[deadkrillin]

Which IE target was mislabeled? Was the mislabeled target using BK5128DF+2015 Harman filters, the result of subtracting Harman OE2015 with Harman in-room?

View attachment 356658

View attachment 356659

It seems Sean used Headphones.com's DF HRTF + Harman's 2015 filters applied (pic related).

Resolve is saying they meant to have him test the 2018 filters on the same baseline instead, which would've had the ~3kHz adjustment we know serves as the difference between Harman 2015 and 2018 AE/OE. This would've made it very close to the SoundGuys target which was the highest scoring target among trained listeners.

 

[USER]

 

[BrooklynNick]

Very interesting presentation. I do question the idea that preference for bass is related to hearing loss. I'd love to see more testing of this.

An alternative explanation for bass preference by age could be musical preference. Perhaps our bass preferences are shaped by the music we were exposed to when we are very young ie it is related to socialization. Popular music has been getting more and more bass heavy over time. No one had a subwoofer in their living room or in their car when Baby Boomers were growing up and hardly any when Gen X was coming of age, for example.

 

[deadkrillin]

View attachment 356665

Those conclusions certainly are interesting! Sean has now said in no uncertain terms that the 5128 is the more accurate hearing simulator, so I guess in that respect Apple, Sony, Samsung, SoundGuys, Headphones.com, RTINGS, Crinacle and whoever else has one for IEM measurements are right to be using it.

I'm not surprised that in this testing Harman IE 2019 wasn't OBLITERATED like all of the reviewers wanted. It stands to reason that it among most of the other curves tested would be roughly equally preferable. Of course, even Sean says we shouldn't form real conclusions from all of this, as it's essentially a very small pilot study with minor if any statistical significance. I think the best I can glean from this is that these findings are in-step with some of the broader conclusions of Harman's work: humans generally don't prefer flat speakers (or in this case, DF without preference adjustments).

I'm a little curious why untrained listener results are being weighted as highly as they are given how inconsistent their ratings were (they rated very similar sounding targets very differently) but even that's probably reading a bit far into something that ought not be paid too much attention. Hopefully the MOA testing he's planning on doing next has more subjects/controls and we can get to the bottom of where we should be going next!

 

[markanini]

View attachment 356663

It seems Sean used Headphones.com's DF HRTF + Harman's 2015 filters applied (pic related).

Resolve is saying they meant to have him test the 2018 filters on the same baseline instead, which would've had the ~3kHz adjustment we know serves as the difference between Harman 2015 and 2018 AE/OE. This would've made it very close to the SoundGuys target which was the highest scoring target among trained listeners.

Funny, I suggested something similar a year ago: https://www.audiosciencereview.com/...target-you-can-try-the-eqs.43209/post-1624317

Those conclusions certainly are interesting! Sean has now said in no uncertain terms that the 5128 is the more accurate hearing simulator, so I guess in that respect Apple, Sony, Samsung, SoundGuys, Headphones.com, RTINGS, Crinacle and whoever else has one for IEM measurements are right to be using it.

I'm not surprised that in this testing Harman IE 2019 wasn't OBLITERATED like all of the reviewers wanted. It stands to reason that it among most of the other curves tested would be roughly equally preferable. Of course, even Sean says we shouldn't form real conclusions from all of this, as it's essentially a very small pilot study with minor if any statistical significance. I think the best I can glean from this is that these findings are in-step with some of the broader conclusions of Harman's work: humans generally don't prefer flat speakers (or in this case, DF without preference adjustments).

I'm a little curious why untrained listener results are being weighted as highly as they are given how inconsistent their ratings were (they rated very similar sounding targets very differently) but even that's probably reading a bit far into something that ought not be paid too much attention. Hopefully the MOA testing he's planning on doing next has more subjects/controls and we can get to the bottom of where we should be going next!

