Thank you very much for the detailed reply, Resolve. And happy July 1st or 4th, whichever is most applicable in your cases. I wanted to give your 6 series comparison video another look before responding to the above and maybe givin a more in-depth take on some of this, because there’s alot of ground covered in this video, in a relatively short space of time.
I also like to quibble about some of the technical stuff. But overall, I think you and DMS are doin a pretty decent job on most of your measurements and reviews. And I find myself agreeing with more of your takes than not. I hope both of you will continue to post your raw/uncalibrated/uncompensated measurements though, since I prefer to use my own methods of compensation and reference curves, rather than the ones Headphones.com is currently offering.
For me, the most logical baseline or reference point for a headphone’s FR is still the in-ear FR of a pair of good neutral speakers in a typical semi-reflective room. And I hope that you, DMS, or some other kind soul at Headphones.com will make a run at this at some point, so we can see how this type of in-ear response compares to the raw headphone measurements. I know some on your staff think DF is the best way to go. But for me, it’s just a stepping stone to get a little closer to the above, until some better in-ear measurements of speakers are available.
I have some gripes about the way y’all are doin DF compensation as well, but will save most of those for another time or place. I like the results I’ve been gettin recently with HBK’s original 1/3-octave 5128 DF measurement though. And would like to try posting a few examples of this in another topic here, maybe in the Technical Forum.
The comparisons of new and worn earpads was very interesting. I wasn’t expecting that much variation in the before and after on these headphones. But my experience with other headphones has been similar. As the pads wear, and become more compressed (and your ears get closer to the drivers), the FR seems to get warmer sounding. This may not be true for all headphones though.
I’ve been wondering about the potential influence of pad flap on some of the closed back measurements as well. So I’m glad you threw in the slide of the HD620S with increased clamp for comparison. And hope this is something you’ll continue to look at, and try to understand better. I know you need fairly high dBs when doing measurements to filter out the background noise. But I wonder if that could be adversely (and maybe unfairly?) effecting the bass measurements on some closed-backs, since it may not necessarily represent the responses at more typical listening levels. And I seem to be noticing a bit less mid-bass emphasis in some of Jude’s (and maybe also the Sound Guys’?) closed-back measurements by comparison.
Tyll put his old Head Acoustics rig in a small insulated box when doing his headphone measurements. And I wonder if that might be something to consider as well. Seal or leakage issues around the 5128’s cheeks might be something else to look at.
On the subject of HpTF variation, I think you’ll have to explain what this refers to a little more, and why you think it’s important… I expect headphones to measure differently on every head, ear, and ear canal. This is more of a problem, imho, with IEMs than with over-ear headphones though (even though some of the technical research seems to contradict this), because an individual’s HRTF interacts much less with an IEM, making it harder to filter out the HRTF effects, and produce a reliable, predictable result. This is why I only use over-ears for my critical headphone listening.
I still expect some variation with over-ears. Assuming we all fall within a more or less normal range of human physical characteristics though, compensation should take care of alot of that. That is really the point of compensation, in fact! To filter out the individual effects of a person’s or rig’s transfer characteristics, so that measurements on two or more individuals or systems can be more easily compared. This is one of the main reasons we do compensation, or “calibration”, as you refer to it in the video.
Re soundstage, since this is largely a subjective effect, I think it’s difficult to actually pin down precisely what’s responsible. I think Tyll and probably also some of the more informed developers and subjective listeners in the hobby would probably say that it’s a bit more involved than just a midrange dip and some extra treble though. This might also be a good topic for another thread though.
I think that a slight dip in the upper midrange can potentially improve the timbral and tonal balance of some recordings though, because the speakers used for mastering recordings will often have a dip or depression in their in-room response due to cross-over and directivity issues in the same frequency range. A good example of this is the Adam T8V…
https://www.spinorama.org/speakers/Adam%20T8V/ASR/index_asr.html
The in-room response of the T8V has a fairly pronounced dip at around 2 kHz in the upper midrange. And it also rises up fairly significantly in the low treble before descending back down again in the higher frequencies. To properly decode a recording mastered on a speaker like this, you’d need a headphone with a similar dip and rise in the same ranges. This might improve the perception of soundstage as well. But it may only be because the headphones are doing a better job of matching the tonal and timbral balance of the transducers used to produce the recording.
The Adam is a somewhat extreme case, but there are many other well-regarded speakers with a similar dip or depression in their in-room/off-axis response, at the cross-over of the midrange and tweeter drivers. So this is not an uncommon thing. Recent speaker designs (with DSP and improved cross-overs) are beginning to address this issue though. And getting much closer to acheiving a linear in-room slope from F0 in the bass to the treble.
