Headphone Measurements: Part One, Prolog.
Dec 10, 2007 at 4:36 PM Post #31 of 40
Quote:

Originally Posted by janderclander14 /img/forum/go_quote.gif
Could it be possible to get data measurements for other headphones? Thanks!


I'm afraid Joe's gotta do other things. But, all good things come to those who wait. We'll get through these discussions and, hopefully both y'all and I will learn a lot. Then when we measure the rest of the cans, we'll have a nice, comprehensive set of data for you to do your comparisons.

Til then, the enjoyment of your cans won't be one lick less for not knowing the technical details.
 
Dec 23, 2007 at 8:45 PM Post #32 of 40
Tyll,
We haven’t talked yet but I want to introduce myself. I go by the blog name of Professor Thump and I am one of the principle engineers at Klipsch for headphones.

I want to commend you and your team for this subject on headphone testing. This topic is quite helpful as a platform for informing not only the general enthusiasts, but also the experts on the issues at hand for acoustic measurements with Circum-aural and In Ear headphones. I find that there are some troubling aspects of comparing headphones to loudspeakers, especially when it comes to measurements. This is analogous to attempting to do a listening test between a loudspeaker in a room and a headphone in/on your ear. To say the least it is not easy when considering reverb and HRTF issues and how they translate to your inner ear.

It is equally troublesome that ear emulators such as a 711 coupler or Zwislocki ear simulator does not appear to be accurate above 8 kHz from my experience. I find that if you get the sonic signature correct below this band you are 95 complete for getting the correct response.

When most people review compensated response curves I think they will be confused because they will want to compare this to the anechoic response of a loudspeaker. The fact of the matter is that you need to include the transform of the room to compare a loudspeaker to this compensated headphone response. But what room transform should be used; an IEC 268-13 standard listening room or ITU-R BS.1116? If so what size? When I started to define Klipsch headphone products I realized that it was not trivial deciding what the target response should be to meet the Klipsch Sonic Signature.

That being said headphones are a wonderful device for conveying the absolute best examples of sonic clarity.
 
Dec 26, 2007 at 6:14 PM Post #33 of 40
Hiya, Thump! Welcome to the conversation!

Quote:

Originally Posted by Prof. Thump
It is equally troublesome that ear emulators such as a 711 coupler or Zwislocki ear simulator does not appear to be accurate above 8 kHz from my experience. I find that if you get the sonic signature correct below this band you are 95 complete for getting the correct response.


That's my experience as well. Too many modal resonances in the ear canal at those frequencies I suspect. Etymotic, however, does seem to be able to measure the last octave quite well, though I think they've got some proprietary gear there. But I agree that it's likely that if you've got everything going on right up till 8k you can just let it run out up the next octave and assume things are good. I still have to make the observation that the Etys seem to get the highs just a little bit 'righter' than others.

Quote:

Originally Posted by Prof. Thump
When I started to define Klipsch headphone products I realized that it was not trivial deciding what the target response should be to meet the Klipsch Sonic Signature.


Well, I very much look forward to your thoughts (as much as you can give) as we progress along.


Also wanted to go back and make a little comment on this:

Quote:

Originally Posted by machead
Variations in perceived bass level have a lot to do with the quality of the acoustic seal between the headphones' pads and the contours of the wearer's face. It is one of the reasons why it is difficult to correlate measurements made on a dummy head with one's own experience.


You're quite right, machead, the dummy head we use is halfway between the average male and female sized head, as a consequence, since headphones seem to be designed around a male's head size, headphones are often slightly large for the head and getting a good seal can be touch and go at times. The problem we have is that clamping the cans to the head to get a better seal will likely change the sound somewhat. We'll have a look at this in a later section, and maybe even try to clamp the cans to the head and see if we can't see the sonic effects of so doing, and maybe even develop an appropriate clamping mechanism.
 
Mar 10, 2008 at 5:24 PM Post #34 of 40
Hi Guys! Just letting you know I haven't forgotten about this thread. We've been working on our exact methodology for headphone measurements and have been making strides. Unfortunately our Audio Precision tester died and will have to be shipped off for repairs.

However, the real holdup is time, what with getting the new power amps and speaker stands out and putting in a new bookkeeping system I've simply been buried. I will get to the next chapter here before too long, I promise.
 
Sep 14, 2008 at 2:16 PM Post #35 of 40
+1 for IMD tests. Ideally with the two tones being something useful like 70hz/2khz. I certainly think this measurement is relevant to headphones, possibly even moreso than it is to loudspeakers due to the fact that headphones, in most cases, use a single full range driver, which ofcourse means the bass is modulating the mids and highs even in top tier cans.

This test sample, for example, when played at high (but reasonable) volume, shows even to the untrained ear that a bass kick will cause IMD that's easily audible in the mids. I've heard it happen on everything from HD650s to GS1000s to Edition9s to H2s. Only in multi-way speakers with a seperate bass woofer does this effect not occur.

Also, if you're feeling really generous, something akin to a waterfall graph for headphones would be great
smily_headphones1.gif
 
Jan 24, 2009 at 5:28 AM Post #38 of 40
Sigh ..... the number of times I've tried to make time for the next installment. It's coming. We're working hard on beating the bugs out of the new measurement data.

I can only say it's got to happen pretty soon as I want to run these threads in parallel as I write the new measurement copty for the website. And that's got to happen soon.
 
Jan 26, 2009 at 4:41 AM Post #39 of 40
Tyll,
Would it be possible, or relevant to see what effects your different level of amps have on the headphones response graph? Example, how does you maxed out desktop "portray" the "whatever" headphone versus a lest costly system like the standard Micro Amp using the same headphones. Like I said, is this even possible to graph and display and is it worth doing?
Kevin
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Back
Top