Master Complaint Thread About Headphone Measurements

[pwjazz]

The day I am not "enjoying" this, you will know as I will stop cold. On that day, you would to reflect back and see if you had anything to do with that and if so, whether you are OK with it.

Okey dokey. I thought I had been contributing over the last couple of years, but since it seems that I'm on a trajectory to spoil the fun, I'll just leave and get out of your hair.

@AdamG247 I'm thinking perhaps I should have left things at my original comment, but I am glad that my bellyaching at least spurred a constructive discussion of Amir's workload. I couldn't find a way to delete my account-please do so for me if it's not asking too much.

 

[Robbo99999]

Okey dokey. I thought I had been contributing over the last couple of years, but since it seems that I'm on a trajectory to spoil the fun, I'll just leave and get out of your hair.

@AdamG247 I'm thinking perhaps I should have left things at my original comment, but I am glad that my bellyaching at least spurred a constructive discussion of Amir's workload. I couldn't find a way to delete my account-please do so for me if it's not asking too much.

You don't have to leave outright, just visit once in a while instead if you're fed up with it.....stick around!

 

[MZKM]

Testing a product before return period? I mean, that sounds like the person is getting a free chance to return a product if they don't like a review. I understand not wanting a bad product or something, but they're basically using you as a service there. Really, if they send it to you, it's on them. You could set some boundaries to lighten the load in that respect.

Disagree about sending in during the return period, this is an excellent way to get products in for review. It is of course though up to Amir whether he can and wants guarantee such to the purchaser.

Mailing in products you already own can be expensive.

Drop shipping to Amir but him going over the return time eliminates the purchaser’s ability to return it if it performs sub-par (maybe they would have returned it based on their subjective evaluation).

 

[Kosta]

So can someone explain the point of these measurements, especially trying to EQ to target. Surely the only things that matters to speakers and headphones is amount of distortion and frequency response ? after that its all subjective whether you like bass, vocals , etc So the recommendations amir is giving out are really just his own subjective tastes as to how he likes a headphone to sound?

 

[solderdude]

Surely the only things that matters to speakers and headphones is amount of distortion and frequency response ?

I would say frequency response and a lot less distortion, unless it becomes substantial.
For speakers I would not rule out room effects and for headphones I would not rule out positioning, pad conditions and seal (for closed headphones).

So the recommendations amir is giving out are really just his own subjective tastes as to how he likes a headphone to sound?

Just like anyone else's EQ is based on measurements based on a specific test fixture with a 'generic' compensation/target curve where computers generate an 'exact' EQ (which isn't exact by definition but is considered exact).

Amir creates an EQ by correcting the overall errors in FR with the Harman target in mind and also listens.
So that's why all EQ suggestions from everyone creating EQ suggestions differ and can differ substantially.
That said in most cases the 'overall' FR improves so the sound improves while not sounding the same.

There is your 'gain' ... improved tonality overall. That is the point of measurements. It is very hard to do this by ear only.

Then... one can change EQ to suit ones taste or circumstances (tone control rather than EQ).

 

[Robbo99999]

So can someone explain the point of these measurements, especially trying to EQ to target. Surely the only things that matters to speakers and headphones is amount of distortion and frequency response ? after that its all subjective whether you like bass, vocals , etc So the recommendations amir is giving out are really just his own subjective tastes as to how he likes a headphone to sound?

Read up on the Harman Research, it's not all subjective, here explained by Oratory:
https://www.reddit.com/r/headphones/comments/78x77b/_/doyj84e
And here some slides from Harman themselves which is really a great resource to understand how it fits together:
https://www.listeninc.com/wp/media/Perception_and_-Measurement_of_Headphones_Sean_Olive.pdf

 

[Kosta]

Read up on the Harman Research, it's not all subjective, here explained by Oratory:
https://www.reddit.com/r/headphones/comments/78x77b/_/doyj84e
And here some slides from Harman themselves which is really a great resource to understand how it fits together:
https://www.listeninc.com/wp/media/Perception_and_-Measurement_of_Headphones_Sean_Olive.pdf

Thanks I think the line that best explains it is:
In other words: The Harman Target is basically the same sound that the artist and engineers heard when creating the music that we hear.

 

[solderdude]

The Harman Target is basically the same sound that the artist and engineers heard when creating the music that we hear.

That would only be true if the monitors used in the studio had a Harman tuning and listening is done in a similar room/circumstances and at the same average SPL.
But kind of 'yes' as tonality would be similar.

 

[Robbo99999]

Thanks I think the line that best explains it is:
In other words: The Harman Target is basically the same sound that the artist and engineers heard when creating the music that we hear.

