I was there since the release yes, and Asobo indeed changed color grading etc. a few times, but none of such changes have any performance implications just like the scattering & absorption coefficients, which is why I think it has nothing to do with Xbox’s performance limitations but rather Asobo’s own artistic vision.
The cloud layer depiction certainly changed, but I believe it’s a result of trying to better capture METARs and not related to Xbox performance, as merging layers actually hurts performance rather than improving it as now the ray marcher has to evaluate density for an even larger distance, despite there being lots of gaps between the layers.
Sorry, I couldn’t edit the post in time. You’re completely right and I also think the cloud depiction in live weather mode changed for the worse, I just don’t think it has anything to do with Xbox performance constraints but the attempts at making the live weather more accurate to the METARs. As said, rendering a single taller layer is actually worse for performance than rendering shorter separate layers because the ray marcher can no longer directly skip the empty gap between the layers, so it would make no sense to make such a change to improve the performance for Xbox. I can still easily get pre-SU5 & pre-SU7 look from the clouds in manual weather mode by placing them accordingly.
This is why I have been against relying more on METARs since the beginning, METARs simply don’t contain enough information to drive clouds over a large area, so the less data from Meteoblue and more data from METARs means less variation. If you haven’t voted the wishlist item for having a separate fully Meteoblue driven live weather option, please do.
I believe I had, but will double check that. Having the option to choose one or the other would be good. I assume that a client side toggle could be used to select which data set you download/stream, MB only or the MB/METAR blend.
Basically yeah, a toggle to not blend in METARs to the Meteoblue coverage would make a massive difference. Another annoyance of mine is that since reliance on METARs increased the amount of sudden weather shifts also increased a lot, likely because METARs are issued at longer intervals while Meteoblue model is much more continuous.
For the atmospheric changes, if some consensus on suitable values could be made, it would perhaps makes sense for Asobo to change to those values, with no GUI interface to change them, while that could be a more long term goal.
looks great but I still say it looks weird to have the darkest part of the sky being the top middle when the source of light is fading off to the right. Why is there light in the upper left? They need to fix that imo.
The image seems to be quite wide-angle, and the sky tends to be darkest at the zenith (shorter travel of light rays through the atmosphere) after dusk. So this seems to be OK for me.
Exactly. I think it’s the same as Alexander’s dark belt - the space between a primary and secondary rainbow is darker because less light reaches your eyes.
BTW with all the talk about Rayleigh and Mie scattering.. it’d be so very cool to have more atmospheric optics phenomena. Fogbows, glories, coronae, iridescence, haloes etc.
My problem is the U shaped dark area. Look at the photo I posted in post 205 and another photo in post 344 – they both show that, with the sun being the source of light, the atmosphere blocks the light in the direction away from the sun and the strength of light fades fairly evenly away from the sun.
During the day, lower altitudes are brighter because there are more particles to reflect light and those lower altitudes are fairly evenly lit due to the sun’s position but, as the sun sets, it tends to light up the whole sky more evenly because the light is cutting across the Earth’s surface and this reduces that “upper sky dark” effect. As it continues to set and the Earths shadow is seen, light fairly uniformly fades from away from the sun towards the sun as it sets.
Maybe this effect is different as seen from higher latitudes but for every place I’ve lived, thats the way it is.
I think it’s mostly accurate, and caused by the directionality of Rayleigh scattering. If you look at the Rayleigh scattering phase function, which shows in what directions light is more likely to scatter, you will see that it’s lobe-shaped, which means there will be two lobes in the sky where it will be darker. It’s something I always pay attention to during sunsets / sunrises in real life and it’s indeed always there, even during day, though the sheer brightness of the sky makes it harder to see. I think it’s just amplified in the screenshots due to the high FOV. If the multiple scattering strength indeed needs to be increased (because I’m still not sure if that’s what the causes the yellow coloration with ARPC coefficients alongside with dark sunsets / sunrises or camera exposure not being enough), that would also help lighting up the sky more uniformly.
Sadly I really don’t think we are going to see this implemented unless it can become a mod made available that goes into the community folder. I cannot see how Asobo will support this, given the resentment/resistance we have seen so far from them on the topic such as the threats/warnings issued by moderators about discussing this topic. Such a shame as it is/was shaping up to be the most significant environmental realism mod for the sim we have seen. I hope I’m wrong.
I don’t think it is really resistence… they just want to avoid masses of people without appropriate understanding to do things and later complain about issues caused by this. Something like
It also could allow people on one platform to have a different experience than other people have.
All that’s why post 391 has uid 409 in the url
Back to the topic: I think a short statement about if and in which timeframe this topic is investigated would be enough to relax the atmosphere. Currently people want and fear they may never get.
I mean they are going to bring it up in the upcoming Q&A so don’t be annoyed guys, the moderators are just doing their jobs here.
This small edit we do is not ‘enough’ as Asobo themselves have wider access to their own creation and can adjust multiple values that can elevate this to another level.
Multiple scattering, ozone rgb, rayleigh scattering, mie values etc
Looking forward to Seb talk about this and explain what their plans are.
