Master Complaint Thread About Headphone Measurements

[Deleted member 16543]

Klippel NFS only uses gating at higher frequencies (set to about 1 kHz in my configuration). Below that, it is actually able to separate the direct sound of the speaker versus reflected ones. As such, it produces completely anechoic response at full resolution.

Interesting. How is it able to separate direct vs. reflected sound at low frequencies? Time domain analysis? And how low of a frequency is it able to measure anechoically (in a typically sized room)?
Even a proper anechoic room at low frequencies is not really anechoic, so the meaning behind my quasi- prefix is to intend that the measurement is not done in a real anechoic room, not to put the equipment you use on a lower rung on the 'anechoic scale'.

 

[ADU]

Interesting. How is it able to separate direct vs. reflected sound at low frequencies? Time domain analysis? And how low of a frequency is it able to measure anechoically (in a typically sized room)?
Even a proper anechoic room at low frequencies is not really anechoic, so the meaning behind my quasi- prefix is to intend that the measurement is not done in a real anechoic room, not to put the equipment you use on a lower rung on the 'anechoic scale'.

I don't know if this will help at all, sax512. But there are a couple threads on the subject here....

https://www.audiosciencereview.com/...nderstanding-how-the-klippel-nfs-works.13139/
https://www.audiosciencereview.com/...ner-the-mathematics-and-everything-else.9970/

 

[amirm]

Interesting. How is it able to separate direct vs. reflected sound at low frequencies? Time domain analysis? And how low of a frequency is it able to measure anechoically (in a typically sized room)?

Yes, it uses phase shift/time differential between direct and reflected sounds. This obviously gets messy at higher frequencies so it only limits the correction to certain frequency. You give it this parameter in the form of roundtrip distance to first reflection. It enforces a 5.5 meter minimum in later versions of software so that is what I use. Turns out the power of reflections has shrunk a ton by then anyway:

You can see the reflections in dashed blue. and the impact they haver on in-room measurements ("fitted" in green). By the time you get to the cut off in orange (vertical), the impact is quite low and lost in the noise.

 

[Deleted member 16543]

Yes, it uses phase shift/time differential between direct and reflected sounds. This obviously gets messy at higher frequencies so it only limits the correction to certain frequency. You give it this parameter in the form of roundtrip distance to first reflection. It enforces a 5.5 meter minimum in later versions of software so that is what I use. Turns out the power of reflections has shrunk a ton by then anyway:

View attachment 146600

You can see the reflections in dashed blue. and the impact they haver on in-room measurements ("fitted" in green). By the time you get to the cut off in orange (vertical), the impact is quite low and lost in the noise.

That's pretty clever. Is it able to discern a front firing bass reflex from the cone itself, or does it measure their combined effect?
By the way, that +/- 2-3 dB is in line with the deviation I get following a near field type of listening, in different rooms, with different types of sound treatment as well.. even at low frequencies. That's one of the main reasons I like near field listening: consistency of sound balance no matter where you place your speakers.

 

[solderdude]

just an idea...

https://www.audiosciencereview.com/forum/index.php?threads/complaint-thread-about-speaker-measurements.11139/

 

[Deleted member 16543]

just an idea...

https://www.audiosciencereview.com/forum/index.php?threads/complaint-thread-about-speaker-measurements.11139/

https://en.m.wikipedia.org/wiki/Interdisciplinarity

 

[ADU]

I'm still working my way through this topic. So this may already have been posted somewhere already.

For some of the newbie headphone users that may be a little lost or in need of a brain boost on things like anechoic chambers, spinorama graphs, directivity/dispersion, the circle of confusion, and what have you, this (and also Dr. Toole's book on Sound Reproduction) is a pretty good place to start to get a little better understanding of some of that...

It is just my personal opinion of course (and the above video is a little bit of an advertisement for the Harman philosophy of doing things), but I believe that a true grasp and understanding of how a headphone should sound and measure probably begins with a better understanding of most of what Dr. Toole discusses in the above seminar related to the sound and measurement of speakers. YMMV though on that.

I have watched the above video probably a dozen times (or more) by now though. And still take away different things each time I watch it.

