He just retweetet the video and I think he agrees hah

Yo its DMS from headphone show 👿😱😲😲🙀

Thanks! Was hoping I didn't misrepresent anything.

Do you mean a headphone that measures to harman? Or a loudspeaker? I'm really not the best at this, so someone else would probably be better. So take this with a grain of salt. I could be wrong on this I guess if done right, a loudspeaker could work, although I believe you'd need to basically recreate the Harman listening room and their in room response. Diffuse field is more consistent and able to be replicated by anyone (with the knowledge and tools) on any rig Having a headphone that measures exactly to harman and using that as a way to recreate the target is a lot more tricky as then you're dealing with the difference in acoustic impedance of both the headphone and ear simulators, as well as the differing transfer function. So if you had two headphones that both matched harman on the Gras, but they had a different acoustic impedance due to their drivers and build. When measuring on another ear sim, they could measure quite differently. I'm curious about the 5128 standard. If it will be adopted, or if more of the hybrid approach that headacoustics will catch on where they maintain the 711 coupler, but with the hybrid pinna

gizaudio, the harman target for the 5128 was created by measuring many different headphones at different positions on the GRAS and 5128, and applying the error curve to the GRAS harman target. Some people who don't know what they are talking about claim this is bad since the preference research did not begin with the 5128, but Harman has always been a statistical average. The point of the 5128 Harman target is for there to be no statistically significant difference in preference ratings between GRAS and 5128, and this was achieved aside from possibly a few outliers.

Timmy!

@@Sharur1 problem is that the 5128 harman target calculated this way is not released by harman, but the diffuse field measurement for the new rigs are released by the manufacturers.

Yeah I noticed my mistake after realizing I posted it on meme Monday. And yeah, the whole v-tuber thing definitely will put many off. But if I do eventually gain more of an audience, maybe it will be ones who wouldn't normally click on a headphone video. Either way, we'll see how it goes! Thanks!

Yes that's the point. Measuring the DF HRTF response of the different rigs so that this method can be applied to it. I only used the KEMAR DF HRTF since it's the closest thing to my pinna which was made from the mould of the Gras KB0065 pinna from the KEMAR, but for the 5128 and other rigs they will need their own individual DF HRTF measured for this to work. While B&K do have their own DF for the 5128, there are some in the community who are doing their own DF HRTF measurment to ensure the best results.

Downward tilt sucks too. Just stick to flat bass... why is that so hard?

blame harman themselves. why don't them release it? you can tilt the accurate diffuse field targets in no time but you have to do this inaccurate and hacky digitization of an image to get your harman 5128.

Yeah, unfortunately I live with others and was trying to be more discrete

Yeah I know. He wasn't the first to do it this way. Harman internally created a new target using known headphone data and comparing for a quick target, as well as the community a long time ago figured out a Harman target when 5128 measurments were being made by other sources and comparing with posts by Sean. I believe soundguys target is also done this way. The issue is doing it this way does still have some flaws, and it's still not reliable to just use another headphone as a refrence to recreate the target since headphones perform different on different heads, so it can lead to inaccuracies doing it this way, and using multiple headphones like this, you may end up with an average you can use but its not ideal. While Harman did this at the beginning to make a quick target, they also understand the issues and will likely recreate the target from the ground up (if they haven't already, although Sean said they will not release it to the public this time). As well, by making the target this way, you are able to keep the high res information in the upper mids and treble that was lost in the 2018 Harman target due to it being 1/2 octave smoothed. This is also meant to be bigger than the 5128. If it was just about the 5128, than these methods could be fine for now. But the goal is to make a reliable standard that could quickly and reliably be recreated on anything. B&K, Gras, Headacoustics, or even another HATS or clone gear if someone so desired and had the means to get a DF HRTF. Simply making a Harman target for the 5128 isn't the end goal, it's making a universal target standard and methodology

You mean "the guy who says everything is terrible except his DT880 and Etymotics" Sharur?

​@@AliArghavan he may be wrong at sometimes but he got the spirit. He's far better than pure subjectivists who thinks cables, dacs and amps changes sound dramatically. I didn't came across where he said DT 880 is good, which in my opinion they're very bad. Also nowadays he doesn't tell Etymotics are good.

@@sidesaladaudio the Harman 2018 target is not anymore smoothed than the flat in-room HRTF created in 2013 which was already very similar to diffuse field. A problem with the 5128 is that the measurements are literally wrong by showing high leakage in the bass and a huge 8 kHz spike compared to real humans. Harman's over-ear target for the 5128 makes full sense if you understand just how useless the 5128 actually is. Since the std dev from actual humans is so high, the 5128 target would at best still be less accurate for predicted preference than the target for the GRAS. The incorrect HRTF problem of the 5128 becomes far worse for IEMs which bypass the incorrect outer ear of the 5128. The diffuse field HRTF of the 5128 is ridiculously bright compared to the DF HRTF of the average human, so compensating for that is very misleading on the 5128.

Owners of the DCA Stealth like me will agree; it's a great closed-back planar, but it has plenty of flaws (way too bright) and needs EQ to be comparable to any well-targeted IEM. I wouldn't use the DCA Stealth FR curve for anything other than forming an EQ correction curve.

Headphone sound isn't localized in front of you, though! When was the last time you put on your headphones and perceived the sound source as in your room the same way you do with speakers? If it's happened at all, odds are it was with spatial content (binaural recordings or HRTF-based post-processing) as opposed to normal stereo. Gunther Theile's original argument against the free field target - which every subsequent subjective test has found to be abhorrent to listeners - was that it would be timbrally incorrect because we do not perceive headphone sound as coming from in front of us, but rather from "inside the head" or "nowhere" (in-head localization). This is true for the same reason that a tiny adjustment to your head's angle doesn't vastly change the timbre of your speakers' sound, even though the eardrum FR does change quite substantially: your brain separates the "sound of the direction" and the "sound of the source" quite efficiently, in effect applying an inverse EQ for the perceived directional HRTF.

@@blainelacross5311 the first step is to stop calling magnitude response frequency response. A HRTF includes both magnitude and phase information, but the latter is often ignored.