[mrtim6]

Thank you arnaud

 

[sander99]

How does that even work without manLOUD in terms of spatialization? I was under the impression that HPEQs won't really make that much of a difference alone in that regard, and that they primarily are used to remove some of tonal characteristics of the headphones, so that you are left mostly hearing just the speakers. I thought manLOUD is what helped more to somewhat normalize PRIRs from other sources.

Anyhow @Litlgi74 insisted I gave him my HD800 HPEQ without manLOUD.

Normally you create a manLOUD for and using a specific PRIR.
If you have one of @Litlgi74's PRIRs on your A16 then you can make a manLOUD for that PRIR, and probably you should (at least to try if you get an improvement).
But with the demo audio fragments you can not make a manLOUD.
Listening to the demo audio fragments using a manLOUD created with yet another PRIR may be worse than using an autoEQ HPEQ.

 

[Richter Di]

Normally you create a manLOUD for and using a specific PRIR.

Really. I was not aware that this is the case. It is hidden somewhere in the handbook?

 

[arnaud]

Really. I was not aware that this is the case. It is hidden somewhere in the handbook?

You are doing a equal loudness correction of the HPEQ but using a specific PRIR to do so (typically the front channels or possibly the center channel?), it’s strictly just a correction for this PRIR/HPEQ pair but I guess the gist is that it helps for all other speaker headings as well.

from the you tube video:

and the key:

This looks very similar to thus crude compensation method for headphones described in this video:

 

[arnaud]

This one is a really interesting talk by the same researcher who explains that, for realistic virtualisation through headphone, the key is to have a tonally correct reproduction of front channel hrtf, not just at the blocked ear canal entrance but at the ear drum…

In my understanding, this is the reason why even PRIR/HPEQ measured with own head sound off in the mid/highs because we’re ignoring ear canal amplification effect (a highly individual type amplification with 20dB boost differences between subjects at some frequencies, not something insignificant ).

When doing the manLOUD Eq, I think you’re simply dialing these back in using subband equal loudness test procedure to correct for the inaccurate blocked ear canal measurements. Maybe @jgazal or others fascinated with the subject may correct or add comment here.

 

[arnaud]

Interesting to note also that the conclusions are in line with some of us in the thread who feel the head tracking is a bit of a gimmick and all it really takes is a tonally correct hrtf measurement in open ear canal conditions (e.g for me, head tracker is off as I can’t stand the tangling wires but I found the manLOUD correction critical to achieve a solid virtualization):

 

[Richter Di]

@arnaud Thank you. Super helpful, as I completely forgot about the manLoud being played through the PRIR.
And the rest super interesting. Again. Thank you!

 

[Om ma Ni]

Interesting to note also that the conclusions are in line with some of us in the thread who feel the head tracking is a bit of a gimmick and all it really takes is a tonally correct hrtf measurement in open ear canal conditions (e.g for me, head tracker is off as I can’t stand the tangling wires but I found the manLOUD correction critical to achieve a solid virtualization):

One interesting point: according to them, they concluded that on-ear and insert phones are way better than open hp....bit of a concern for us as the A16 has been tuned mainly by open hp like the HD800 etc...

 

[Sanctuary]

One interesting point: according to them, they concluded that on-ear and insert phones are way better than open hp....bit of a concern for us as the A16 has been tuned mainly by open hp like the HD800 etc...

Yeah, I was wondering about that, and they must be talking about a different application, because closed headphones and IEMs (those I've been able to use at least) sound nowhere near as realistic to me, and sound completely claustrophobic by comparison to my DT1990s and HD800s. I get what actually sounds like a center speaker as it should be, localization anyway with either of those, yet it always sounds like it's inside my head, or pressed right up against my forehead when using closed-back headphones, and even worse with IEMs. With actual binaural recordings though (like the virtual barber shop), it sounds fine with both closed and IEMs.

That guy sounds just like John Malkovich BTW. Even looks like him a little too.

Incidentally, in the video he's talking about how you need to be able to localize with your eyes closed, and not have the eye/ear connection happening for it to actually be working, and whenever I did a PRIR for my friend the first time, I specifically told him to close his eyes and tell me if it sounded accurate. He felt like it wasn't needed, and that the sounds were coming from where they should be. But he was able to look at the speakers at the same time. Later, when we left the room with the speakers, it wasn't as convincing for him, which is exactly what I was trying to ensure against. Thankfully, the latest PRIR that we did for him seems to work just fine in a different room from the source.

