Experimenting with X-Plane cloud system.

[BimmerCop 332]

Maybe Austin could employ these puffs... They seem powerful enough... L'meow

 

[FloB 34]

What I don't understand:

It seems like someone who knows/understands the cloud/view distance code should be able to really improve the look without a lot of work. But actually this problem is not adressed by Laminar or anyone else. Probably there Is more than what meets the eye...

 

Flo

[Muskoka 104]

What I don't understand:

It seems like someone who knows/understands the cloud/view distance code should be able to really improve the look without a lot of work. But actually this problem is not adressed by Laminar or anyone else. Probably there Is more than what meets the eye...

 

Flo

 

I agree, and sometimes don't understand. It's like you've got this far doing some wonderful amazing things, why fall short with this. It's the first thing you see when you look out the cockpit. I think it's just not high on his list, basically to him it's good enough, just speculating of course.

 

Glen

[Mountain Man 142]

I wonder if it would be possible to use 3D clouds up close and 2D clouds for anything a certain distance away from the plane. Would that allow cloud cover to extend further without significantly impacting performance? (Assuming it doesn't do something like this already.)

[Muskoka 104]

I wonder if it would be possible to use 3D clouds up close and 2D clouds for anything a certain distance away from the plane. