Night lighting issues still present - The community solutions

Bug Reporting Hub Resolved
,
1103

Regarding lamp shielding:
The shielding could be calculated with a brightness falloff factor on observers angle relative to the lamp.
This could be done with basic trigonometry by multiplying lamp brightness by Cosine for the angle between the horizon and the lamp from the observers perspective.
Lamp brightness adjustment = Cos(Tan⁻¹(altitude / distance))

But since the earth is round, this will have a slight effect on long distance lamps, but I doubt that the altitude falloff with distance would have big enough impact to motivate a slightly more complex formula.

But the main thing is, this would have to be calculated every frame for every single lamp on the screen, which would be quite a few(tens of thousands).
Or, restricted to closer lamps, as far away lamps will be closer to the horizon and therefore be the least impacted by shielding due to the shallow viewing angle

Right and wrong. The majority of lights are shielded yes, and I agree on the size of the blobby lamps… But we would definitely not see “primarily the reflections from the ground”.

That would only happen at a close to vertical / top-down angle. Real world flying in the night will show you this. Of course low altitude & close range will have the “ground-splash” take up a larger angular area than the actual light source but unless looking straight to the ground, the vast majority of lights will be lamps / point light sources. Like in these photos:

image
image1920×1080 155 KB

image
image1920×1080 167 KB

image
image1920×1080 157 KB

No, the lamps are an extended surface / texture, the polygon would only block the lower half of the textured surface facing the camera, or they would have to calculate a visibility-flag depending on source obscured by polyon, and calculating that for every single lamp, every frame, would certainly have an impact, as well as lamps going on/off instantly when passing the blocking threshold.

Not from me, (even though I’m probably one of the biggest proponents of long lamp visibility range)
Lamps far away (closer to the horizon) are seen from a more shallow angle, therefore the least affected by lamp shielding. So you would still have highways lamps stretching towards the horizon like in real life, but the shielded effect would be most noticeable at a relatively steep downward angle (usually blocked by cockpit unless in a steep turn, looking down with head close to window)

Agreed, that’s the bottom line IMHO too

2 Likes
1104

Yeah, i got that backwards, wasn’t thinking through it properly.

1105

According to a few graphic artists I’ve talked with, we probably won’t see nearly as much improvement as many believes/hopes…
Just a small increase in performance (according to devs) thx to better use of multiple cores, better API, ability to use some functions as raytracing, which would probably give less grainy reflections for those with RTX cards if Asobo decides to implement this. There’s no “magic” that will visually change night flying appearance that couldn’t be done with DX11.

A poorly optimized lamp-texture in DX11 will look equally bad in DX12.
We had excellent looking lamps in the early alphas over a year ago (still DX11), but this got downgraded before release, nothing to do with DirectX version, when they switched to more semitransparent blobby/large bulb-textures and decreased the bloom, giving us a inferior night appearance now. Again nothing to do with DirectX version.

The devs said themselves just a few days ago that they’ve already done lots for the close up lamps, just the long range and world map left…
And I’m seriously worried, since they said during an earlier Q&A, that they would “fix/improve” long range sepia mask by getting higher resolution sepia mask, even though vastly superior methods has been presented via the community over and over again.
They seem to be “locked” on their own “artistic” approach, even though it differs quite significantly from both real world and the better looking early alphas.

I don’t mean to sound pessimistic, I’m one of the bigger optimists since I KNOW for a fact how much better/real night flying can look in MSFS given how great the early alphas looked over a year ago…

But for some reason Asobo suddenly opted to take a completely different approach both less realistic and less eye-pleasing.

The core problem of downgrading the excellent early alpha lighting and the stubbornes to avoid fixing it has been dodged over and over again, to a point it’s becoming an elephant in the room.
Frustrating to say the least. Almost like buying a car firing on 8 cylinders during the test drive, but ending up with a 4 cylinder at release, with zero explanation.

Come on Asobo, we are well aware you knew how to build 8-cylinder awesomeness one year ago, make night lighting great again :stuck_out_tongue:

7 Likes
1106

Perfect example of how much beautiful the night lighting would look just if Asobo could remove the horrible sepia masking as a first step. Unilluminated areas pitch dark instead of brown glowing grounds everywhere…

8 Likes
1107

From this image (and other ones), we can clearly see “non street lights” light sources scattered on the surfaces… Lights from building windows etc are visible until miles away from the observer.

In the sim, buildings no longer have night window lights etc after a few miles (when it starts to go to lower LODs) and a bit further they’re no longer drawn (leaving night lighting only by street light orbs). This is one of the big handicaps in MSFS night lighting.

2 Likes
1108

This! The word “downgrade” not only affecting night lighting, as we can see in the forum.
See threads about draw distance, water, clouds, LOD, photogrammetry etc and we can see similar theme… “Downgrade” felt widely by the community.

Using your example, if we buy a car from a brand / company that promise us “we’ll continuously work and updates your car type for at least 10 years!”, leading customers to believe or expect that their car will get vastly better in 10 years time… And a few months later the company starts “updating” your car in a way that customers feel it’s been downgraded… then again… and again… How’s the company / brand’s REPUTATION to you? Will you buy any other cars made by the brand? I know I won’t!

2 Likes
1109

This is a bit of a weird one. Asobo are of the opinion that they’re improving things. I don’t think they’re lying about that, i think they genuinely are convinced they improved things.

