I replaced the original clouds textures to see how the system works. X-Plane uses different textures to depict cloud puffs at various distances. That's the reason you see rings of different colors around the aircraft. The weather in the image is overcast with MINIMUM cloud puffs density! (10%).

 

Marco

Hey, that's what the clouds used to look like back in the seventies.....hmm....there was probably a reason for that.

 

Edit: It's also why we see a "mushroom" of cloud around the aircraft that I have mentioned in other posts. Your colourful example shows it all too well.

I replaced the original clouds textures to see how the system works.

Very interesting. So do the letters and numbers have any significance code-wise other than acting as placeholders for the purposes of this experiment?

Edit: It's also why we see a "mushroom" of cloud around the aircraft that I have mentioned in other posts. Your colourful example shows it all too well.

 

That's right. I think with a good set of cloud textures the effect could be minimized. The idea of using different textures resolutions depending on distance seems good to me. The real problem is that I experience a gap between the different rings, when in overcast conditions.

 

Very interesting. So do the letters and numbers have any significance code-wise other than acting as placeholders for the purposes of this experiment?

 

Placeholders for now. There are 6 cloud textures, each of them made of 4x4 subtextures. In my picture each letter+number identifies a single subtexture/puff.

 

The first 2 hi-res texture sets are not used when the cloud rendering density is below 108%. Here is a picture with cloud puffs at max density (150%). The closer red and green rings are the two texture sets I'm talking about.

 

 

Yes, I was just thinking that myself. If there was perhaps a greater variety of shapes or something it might help with the effect. They still need to extend to the horizon in overcast conditions though, just make the rings extend further?(I feel)

 

Glen

 

Edit: That picture with the aircraft in it, what does it look like from the ground? Do the colours extend to the horizon? There is something that shows to the horizon in overcast, but it's not "shaped" clouds, was just curious if a colour went to the horizon in your test, from the ground.

Yes, I was just thinking that myself. If there was perhaps a greater variety of shapes or something it might help with the effect. They still need to extend to the horizon in overcast conditions though, just make the rings extend further?(I feel)

 

Here's my continued experiment. I copied one of the hi-res texture sets (the "yellow" one, in my first pic) to the other sets, basically having only one single texture set for all distances to try minimizing the "ring" effect. The pics below are with 50% cloud density, overcast conditions and max visibility (100 miles). First pic is 30.000 feet above the clouds, second pic is at 1000 ft, 9000 ft below cloud deck. Seems not too bad to me, the ring effect is not very noticeable and the clouds seem to extend quite far. With better puff textures the result could be even better maybe.

 

Marco

 

 

 

 

Edit: That picture with the aircraft in it, what does it look like from the ground? Do the colours extend to the horizon? There is something that shows to the horizon in overcast, but it's not "shaped" clouds, was just curious if a colour went to the horizon in your test, from the ground.

 

Here are overcast conditions on the ground (50% cloud puffs density, 100 miles visibility, 90 deg field of view). Extending pretty far! :smile:

 

 

 

 

Seems not too bad to me, the ring effect is not very noticeable and the clouds seem to extend quite far. With better puff textures the result could be even better maybe.

The density is very impressive, and the clouds really have a gloomy and moody look, but I think they appear slightly rigid and unnatural in some of the images. To me, the second image in the post I linked depicts the most-realistic clouds, as they show a greater sense of tension, turbulence, and variety.

Hmmm Very interesting seeing how XP generates it's clouds... Good job on your findings so far... I can't imagine how XP's weather would look like if it was fully understood how it works and developers of OPUS and AS would jump in...

Looks ok, but 90 deg field of view is way too much for anything other than very wide view setups like mine at 45". Put it at a reasonable 50 deg and the clouds won't come close to the horizon, your zoom is bringing the cloud and horizon together. They might extend to the horizon at 150% puffs but no system can run at that if you want any other settings high(er), and or a descent aa setting, and definitely not with hdr on. The clouds at a reasonable setting and view angle extend out a few miles around the aircraft and that's it. Not trying to diminish your testing, just stating facts. The rings still exist, hope that explains better? The only present way to get rid of the rings "mushroom" is to run your cloud settings too high, at the sacrifice of too much else. Even then you still have a gap from the cloud to the horizon. Maybe it's not physically possible to do what I'm asking with computer simulators?

Looks ok, but 90 deg field of view is way too much for anything other than very wide view setups like mine at 45". Put it at a reasonable 50 deg and the clouds won't come close to the horizon, your zoom is bringing the cloud and horizon together. They might extend to the horizon at 150% puffs but no system can run at that if you want any other settings high(er), and or a descent aa setting. The clouds at a reasonable setting and view angle extend out a few miles around the aircraft and that's it.

 

Ok, I think I've pinned down the issue. Actually it's not related to cloud puffs density nor view angle. Increasing cloud puffs density should not increase the distance the clouds are rendered.

 

The problem seems to lie instead in how X-Plane depicts visibility. I'll explain: in the first pic below, there is max visibility and the clouds are visible near till the horizon, so the mushroom effect is not very noticeable. In the second pic, the visibility has been reduced a lot (10 miles IIRC) and the clouds disappear at 10 miles, giving the mushroom effect. Now the problem is, why does the final effect is not realistic in X-Plane? What would one see in real life with overcast clouds and 10 miles visibility?

 

 

 

 

That 3rd pic doesn't look too bad, actually good, but could be better if the "darker" cloud extended out even twice what it does now in your picture? Then the "illusion" of overcast would be a little better represented? Even if it faded would be better, as it stands right now if just stops, it's too abrupt to the eye.

Murmur, it seems like XP is trying to limit the actual cloud coverage to the same distance as the visibility but the transition is so abrupt, that it creates the mushroom effect. I like their approach but it just needs to be tweaked or refined... I've learned a ton by just a couple of your posts on this thread! lol

Murmur, it seems like XP is trying to limit the actual cloud coverage to the same distance as the visibility but the transition is so abrupt, that it creates the mushroom effect

 

+1

 

In RL visibility is not usually even in all directions, so, a 10 mile vis will be the minimum, but there are variations, and certainly the transition is not abrupt like in XP10 cloud rendering...

I would also like to take this opportunity to focus in yet another limitation in XP10 - cloud layer thickness limited to minimum 2000'...

 

Murmur, thanks A LOT for your tests!

By the way, there was an article by Austin (even though his language might not be appealing to everyone :lol: ) about the XP10 cloud engine way back:

http://www.x-plane.com/2010/12/cloud-rendering-in-x-plane-10/

 

And somehow I don't see (currently) the images in the x-plane.com site, but the same article was reposted (with the picture) at the ORG:

http://forums.x-plane.org/index.php?showtopic=49687

 

This underlines, that - and why - your observations are correct.