So some people complaining about sun glare in XP12 ?

[Murmur 3,944]

15 minutes ago, mSparks said:

add both pictures, be clear how recreate them. I could be wrong but pretty sure that is not intentional. it may even be what some people are complaining about the sun glare. Thats why I like pictures and videos.

It was already missing in XP11 though. The sun glare in XP8, 9 and 10 was better.

[mSparks 2,094]

45 minutes ago, Murmur said:

It was already missing in XP11 though.

Not mine, nothing that bad

Its also the complete opposite of everything we have talked about in the thread before now isnt it? a complete lack of sun glare at altitude?

Edited November 25, 2022 by mSparks

[Murmur 3,944]

1 hour ago, mSparks said:

Not mine, nothing that bad

Not the bloom of light around the sun, but the glare/blinding effect on the objects around.

XP11:

XP10:

 

[Bjoern 1,158]

Reminder for the compulsive complainers in here:

 

[blingthinger 649]

On 11/25/2022 at 10:42 AM, Bjoern said:

 

 

I haven't played with these datarefs, but I imagine they could be used to make a "squinting" addon. Bind it to a button and we could effectively squint our little digital peepers for a split second: flex those eyeball muscles a bit to close that eyehole aperture down a couple f-stops.

 

 

[jcomm 5,985]

Well, because a few posts above I mentioned something about the model of temperature lapse in XP12, and namely the Tropopause height and temperature, I would like to add that I was not right about it:

- XP12 when using model data does indeed take into consideration the data regarding temperatures, up to FL390, and then for simplification propagates the temperature profile above that level using ISA deltas;

- It is possible to get lower than -56.5 ºC at higher levels provided the models show those temps for the higher levels.

 

[Biology 1,347]

I'm sorry to break it for everyone, but this has nothing to do with bright sun glare reducing visibility and it very likely is not intentional. Here's what's happening:

Since X-Plane 12, the sky colors (and aerial perspective) are calculated at real time using physically based equations. In short, a light is casted for each fragment (or more loosely, pixel) and then scattered & absorbed light for each main color is calculated. Phase functions are very important in these calculations as they describe how much light scatters to a particular direction. The issue in X-Plane 12 is happens during the calculation of Mie scattering, which is exhibited by relatively large particles, like the aerosols in the atmosphere which go up to an altitude of approximately 2 kilometers, such as mist and pollution. Mie scattering does not depend on wavelength and it's known with its strong forward peak in its phase function. Phase function for Mie scattering also depends on a lot of factors like particle diameter. You can see a polar plot of Mie scattering phase function for various particle diameters below:

However Mie scattering phase function is very hard to calculate in real time and due to how drastically it changes it is not feasible to bake it into an LUT either. Therefore, approximations such as Henyey-Greenstein phase function are oftenly used instead. It's normalized so that its integral over a sphere returns 1, which makes a lot of physical sense, given that phase functions give the ratio of light scattered to a particular direction and more than 100% percent of the light cannot get scattered. You can see a polar plot of Henyey-Greenstein phase function below for various directionality coefficients (analogous to particle diameter) below:

Now here's the issue - there seems to be a bug in the X-Plane 12 so that the Mie scattering phase function returns abnormally high values when the scattering angle approaches sun angle. This overwhelms the tonemapper and clamps everything to white. Not only that, Mie scattering is not the main phenomenon that causes reduced visibility due to sun, it is the sun itself! Sun is extremely bright on its own and even without Mie scattering spreading sun light it is bright enough to overwhelm cameras or eyes looking at it, which is not the case with X-Plane 12, proving this is a Mie scattering issue, not a simulation of the effect of sun's brightness on eyes. While sun glare is partially caused by Mie scattering, it also happens without Mie scattering because of the nature of light itself - light behaves as a wave, therefore even parallel light rays create spreading out light waves when they hit an "aperture", be it a camera aperture or pupils in eyes. Otherwise neither sun glare nor the blinding / visibility reducing effect of the sun would exist in space or at high altitude, but obviously this is not the case in real life. Conversely this is exactly what happens with X-Plane 12, again due to the fact that this is a Mie scattering issue.

Not only that, people who posted real-life photographs posted here as an evidence are missing one important point - eyes have a much higher dynamic range and while sun has a similar visibility reducing effect on eyes, it certainly does not look the way they look in those photographs. If anything, I think those photographs do a great way showing that the issue in X-Plane 12 is caused by something completely different. Here's a random screenshot from X-Plane.org, showing exactly what I mean:

Notice how the overexposed part descends into the horizon and it gets laterally larger as it descends. This is because as I've mentioned aerosols (which mainly exhibit Mie scattering) tend to not go much higher than 2 kilometers. This is not the case for the photographs shown as evidence in this thread, because as I've mentioned it is mainly the sun's brightness itself that causes visibility reduction.

Regardless, this issue has been known since X-Plane 12 was in early testing, it is acknowledged by Laminar Research and it will be fixed in the near future. It is not an intended feature and while there likely are plans to properly simulate visibility reduction by the sun, this is not one of them. It is not a feature, it is a bug that requires changes to the approximation for the Mie scattering phase function and the tonemapper.