Also, kind of a tangent, but has anyone here ever used not just one, but two front speakers as a center for a PRIR? Could it even work? It should be easy enough to get a Y shaped analog that could work as the interface between the A8/A16 and the receiver, but I am only assuming that either Realiser would send the sweeps to both speakers at the same time. Also, if it did work like that, would even the slightest discrepancy between the speakers (even when they are the same model and age) maybe be a negative factor, and make it not worth the trouble? I just thought that maybe having two fronts right next to each other might in a way offset any issues with a person's head not being 100% dead center with a single front since those are much more narrow (obviously) vertically than a typical center is when lying horizontally. I managed to get a really great center from a front recently, but it required more than a few tries to finally get one that not only sounded really good, but actually sounded absolutely centered.

 

[GeorgeA]

As part of his Post #11,837, @phoenixdogfan provided a link to a youtube video “about how to create the single unified layout needed to do all three immersive formats.” Other useful resource in that respect is available at the following link:

http://www.grobi.tv/media/pdf/54/36...-Guide_V5-0_englisch_publishigPUdX9hPlgSJ.pdf

It’s about an approach for the height layer of various speaker layouts that can be used with all currently available immersive formats.

 

[Eich1eeF]

[...]
Also, kind of a tangent, but has anyone here ever used not just one, but two front speakers as a center for a PRIR? Could it even work? It should be easy enough to get a Y shaped analog that could work as the interface between the A8/A16 and the receiver, but I am only assuming that either Realiser would send the sweeps to both speakers at the same time. Also, if it did work like that, would even the slightest discrepancy between the speakers (even when they are the same model and age) maybe be a negative factor, and make it not worth the trouble? I just thought that maybe having two fronts right next to each other might in a way offset any issues with a person's head not being 100% dead center with a single front since those are much more narrow (obviously) vertically than a typical center is when lying horizontally. I managed to get a really great center from a front recently, but it required more than a few tries to finally get one that not only sounded really good, but actually sounded absolutely centered.

Sounds like a bad idea to me, since the two speakers will create interference and combine to create a comb filter precisely if you're not at exactly the same distance to both of them. If you want a larger center, use larger speakers, maybe electrostatic or magnetostatic ones. Otherwise just make it easier for the test subject to align his head, by attaching something that works like a sight to the center. That should be more than good enough to get the sound right,

 

[Sanctuary]

Otherwise just make it easier for the test subject to align his head, by attaching something that works like a sight to the center. That should be more than good enough to get the sound right,

This is actually something I was wondering about in the first place. Something like a laser pointer that you could have affixed to someone's forehead, but I have no idea if they actually make something like that. Need something like a dentist's headlamp, but with a laser instead.

edit: Found some!

 

[arnaud]

Yeah, I was wondering about that, and they must be talking about a different application, because closed headphones and IEMs (those I've been able to use at least) sound nowhere near as realistic to me, and sound completely claustrophobic by comparison to my DT1990s and HD800s. I get what actually sounds like a center speaker as it should be, localization anyway with either of those, yet it always sounds like it's inside my head, or pressed right up against my forehead when using closed-back headphones, and even worse with IEMs. With actual binaural recordings though (like the virtual barber shop), it sounds fine with both closed and IEMs.

That guy sounds just like John Malkovich BTW. Even looks like him a little too.

Yes, I also vaguely thought I had seen him somewhere, john malkovich, eh!!

As for his recommendation for using in ear mics, it boils down to consistency. I suppose we should be running into these issues with the realiser and headphones with wide adjustments in seating possible, I don’t think I’ve noticed a big issue with my hd800 though.

Just guessing but I wonder if some of his reasoning is because he’s not doing a proper headphone compensation like the HPEQ inverse filter in the realiser?

 

[VandyMan]

I tried to do a manLoad EQ, but I was not happy with the results. When I'm doing the manLoud, I perceive the higher tones as being much louder, but when I listen back with music it sounds wrong. I think I'm going to try again, but this time I will not touch the EQ above maybe 10K.

 

[Eich1eeF]

This is actually something I was wondering about in the first place. Something like a laser pointer that you could have affixed to someone's forehead, but I have no idea if they actually make something like that. Need something like a dentist's headlamp, but with a laser instead.

edit: Found some!

A simple piece of wood with two nails sticking out, laying on top of the center would enable the subject to adjust position (2D, in the plane parallel to the front wall) and direction. Unlike a laser stuck to their head, it also does not have the effect of ridiculously amplifying small, entirely irrelevant changes.