I'm not surprised either. Although I don't care about reviwer incentives as much, and I still think it might be not be preferred for a sizable percentage of listeners. Given that the individual FR variation of the IEM tested is as much as 10dB at 5kHz. I hope there is solution to handle conditions like these that's still fully data driven. At worst compliance in the top octaves will have to be done with some leniency when in situ FR is can have such high variability.

 

[deadkrillin]

Funny, I suggested something similar a year ago: https://www.audiosciencereview.com/...target-you-can-try-the-eqs.43209/post-1624317

I'm not surprised either. Although I don't care about reviwer incentives as much, and I still think it might be not be preferred for a sizable percentage of listeners. Given that the individual FR variation of the IEM tested is as much as 10dB at 5kHz. I hope there is solution to handle conditions like these that's still fully data driven. At worst compliance in the top octaves will have to be done with some leniency when in situ FR is can have such high variability.
View attachment 356680

I largely agree, yeah. The data you've posted, as well as the fact that the majority of difference between the curves tested lies in the 4-8kHz area that a lot of people don't like about Harman's IE target, makes me think the MOA testing might do well to incorporate an adjustment band in this area. Can deal with the preference variance between listeners as well as anatomical variance all at once!

 

[Resolve]

View attachment 356663

It seems Sean used Headphones.com's DF HRTF + Harman's 2015 filters applied (pic related).

Resolve is saying they meant to have him test the 2018 filters on the same baseline instead, which would've had the ~3kHz adjustment we know serves as the difference between Harman 2015 and 2018 AE/OE. This would've made it very close to the SoundGuys target which was the highest scoring target among trained listeners.

That's right. We don't use an alternative target curve given the way we present the data, we were just trying to guess what class 1 would be for a population average DFHRTF baseline. While all of them fall within the preference bounds, it's still unclear how much bass most people would prefer when faced with such a baseline. The 2018 filters would look like this, which is similar to the SoundGuys result barring a few differences in bass and treble. But even that could potentially not be most preferred with a larger study, which we hopefully get soon.

 

[deadkrillin]

That's right. We don't use an alternative target curve given the way we present the data, we were just trying to guess what class 1 would be for a population average DFHRTF baseline. While all of them fall within the preference bounds, it's still unclear how much bass most people would prefer when faced with such a baseline. The 2018 filters would look like this, which is similar to the SoundGuys result barring a few differences in bass and treble. But even that could potentially not be most preferred with a larger study, which is what we're hoping to see.

Oh, hello!

I'm aware you guys use the bounds instead of a single curve anyway (was cool to see Dr. Olive use this as well) so I didn't put much stock into the "Headphones.com target" anyway. Now that I mention it, do any of you on the Headphones.com team use this target for personal use when equalizing IEMs? It seems like it'd be a pretty nice target but I confess I've not tried it.

 

[Resolve]

Oh, hello!

I'm aware you guys use the bounds instead of a single curve anyway (was cool to see Dr. Olive use this as well) so I didn't put much stock into the "Headphones.com target" anyway. Now that I mention it, do any of you on the Headphones.com team use this target for personal use when equalizing IEMs? It seems like it'd be a pretty nice target but I confess I've not tried it.

It could be used, although I think most of us who do EQ are supplementing that with our own personalization for individual length mode issues and such with a given IEM - certainly I am. It's a reasonable starting point. I think what would be useful for people is to try comparing population average DFHRTF baseline vs 5128 DFHRTF baselines, both with 2018 filters applied.

 

[CTWily]

Those conclusions certainly are interesting! Sean has now said in no uncertain terms that the 5128 is the more accurate hearing simulator, so I guess in that respect Apple, Sony, Samsung, SoundGuys, Headphones.com, RTINGS, Crinacle and whoever else has one for IEM measurements are right to be using it.

It's definitely good to see after all this time! Saves us all a lot of re-measuring.

I'm interested to see more about bespoke 3D-printed ears for the fixture, that's a cool idea.

 

[Somafunk]

While all of them fall within the preference bounds, it's still unclear how much bass most people would prefer when faced with such a baseline.