Edited November 27, 2022 by Biology

[UKflyer 630]

13 hours ago, Biology said:

I'm sorry to break it for everyone, but this has nothing to do with bright sun glare reducing visibility and it very likely is not intentional. Here's what's happening:

Since X-Plane 12, the sky colors (and aerial perspective) are calculated at real time using physically based equations. In short, a light is casted for each fragment (or more loosely, pixel) and then scattered & absorbed light for each main color is calculated. Phase functions are very important in these calculations as they describe how much light scatters to a particular direction. The issue in X-Plane 12 is happens during the calculation of Mie scattering, which is exhibited by relatively large particles, like the aerosols in the atmosphere which go up to an altitude of approximately 2 kilometers, such as mist and pollution. Mie scattering does not depend on wavelength and it's known with its strong forward peak in its phase function. Phase function for Mie scattering also depends on a lot of factors like particle diameter. You can see a polar plot of Mie scattering phase function for various particle diameters below:

However Mie scattering phase function is very hard to calculate in real time and due to how drastically it changes it is not feasible to bake it into an LUT either. Therefore, approximations such as Henyey-Greenstein phase function are oftenly used instead. It's normalized so that its integral over a sphere returns 1, which makes a lot of physical sense, given that phase functions give the ratio of light scattered to a particular direction and more than 100% percent of the light cannot get scattered. You can see a polar plot of Henyey-Greenstein phase function below for various directionality coefficients (analogous to particle diameter) below:

Now here's the issue - there seems to be a bug in the X-Plane 12 so that the Mie scattering phase function returns abnormally high values when the scattering angle approaches sun angle. This overwhelms the tonemapper and clamps everything to white. Not only that, Mie scattering is not the main phenomenon that causes reduced visibility due to sun, it is the sun itself! Sun is extremely bright on its own and even without Mie scattering spreading sun light it is bright enough to overwhelm cameras or eyes looking at it, which is not the case with X-Plane 12, proving this is a Mie scattering issue, not a simulation of the effect of sun's brightness on eyes. While sun glare is partially caused by Mie scattering, it also happens without Mie scattering because of the nature of light itself - light behaves as a wave, therefore even parallel light rays create spreading out light waves when they hit an "aperture", be it a camera aperture or pupils in eyes. Otherwise neither sun glare nor the blinding / visibility reducing effect of the sun would exist in space or at high altitude, but obviously this is not the case in real life. Conversely this is exactly what happens with X-Plane 12, again due to the fact that this is a Mie scattering issue.

Not only that, people who posted real-life photographs posted here as an evidence are missing one important point - eyes have a much higher dynamic range and while sun has a similar visibility reducing effect on eyes, it certainly does not look the way they look in those photographs. If anything, I think those photographs do a great way showing that the issue in X-Plane 12 is caused by something completely different. Here's a random screenshot from X-Plane.org, showing exactly what I mean:

Notice how the overexposed part descends into the horizon and it gets laterally larger as it descends. This is because as I've mentioned aerosols (which mainly exhibit Mie scattering) tend to not go much higher than 2 kilometers. This is not the case for the photographs shown as evidence in this thread, because as I've mentioned it is mainly the sun's brightness itself that causes visibility reduction.

Regardless, this issue has been known since X-Plane 12 was in early testing, it is acknowledged by Laminar Research and it will be fixed in the near future. It is not an intended feature and while there likely are plans to properly simulate visibility reduction by the sun, this is not one of them. It is not a feature, it is a bug that requires changes to the approximation for the Mie scattering phase function and the tonemapper.

wow, ever thought of joining the LR team, they are hiring!! 👍

[Biology 1,347]

1 hour ago, UKflyer said:

wow, ever thought of joining the LR team, they are hiring!! 👍

This is actually something I've always wanted, but based on the announcement it seems like they're not hiring anyone for atmospherics or lighting 😞. Not even mentioning the fact that they already have 2 very skilled developers (Ben and Sid) working on atmospherics, lighting etc. and I don't think there's anything I can do that they wouldn't be able to. I mean if anyone from Laminar is reading this and is fine with me being nowhere near Ben or Sid in terms of graphical programming skills, I certainly would love to work for Laminar.

Edited November 28, 2022 by Biology

[peroni 541]

@Biology all you have to do is to apply. Don't expect them to call you

[Biology 1,347]

19 minutes ago, peroni said:

@Biology all you have to do is to apply. Don't expect them to call you

I definitely want to give it a try, but as I've said it seems like they are not hiring anyone for atmospherics or lighting, at least they weren't mentioned in yesterday's livestream.

[Gulfstream 416]

3 hours ago, Biology said:

I definitely want to give it a try, but as I've said it seems like they are not hiring anyone for atmospherics or lighting, at least they weren't mentioned in yesterday's livestream.

I've gotten the impression over the years that Laminar likes to run it very lean and mean and let the money flow in via that route.

That have some extremely talented developers, who, up to this point, have managed to make it work.  Sure, they've hired for things like mobile, but that's another cash cow.

It wold seem hiring full-time developers to work on tweaking atmospherics is not high on the priority list.   But you never know, I'd send them a resume.

Edited November 28, 2022 by Gulfstream

[Baber20 1,474]

@Biology 

4 hours ago, Biology said:

I mean if anyone from Laminar is reading this and is fine with me being nowhere near Ben or Sid in terms of graphical programming skills, I certainly would love to work for Laminar.

I believe you should atleast send an email to Austin @ "austin@x-plane.com" and share your findings. While you are at it, ask him to forward your theory to Ben Supnik who is the graphics lead. Maybe he can connect you directly to Ben. But do send an email. Austin almost always replies back.

[rka 916]

@Biology Dude, apply. Now. 😉

[SAS443 1,729]

To answer the thread title
"So some people complaining about sun glare in XP12 ?

Yes some of us are still complaining, and looks like we had a reason for it. 😜