This is the disconnect people are talking about in multiple threads on this forum. I don’t think it’s done intentionally. It seems to me that Asobo genuinely care about community feedback and what to improve user experience. But it seems to me like the overarching issue here is that the method through which feedback is passed from the community to the devs is fundamentally broken. Look at the Q&As. People spend time writing elaborate posts, bringing in evidence, suggesting fixes, discussing user experience. Out of that, all that makes it to Jorg, Seb, and Martial’s ears is one, maybe two sentences at best. Take the question about the 3 cloud layers in ATIS. It took half of chat screaming about it for them to finally realize what we were talking about, that’s how abbreviated the question was. They are not shown before and after photos of the night lighting. They are not given any detailed information on what the issue is. So short of them personally reading the thread they are not getting any of this information. And even when they do it, this thread has over 1000 messages. I’m pretty sure they don’t have time to read through that.

I think what we need to do right now is, before complaining about this bug, or that bug, or this downgrade, or this issue, we should make it a top priority for Asobo to change their feedback system, especially during Q&As. More detail needs to be provided to them when the question is asked. I think that there must be a team of people, maybe dedicated users from the forum, that take the questions suggested for Q&As and compile a slide or two of evidence and descriptions which is then passed to the devs. That way they can see what we’re talking about.

4 Likes
1110

I decided to open a topic of discussion on this. Maybe we can get somewhere with this.

2 Likes
1111

You might want to consider starting with these:
Night lighting issues still present - The community solutions - #1013 by CptLucky8
Night lighting issues still present - The community solutions - #719 by CptLucky8
Night lighting issues still present - The community solutions - #1057 by CptLucky8
Runway approach lights - should be directional - #6 by CptLucky8

4 Likes
1112

Just to be clear, i didn’t mean that we, as a community, are doing a bad job of giving them this information. I meant that there is an informational breakdown between the information we post and the information the devs receive.

Other than that, yes, your posts are exactly what i am talking about. It’s a crying shame that you put so much detail and effort into those posts but all the devs get to hear about it is: “The community wants to know what’s happening with the night lighting.”

3 Likes
1113

Or any one of them could just take off, climb to FL200, look out the window (at night) and see that it looks rubbish? Ditto the tree LOD issue?

I don’t know why we have to do all this bargaining too and fro to get results from what should have been an astonishing sim, but now feels more like X-Plane with worse flight dyamics and slightly blurry ortho you don’t have to download.

2 Likes
1114

But you’re talking about subjective experiences here.

The community: “This doesn’t look right to me!” talking about lights being big and blurry.
The devs: “You’re right! I see it! We shall fix it!” talking about how far away lights extend.

What looks bad to you might look good to someone else. That’s why a picture is worth a thousand words. We can show them exactly what we’re talking about, with no room left for subjective interpretation.

To me, it’s simple: This is important enough for me to try and take the extra step. I can sit around complaining that the devs don’t see what we post, or i can try and make things better. If the devs aren’t open to this at all, well, at least we would have learnt something.

1 Like
1115

thank you for sharing this indepth comparison, sadly it seems like Asobo/MS doesn’t care at all about what this “sim” looks like, but what the numbers say, performance, Xbox compatibility, money…
And regarding all the “updates” since release, they barely, if at all, listened to the community/customer.

1116

The reason is simple: they are expecting the community to using the voting system!

In other words: the more votes, the more attention*

This is the way they told us how to raise the most important questions from the community!

*although it doesn’t always work like this so it seems.

Example #1:
The following topic back from October 2020 was engaging the community to better reporting in order to help Asobo tracking bugs. It was taking as an example the LOD problems I’ve extensively documented which are still present 5 months later. However it was closed:

Example #2:
The 2 most voted topics for the guided Q&A where nearly the only ones not discussed at all:
Live Dev Q&A: Guided Question - #14 by CptLucky8
Live Dev Q&A: Guided Question - #59 by EvidencePlz

Example #3:
Discussion: April 8th, 2021 Development Update - #75 by CptLucky8

1 Like
1117

I’ve just added some comments on this topic which is related but talking about another type of night lighting problem:

1118

It looks like it cannot be solved at all. Waiting for years for a problem to disappear should not take that long in today’s technology. As if they were deliberately lying and doing it deliberately. They don’t care about the community. And every update they’ve made to date has brought new issues. Conclusion There are tons of issues still.

1119

Unfortunately it took about 5 mins to be able to go that altitude for just cairo

1 Like
1120

I’m a month late in seeing this, but those Rio comparisons really are great. And the biggest problem to me is the window lighting. Just left of the compass is a cluster of buildings, and the alpha version is the ONLY ONE with the windows lit up. Flying in big cities has been such a letdown because downtown is always a black hole where the buildings block the street lights but don’t add any light of their own (until you’re right on top of them).

Is this being addressed in SU6? Asobo really should be more specific about what “long-range quality fix” actually covers.

1121

At the risk of drawing out the ire of the forum militia, one does have to wonder why these changes were made… First reducing the range of visible lights, then simplifying the lights, then no more lights in windows… That, to me, sounds like trying to reduce lights in the scene in order to improve performance. And while I certainly appreciate it on my aging machine, i’m sure people with new configurations don’t exactly welcome it. But why would they be reducing lights and improving performance, if that is what they’re doing? Is it for the Xbox, maybe? Would be strange, since we were told graphics won’t be reduced on PC in order to cater for Xbox. Or maybe it’s a conflict of terms. What we understand by “graphics won’t be reduced” is “graphics won’t be reduced”, while what Asobo may be understanding by that is “graphics won’t be reduced noticeably”. And then we can get into the subjective topic of what is noticeable and what isn’t… It’s all speculation, but one cannot help wonder why this is happening.

5 Likes
1122

They always leave out the word PC, i.e. graphics won’t be reduced on the xbox