[PerryPlatypus]

Not sure if this belongs in this subforum or the general Discussion one, but if it needs to be moved, admins feel free to do so. :)

My apologies if this has been discussed before. This has bugged me for quite a while, but it is a pretty big visual issue, in my opinion, that distant objects are still visible, especially when silhouetted against the sky, even with very thick fog. Additionally, even at very thick fog settings, Distant Mountains are still very visible (they should be completely invisible in thick fog), and headlights and lights from signals are still very clearly visible at long ranges.

The simple answer of course would be to change the "Viewing distance" before starting a run, but I think this defeats the point of being able to have dynamic weather during a run, especially if you want an activity that starts off on a foggy morning, with the fog gradually burning off throughout your run (as one example).

I am wondering, is there an issue with not being able to easily implement (on the programming side) a dynamic "Viewing distance"? In an ideal world, I would recommend having the visual distance directly tied to the Fog level, at least once the Fog range is down less than a mile or so.

The following screenshots were taken with a "Telephoto" FOV, with a distance of about 700 meters from the camera to the signal, and a distance of about 1200 meters to the distant trees. Note that in all three screenshots, the Red signal to the right of the locomotives, the Headlights, the distant tree outline, and the distant terrain outline against the sky are all visible.

http://i1064.photobucket.com/albums/u380/PerryPlatypus7/MSTS/MRL%20Mullan%20Pass/Open%20Rails%202016-04-11%2007-53-34_zpsnppk02oo.png

http://i1064.photobucket.com/albums/u380/PerryPlatypus7/MSTS/MRL%20Mullan%20Pass/Open%20Rails%202016-04-11%2007-53-40_zpsxmatofu7.png

http://i1064.photobucket.com/albums/u380/PerryPlatypus7/MSTS/MRL%20Mullan%20Pass/Open%20Rails%202016-04-11%2007-53-53_zpseyyxfrar.png

[railguy]

I believe that this is a well-known issue. The weather module is very much still a work in progress, though I believe it has a lower priority right now--the OR team could comment on that. One can get some nice effects by adjusting both the cloud and fog layers, but neither one by itself gets the nice nuances of real weather. That said, being able to change weather in-game, either manually or within the .act file, is leaps ahead of MSTS. That's a tradeoff that I'm willing to accept.

In my opinion, the sky layers, clouds, and fog need some additional work. In the sky layer, the stars are way too large (and I live in a real world place with some of the best star-gazing available anywhere in the world to use for reference). To be effective, I think the cloud layers need to have three layers--high, medium, and low--just like the real sky does. The altitude of the clouds would dictate the fog layer. Realistic thunderstorm clouds, lightning, and thunder is also a big wish of mine. About the closest atmosphere in a sim that I can think of is the Rex4 weather accessory for Flight Simulator. That would be a good example to emulate. At one point, I suggested that Rex4 be approached about adapting their weather module to OR, but I doubt that would be practical.

For the record, I don't use MSTS environments in OR because a.) it breaks the dynamic weather option and b.) the cloud types that I like best from Kosmos don't display on a correct axis in OR.

[James Ross]

PerryPlatypus, on 11 April 2016 - 07:18 PM, said:

Not sure if this belongs in this subforum or the general Discussion one, but if it needs to be moved, admins feel free to do so. :)

It's a mixture, really; it is a known issue with the current code in the renderer.

PerryPlatypus, on 11 April 2016 - 07:18 PM, said:

My apologies if this has been discussed before. This has bugged me for quite a while, but it is a pretty big visual issue, in my opinion, that distant objects are still visible, especially when silhouetted against the sky, even with very thick fog. Additionally, even at very thick fog settings, Distant Mountains are still very visible (they should be completely invisible in thick fog), and headlights and lights from signals are still very clearly visible at long ranges.

This is the known issue; we know it happens, but nobody's gone in and solved it yet. :)

I don't mind the lights issue so much as they should be visible a bit further away than unlit scenery, but not infinitely further away as is currently the case.

PerryPlatypus, on 11 April 2016 - 07:18 PM, said:

I am wondering, is there an issue with not being able to easily implement (on the programming side) a dynamic "Viewing distance"? In an ideal world, I would recommend having the visual distance directly tied to the Fog level, at least once the Fog range is down less than a mile or so.

This would be doable but it's probably simpler to fix the visual issues, as we have a number of other things involved in viewing distances.

[PerryPlatypus]

James,

It sounds like we have a similar understanding of the issues and what it should look like. :) And I agree that fixing it visually would be the best solution.

It sounds like what is needed is to have the fog work a bit differently. As the fog increases, distant mountains should first gradually become hazy and then disappear entirely, then a medium-range "layer" of the fog that 100% blocks distant objects, and then finally a closer range layer for the actual "Fog" appearance of haziness in the nearby air. For example:

Fog Level 0%: No fog present, distant mountains fully visible
Fog Level 25%: Low clouds visible, distant mountains partially obscured/hazy
Fog Level 50%: Thick low clouds, some thin close-range fog, distant mountains completely invisible.
Fog Level 75%: Normal scenery fading in visibility until approximately 250 meters, lights visible until approximately 500 meters (diminishing in strength between 250 and 500 meters), no distant mountains visible.
Fog Level 100%: Normal scenery fading in visibility until approximately 30 meters, lights visible until approximately 60 meters (diminishing in strength between 30 and 60 meters), no distant mountains visible.

I do wish I had knowledge of the inner-workings of this kind of stuff myself, as I would try to experiment with it myself.