Yes, it's quite a close approximation of this for a lot of people, but because our HRTF's are somewhat different it means that one Target can't cater 100% accurately for everyone. An ideal situation would be your own head & torso being used in the Harman experiments that determined the Harman Curve (instead of a dummy head), but then that would only be applicable to you, so couldn't be used for anyone else......but really a personalised/measured/calculated target would be the most accurate, but most people don't have access to such listening rooms and the equipment and know-how of how to create these curves. The Impulcifier Project (https://github.com/jaakkopasanen/Impulcifer) is a home-brewed version of this, and the Smyth Realizer (https://smyth-research.com/) is the very expensive version.....but beyond doing all that the Harman Target for headphones is a good approximation for a lot of people.....it's what I use, I think it's good and quite accurate.

 

[Sombreuil]

Would it be difficult to mesure a game console? A Playstation 3/4/5 for example. Always wondered what it'd be like.

 

[amirm]

Would it be difficult to mesure a game console? A Playstation 3/4/5 for example. Always wondered what it'd be like.

I don't know really because I am not a gamer. Can you bring your own music to it and have it play locally? If so, then that solves the problem of having it play a test track. Next is what to measure. Do they have analog audio out? If so, then yes, I can measure it.

 

[Sombreuil]

For the first question, yes. For the latter I believe it's not the case when it comes to recent consoles .

 

[amirm]

How do they connect to gaming headset?

 

[Sombreuil]

Bluetooth or directly into the pad.

 

[sprellemannen]

Haven't taken a vacation in year and half.

Treat yourself with a good vacation.
The work you do and have done regarding ASR is beyond amazing.

 

[John B]

Anyone that lasts as long as Amir at MSFT possibly burnout proof! I’m not worried

 

[ADU]

Howdy Amir & other ASR members,

I am an ASR newbie. But just wanted to say that I'm glad you've decided to start doing some headphone measurements, in addition to the other audio gear you've been doing. I wish they could've been done with the new HBK 5128 system. But I understand that was probably too big a lift to start with.

With a little help from Pierre Aubert's database, I've found both your reviews and spinorama plots of loudspeakers very helpful indeed! So thank you sincerely for all the work that you and your members have put into those!

I have one small request. And that is to include diffuse field measurements with your headphone reviews and measurements, in addition to the raw and Harman compensated plots.

Diffuse field graphs would have a couple of advantages. They'd remove the dominant ear resonances making it easier to assess a headphone's linearity and tonal balance (see my comments below re sound power). They would also allow your measurements to be more easily compared to the DF compensated measurements of some other sites. Including those which use a different measurement apparatus than the GRAS system you're currently using. And they would make it easier to compare your headphone measurements to the frequency response of loudspeakers, using the diffuse field equivalent in that domain, otherwise known as sound power.

If you ever decide to invest in a full head and torso rig, such as the HBK 5128-C or D. Then it would also be interesting to see some raw in-ear frequency response measurements of loudspeakers in a typical home listening space, made on the same rig that you use for headphone measurements. For comparison to your raw headphone plots. Maybe there is also a similar dummy head and torso that could be used with your current GRAS ear simulator for something like this as well.

In the meantime though, I think comparing the diffuse field headphone measurements to sound power is probably the next best way to accomplish this, rather than using the Harman curve. Others may disagree though.

Thank you again sincerely, for all your time and efforts. And the amazing resources you provide here at ASR.

 

[Robbo99999]

Howdy Amir & other ASR members,

I am an ASR newbie. But just wanted to say that I'm glad you've decided to start doing some headphone measurements, in addition to the other audio gear you've been doing. I wish they could've been done with the new HBK 5128 system. But I understand that was probably too big a lift to start with.

With a little help from Pierre Aubert's database, I've found both your reviews and spinorama plots of loudspeakers very helpful indeed! So thank you sincerely for all the work that you and your members have put into those!

I have one small request. And that is to include diffuse field measurements with your headphone reviews and measurements, in addition to the raw and Harman compensated plots.

Diffuse field graphs would have a couple of advantages. They'd remove the dominant ear resonances making it easier to assess a headphone's linearity and tonal balance (see my comments below re sound power). They would also allow your measurements to be more easily compared to the DF compensated measurements of some other sites. Including those which use a different measurement apparatus than the GRAS system you're currently using. And they would make it easier to compare your headphone measurements to the frequency response of loudspeakers, using the diffuse field equivalent in that domain, otherwise known as sound power.

If you ever decide to invest in a full head and torso rig, such as the HBK 5128-C or D. Then it would also be interesting to see some raw in-ear frequency response measurements of loudspeakers in a typical home listening space, made on the same rig that you use for headphone measurements. For comparison to your raw headphone plots. Maybe there is also a similar dummy head and torso that could be used with your current GRAS ear simulator for something like this as well.

In the meantime though, I think comparing the diffuse field headphone measurements to sound power is probably the next best way to accomplish this, rather than using the Harman curve. Others may disagree though.

Thank you again sincerely, for all your time and efforts. And the amazing resources you provide here at ASR.

