I have a couple of questions regarding amplifiers.
Could it be that Sinad transfer poorly because its based on sinus tones and music does not. Its simply not what we hear when we listen to a product.
Could IM distortion (maybe two pairs) be a better way to compare as it more resembles music?
Mike
This isn’t the issue so much. We sometimes see a critique of measurements in general that “but test tones aren’t music” (not suggesting you’re saying this), but at least with headphones you can in fact measure them with music and get the same FR result. Test tones just give you a more precise way to test a product’s behavior.
With that said, from my understanding, there is something to the issue of identifying distortion products based on a single tone stimulus, where the rest of the spectrum isn’t being considered. I believe this is something @GoldenSound goes into in the video.
This is definitely relevant, and a good thing to test, and for that you’d want to look at the IMD results, not SINAD.
With source equipment, we recently found an example where IMD was highly relevant (this was one example where I was very wrong in my initial assessment), where the type of content being played was highly relevant for one of the amp’s functions.
While perhaps not as relevant with headphones, for any active designs that make use of volume-dependent dynamic equalization (like the Apple ones for example), we use an M-noise test, which is meant to better emulate a musical stimulus. We do this because depending on the crest factor of the music the headphone performance may change for those types of devices.
Is that from this particular video about measurements? Are there any other sources or articles about it? Seems like an interesting topic.
Thanks
I also like to suggest that spectral decay plots say a lot more - though not all - than a FR plot and and better reflects how we perceive headphone sound.
While we have no best practice for this that could possible over time be agreed upon.
One example done by purr1n
Mike
Nope.
This is the horse that is dust at this point, and we should probably put it to the other thread. But I’ll say it again for folks who haven’t engaged with that yet:
Headphones are minimum phase systems 99% of the time (except when something is super wrong), which means time domain information is proportional to FR. Fix the resonance in FR with EQ and the ‘decay’ goes away with it.
In other words… this is just a worse view of frequency response.
And yes, I’m sure this will come up again.
It would be very helpful if you could demonstrate that in a video or with some pictures in a forum thread.
I see you and others posting this explanation time and time again, and I believe you because that’s been my experience EQing headphones as well, but I don’t have a measurement rig to demonstrate this effect on a CSD graph.
“It is immensely frustrating to me that certain audible effects like sound stage, imaging, “punchiness” “tactility” continue to be described, heard, but not quantified”
If it helps, I’ve found some subjective commonalities with the tactility you mentioned as I love that feeling and have chased it in many headphones.
Again this is all subjective, feel free to correct me!
This actually exists on head-fi already… Unfortunately I don’t know where the post is but I’m sure @Mad_Economist can dig it out.
But we featured that in this video:
Thanks! That’s exactly what I needed.
@Resolve Thanks, I forgot I’ve seen that in your video already. My mind’s like a sieve with one inch holes.
Interesting list. Do you have any particular small sections of music that’s particular “tactile” which are easily isolated? I’d like to listen to that back to back with my various headphones.
Sound transducers merely move air back and forth, they’re not one way pumps, so I don’t think flow measurement would make sense.
It seem to make some logical sense at first glance, but since headphones are minimum phase system, isn’t the “speed” simply treble extension ability up to the highest audible frequencies?
Would like to circle back to my original request of small sound bites that demonstrates tactility so we could have a more meaningful conversation. Because on this point I find my experience deviate from yours.
That is certain a sensation I have experienced, air seemingly being “moved”. However I’m not entirely sure if this is a separate measurement from FR graph? If it is, what is it?
+1
Mike
Some interesting takes on MDAQS.
If MDAQS is based at least in part on frequency response, then I wonder if a headphone’s MDAQS score could be changed/improved by simply applying some EQ. Might be an interesting test to try.
Maybe there are some other similar ways that Headphones.com could also rate certain technicalities though for audiophiles who prefer to think in these terms?
I thought the video was pretty good overall btw.
I wish Headphones.com had some different approaches though to FR. Because the current approaches just aren’t workin for me, except for some of the raw measurements.
As I mentioned above, the only objective reference point that really matters to me at this point is the in-ear response of neutral speakers in a semi-reflective room. And if there isn’t a way for Headphones.com to more precisely measure and show that on their graphs, then it would be helpful if there was a system to make it easier for users to apply their own target response/compensation curves to the measurements.