As someone who listens to electronic music with a lot of artificial bass created in the box or with analogue synths, I’d prefer greater bass rising from 100hz down to a 5-8db rise at 20hz, I honestly can’t say when I last listened to music created by actual instruments and recorded live before being mixed down/mastered for release.

 

[MacClintock]

View attachment 356663

It seems Sean used Headphones.com's DF HRTF + Harman's 2015 filters applied (pic related).

Resolve is saying they meant to have him test the 2018 filters on the same baseline instead, which would've had the ~3kHz adjustment we know serves as the difference between Harman 2015 and 2018 AE/OE. This would've made it very close to the SoundGuys target which was the highest scoring target among trained listeners.

Resolve has to say something to get out of this embarrasment, critizising heavily the Harman target and then seeing his own target coming out much less prefered and Harman basically being vindicated, in spite of 5128.

 

[amirm]

Those conclusions certainly are interesting! Sean has now said in no uncertain terms that the 5128 is the more accurate hearing simulator, so I guess in that respect Apple, Sony, Samsung, SoundGuys, Headphones.com, RTINGS, Crinacle and whoever else has one for IEM measurements are right to be using it.

What does "using" mean? As long as they throw out target curves that are not shown to be correct in human studies, they are producing worse results, not better. You can have the world's most accurate measurement rig but if you don't have a reference of what the response should be, it is for not.

 

[amirm]

I do question the idea that preference for bass is related to hearing loss.

It makes intuitive sense. If you have lost your high frequency hearing and you boost the bass, you might think the overall balance is wrong. Someone who hears both ends of the spectrum will not face this. As I EQ headphones and speakers, I frequently run into this in that I boost the bass to correct response and it may sound too much. But once I fix treble as well, the balance is there.

But yes, specific research would be good as I suspect good portion of headphone enthusiasts have less than ideal frequency response.

 

[MacClintock]

What does "using" mean? As long as they throw out target curves that are not shown to be correct in human studies, they are producing worse results, not better. You can have the world's most accurate measurement rig but if you don't have a reference of what the response should be, it is for not.

But did Sean Olive not present the "mapping" of the G.R.A.S. Harman target to the 5128 Harman target in the talk?

 

[MacClintock]

BTW, congrats to ASR and Amir for being featured prominently in this talk, also with the speaker measurement.

 

[Resolve]

Resolve has to say something to get out of this embarrasment, critizising heavily the Harman target and then seeing his own target coming out much less prefered and Harman basically being vindicated, in spite of 5128.

I do not have a target curve. And what I'm saying is the truth of it, it was the 2015 filters that got tested, not 2018. Spin it however you want lol.

Edit: it even says it's not the one with ear gain adjustment in the slide.

 

[MacClintock]

I do not have a target curve. And what I'm saying is the truth of it, it was the 2015 filters that got tested, not 2018. Spin it however you want lol.

Edit: it even says it's not the one with ear gain adjustment in the slide.

Whatever is your excuse, if don't even have a target curve, it is even worse, Sean Olive and Harman have one, or a few, and they are backed up by a lot of research, their's and other's, and now even on the 5128. What do you have? Polemical clickbait videos on YouTube and being practically working in the marketing department of a headphone seller.

 

[amirm]

But did Sean Olive not present the "mapping" of the G.R.A.S. Harman target to the 5128 Harman target in the talk?

He did but until they use it, they are producing useless analysis.

 

[SalutaryGraveness]

A more sophisticated and accurate simulator is fine and great, but without the implementation of a proper scientific protocol to develop a reference target curve with it, I don't see the point of advertising alternatives as "superior" out of thin air. You first develop the instrument, do the research with it, and then you discover how the transducer is more or less supposed to be tuned. Because honestly it feels a bit "cargo culty" to present an alternative as superior, do the advertisement second, develop the instrument third and then do a research if any.

That doesn't mean that "artisan" alternatives are bad and can't be superior according to whatever relevant metric you choose. But good advertisement, marketing, influencing on social media, infotainment and "charisma" are not a replacement for proper research to validate any claim.