About Diffuse Field Target, if you want to compare the measurements to that then you could download the attached zipfile of measurements that Amir puts at the end of his review and then import it into REW, then set it against Diffuse Field Target to see how different it is and even create your own EQ's using REW to that diffuse field target. You could get the Diffuse Field Target from one of Oratory's pdf's, he does the occasional EQ to Diffuse Field - you could use Virtuix CAD and the SPL Trace tool within that program to trace the Diffuse Field Target & import it into REW. That's all a bit of a faff if all you're curious about is how it compares to Diffuse Field Target, but I'm not sure how much interest there is in Diffuse Field Targets on this site to warrant Amir including an additional graph of the measurement with Diffuse Field Target overlaid.

You talk about comparing results from Amir's GRAS unit with different measurement systems that aren't GRAS - well you can't do that because they don't have the same design and therefore aren't comparable in their measurements, and this doesn't change based on whether or not you choose a Harman Curve or a Diffuse Field Target. (Harman Headphone Target is only applicable/useable with GRAS measurement devices as this is what the research was based on....and the Diffuse Field Targets will only be compatible with the specific measurement systems too).

I'm not with you on when you talk about comparing headphone measurements to speaker measurements, they're not comparable, and I'm not sure what you'd want to achieve?

 

[ADU]

Thank you very much for the reply, Robbo99999.

I appreciate the tips on exporting the raw data.

I currently use the Equalizer APO's Configuration Editor as my EQ, and also for some general graphing and plotting jobs. And have already made a high resolution plot of Oratory's DF curve in there. So if I can figure out some way to import the raw ASR headphone data, then I could probably use the Editor's stacking function to compute some equivalent DF curves from that.

I'd be curious to see the DF curve that was provided with the ASR GRAS setup as well, for comparison to the one that Oratory is currently using. Because Oratory recently made some changes to his DF compensation curve. And I'm not really sure why that was done.

It is getting rather late here. So I'll have to get back to you on some of the other stuff. Maybe there are some other knowledgeable folks here who can dig back in their memories a bit, and try to explain how DF compensation was supposed to be used to make the data between different rigs more easily comparable and interchangeable though. Because I haven't had much luck explaining this concept to folks on some of the other forums. (I'll give it a try though, if nobody else feels like it.)

I'm afraid the hype about the Harman curve has led to alot of confusion and misunderstanding though about some of the original uses and purposes for both diffuse and free-field compensation. Because these still can have some very beneficial uses for data comparison purposes.

 

[ADU]

I'm not with you on when you talk about comparing headphone measurements to speaker measurements, they're not comparable, and I'm not sure what you'd want to achieve?

Compare the sound power curves, especially on some of the better quality speakers with good bass extension here...

https://pierreaubert.github.io/spinorama/

...to the diffuse field plots of some of Oratory's headphones here...

https://headphonedatabase.com/oratory/headphones?ids=81,76,66,12

...and I think you'll begin to understand a little better what I'm referring to. The response curves are very similar.

To see the DF compensated plots on Oratory's graph, you will have to select "Diffuse Field" from the "Target Response" selector. And change the "Include Raw" setting to "No".

Here's a screen grab of the 4 headphones I've selected as examples above, so you can see what their plots should look like. You can select some others though from the "Headphones" drop-down menu. But can only display up to 5 headphones at time on the same graph.

The diffuse field plots of many of the better-sounding headphones with good bass extension will have essentially the same downward slope from the bass to the treble as the speakers' sound power curves. And they will also frequently (though not always) have a similar dip in their response in the upper midrange at approximately the same frequency as the midrange-tweeter crossover on alot of loudspeakers, which is usually centered somewhere around 2 to 2.5 kHz.

You can see this same type of dip at around 2k or so in the sound power responses of the following speakers. And also inverted as a small peak in some of their sound power directivity curves...

https://pierreaubert.github.io/spinorama/Infinity Prelude MTS/Infinity/index_vendor.html
https://pierreaubert.github.io/spinorama/Adam T8V/ASR/index_asr.html
https://pierreaubert.github.io/spinorama/Buchardt Audio S400/ASR/index_asr.html
https://pierreaubert.github.io/spinorama/Genelec 8050B/ASR/index_asr.html
https://pierreaubert.github.io/spinorama/Revel F35/ASR/index_asr-v2-20210415.html
https://pierreaubert.github.io/spinorama/Revel Ultima2 Salon2/Revel/index_harman-v1-2007.html

The fairly pronounced notch between 9 and 10 kHz on many of Oratory's headphones plots is a pinna artifact, btw. And not something which should be corrected, or EQ'd out.

Although it's somewhat harder to see, there are also some similar notches in the treble above that, particularly at around 15 kHz and sometimes also around 20 kHz (or slightly above), which are caused by the same effect. Those are all normal features and resonances in his particular rig and measurements, that should not be corrected with an EQ.