This is probably off-topic, but since I don’t have the $$ for an accurate measurement rig to measure my own headphones, I also wonder if Headphones.com or its affiliates might be willing or interested in offering this as a service to headphone users, for a fee.
They showed up at CanJam Dallas with DMS’s 4128 rig and did a bunch of measurements (for free!). There is talk of that becoming a thing at future US CanJams.
The thing is we know that this isn’t actually the ideal response. That’s why Harman has separate research and overall preference curves for headphones vs speakers for example rather than just measuring the harman speaker curve in a room on a HATS and calling it a day.
With headphones we don’t have any reflections and so things are different.
I think generally speaking, if you have a pair of ‘relevant’ headphones that you’d like to see measured, any of us would be more than happy to measure them for free if you’re willing to send them in. Or if you bring them to a CanJam, I believe we are going to be bringing measurement rigs to these events more in future as it worked quite well at CanJam Dallas.
I’m also more than happy to measure any piece of source gear if someone is willing to lend it.
Measuring does take time though, so if the headphones are something more esoteric it’ll depend on available time, but do reach out!
Anything with heavy bass will be easier to feel, but certain headphones have more feeling even in the mid range. In ‘Polyphia - Playing God’ with certain headphones/iems I can feel each pluck. I’ve noticed this with xenns top, focal clear, heddphone one, arya organic, the composer, noble spartacus. Especially if low sub bass is EQ’d up by a couple DB. I’ve also tried this with sundara, arya, he1000, hd650, hd599, dca e3, hd800s, px8, bathys, radiance, stellia, maxwell, auteur w/leather … No tactility in those that I noticed
“Isn’t the “speed” simply treble extension ability up to the highest audible frequencies?”
I don’t think the speed of continued vibration i.e. high freq is the same thing I’m describing, maybe inertia is a better word for it. I don’t want to overcomplicate things but AMT drivers for example move very quickly, they expel air at a speed you can feel with the initial force of the driver. You make a good point that they aren’t 1 way pumps, I guess CFM doesn’t apply.
It definitely doesn’t show up on an FR graph, at least for notable ones I’ve tried. Sure most of these are more on the bassy side, but no more bass than others that didn’t. I’ve noticed they’re all open back or vented, they have decent sub bass extension, and they’re all fast drivers. It may just be a side effect of sub bass colouring all the notes in the spectrum, but few drivers having the capability to have that clean sub bass extension
Great song to test this. I’m trying to narrow this down to shorter sound clips to make it easier to put on repeat and test. Of the two sections, which one would you say you feel this tactile pluck effect more? 0:18 - 0:23, or 0:57 - 1:02?
Appreciate the reply on this, GoldenSound. And your invitation to get more involved in discussions of measurements.
Things are different. But they’re also sort of the same, according to the folks at Harman and their research. Speakers will of course measure differently in a room than they do in the ear of a mannikin. And you can look at and consider both their on and off-axis responses in assessing them, which is something you can’t really do with most or all headphones.
Still, in interviews, conferences, promos, etc., the folks at Harman (like Dr. Olive) are much more apt to stress the strong parallels and throughlines in their research on loudspeakers and headphones. And the parallel between the in-ear response of neutral speakers in a semi-reflective room and the preferred in-ear response of headphones that they found in their research. I’ve posted a couple examples of this below for your and others perusal.
According to Dr. Olive, roughly a 2/3 majority of the listeners in their early studies preferred an in-ear response in headphones close to the in-ear measured response of speakers in a semi-reflective room. The remaining 1/3 also preferred the same type of response with a little more or less bass or treble, depending on their age, testosterone levels, and hearing loss.
Another video featuring Dr. Olive and Omid Khonsaripour of Harman that draws comparisons between headphones based on their target and the in-ear sound of speakers in a room.
Will do. And I appreciate the offer on this as well, Cam. (Though maybe you’ll want to retract it after my last post above. 
) Good measurements add value to a headphone imo. So I’m happy to pay something for such services for the few headphones I’d like measured.
Maybe this is also something Headphones.com would consider offering as an add-on to the headphones they sell.
Another recent interview where Dr. Olive describes the strong relationship between the in-ear response of neutral speakers in a semi-reflective room and preferred in-ear response of headphones. The audio isn’t too good on this btw. So I suggest turning on the closed captions.