Thank you, btw, to solderdude for the above link to the sister topic in the speaker forum. I'm sure I'll also enjoy reading some of the stuff in there as well.

 

[ADU]

Another good followup discussion to the above, from earlier this year.

 

[ADU]

This is mostly for sax512, since he was asking how the Klippel works, and is able to separate the direct from the reflected sounds in a room.

Klippel's Near-Field Scanner vs Anechoic Chamber: Discussion with Christian Bellmann

The video is nearly two hours long, btw, and goes into a fair amount of detail about how the Klippel system works, including some of the things Amir mentioned above. In the first half, Christian describes some of the problems associated with making direct sound measurements, and how the design of their Klippel rig evolved. Then at around 48:00 mins he starts to go into some of the geekier aspects of how the physics, math, and sound modeling works in their setup.

It's interesting, but most of it goes over my head. So I'd probably have to watch and listen to this a bunch more times before I could really start to understand some of how this works. Maybe sax512 and a few others out there will be able understand it though.

 

[AdamG]

Thread notice: Please try to keep your posts related to the Thread subject:

“Master Complaint Thread About Headphone Measurements”

Please and thank you

 

[Deleted member 16543]

Let's measure headphones.
We need a target curve.
What does that target curve look like?
Some believe it should be based on what our eardrums perceive as a balanced sound when listening to speakers.

That's how we ended up talking about speakers.
It's all related, folks. The common denominator being our own eardrums!
Over and out.

 

[AdamG]

Let's measure headphones.
We need a target curve.
What does that target curve look like?
Some believe it should be based on what our eardrums perceive as a balanced sound when listening to speakers.

That's how we ended up talking about speakers.
It's all related, folks. The common denominator being our own eardrums!
Over and out.

It’s called Thread Drift. It happens all the time. A little is expected and acceptable. However, when the Drift becomes the dominant topic of discussion it’s my job to redirect the conversation. If you wish to continue this line of drift discussion please start a new thread.

 

[sonhouse59]

Hi Amir,

I know it would be work more to do, and you are doing a lot, but could redo the frequency response measurement after eq so that we could see the difference ?

Regards

 

[John B]

This would be interesting. You could measure how accurate the EQ is as well. A bit meta but maybe useful. ; )

 

[Tadgh]

Could we get measurements with the ear located in different parts of the headphone cup?

Measuring how the performance of a headphone changes when worn differently provides really useful information. Some headphones (K240s etc.) have dramatically altered performance when you're outside of their "sweet spot"

Furthermore, information on how headphones perform with a partially / fully broken seal would provide useful insight into a driver's resonant frequency, as well as indicating how a given pair will perform for users with imperfect seal (due to fit issues, glasses, etc).

 

[Jabinho]

Please post all generic (non review specific) complaints, comments, etc. in this thread.

In my opinion these headphone reviews are quite useless. I don't like harman curves as reference curve. There is just too much low end for my taste.

 

[John B]

In my opinion these headphone reviews are quite useless. I don't like harman curves as reference curve. There is just too much low end for my taste.

I feel the same about the high end on the Harmon curve, but it gives you a fixed reference to anchor your personal preference curve to, and a general idea of headroom to noodle on things in EQ relative to distortion introduced to get there.

 

[Jabinho]

I feel the same about the high end on the Harmon curve, but it gives you a fixed reference to anchor your personal preference curve to, and a general idea of headroom to noodle on things in EQ relative to distortion introduced to get there.

Correct me if I am wrong. Harman curve is created based on opinions of 250 random people from Connecticut.

 

[amirm]

Correct me if I am wrong. Harman curve is created based on opinions of 250 random people from Connecticut.

Consider yourself corrected. This is just one of many studies:

Where do you get stuff like that? If you are going to put down the research, least you can do is read it first hand.

 

[amirm]

In my opinion these headphone reviews are quite useless. I don't like harman curves as reference curve. There is just too much low end for my taste.

Measurements are independent of the target curve. I don't know why you are mixing the two. That aside, what do you think the situation is? Wild west where anything goes?

And how do you know you don't like it?