Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Now that Denon headphones have gone to shit, what is the next logical  (Read 22866 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Now that Denon headphones have gone to shit, what is the next logical

Reply #25 –
That's just the typical "Beyer highs". Not quite my taste either. They've only got a few models that don't features these, like DT250 (once described as a "closed HD600") or DT150 (which even is on the somewhat dark side of things). Older samples of their open models (like '80s/'90s) are supposed to be a fair bit darker, too.

Quote from: Palladium on
The only headphone audio gear reviews I remotely trust is Sonic Sense where they rerecord the speaker/headphone outputs and compare them to the original source track, still sighted but at least still a far more robust methodology than the typical placebophile nonsense.  Even then, my first hand experience is that the LSR305 sounds incredible but the ATH-M40X sounded like complete ass despite both of them staying super faithful to the source track in their testing.

Most probably their headphone testing setup is no good then. I actually found this discussed here. While picking up the sound of speakers with mics in a halfway realistic way is quite doable, I don't see how that should work for headphones. They're playing sound through the headphones under test which interacts with the dummy head, is then recorded, is then played back over your headphones interacting with your head. Sure doesn't sound right to me! Their "how we measure" video (see previous link) is just blinding you with science... or as many numbers as possible, rather.

What I guess they should be doing is reverse the recording setup, putting the sound source inside the dummy head and recording with the headphones under test. This is not without its pitfalls, of course. As most headphones are designed to be driven from a low-impedance output, you need a custom "headcrophone" preamp having a low input impedance, i.e. current input - the classic Hiraga MC prepre would be one example, which scaled to 1 ohm or less would probably mean using a medium power transistor at Ic > 26 mA (this is because r_e ~= 1/gm ~= 26 mV / Ic). And as for wideband omnidirectional loudspeakers, uffff. I guess piping something through a small hole should work, but getting rid of standing waves behind the baffle may not be easy.

Even then, a dummy head will never be interacting with headphones in the same exact way that your head does. Plus I'm still not sure whether the effect wouldn't be exactly the inverse of what you want (would have to think about it some more), but then you could still derive an EQ / impulse response from that and apply the inverse (see deconvolver) to the music material, so not a total deal-breaker at least.
Hmm. Maybe that's exactly what they should be doing with their present setup, actually. Reciprocity and all.

No, that still doesn't seem quite right. Like this you're getting H_yourcans(f)/H_DUT(f) on the music material, while what you want would be the exact inverse. IOW, to do this right you would have to convolve the music material with the inverse of your headphone + ear transfer function first and play this on the DUT cans on the dummy head where it is then recorded. No, wait, it would have to be the inverse transfer function of your headphones + dummy head TF. Gah, I should just have done the math properly. Anyway, then the dummy head TF should cancel upon playback, likewise the one of your cans, leaving only the DUT cans' TF + your ear transfer function.

Now obviously it's a tad impractical if you have to send in your cans first (unless another very similar sample of the same model has already undergone response recording). Plus as said earlier, a dummy head will never be interacting with headphones in the same exact way that your head does, so the two dummy head TFs will not be exactly the same.

(Please excuse the braindump, it's late and I wasn't planning on doing this.)

Measuring headphones enjoys the reputation of being a mess, but I'm afraid that trying to reproduce the sound of another headphone is an even bigger mess. The Harman guys are doing something like this though, by means of EQ if memory serves. It works for them because they only have one model headphone for playback (HD518) with correspondingly well-known characteristics, and they're only aiming at ballparking the response, which is quite effective as-is.
Quote from: Palladium on
(BTW, why its is ALWAYS the pro-audio guys who are doing the sensible things?)

Pro audio users have more at stake and less money to burn?

That being said, hi-fi reviews used to be much more rational, too. But that was back in the '70s. Then they discovered that there was more money to be made with audiophools...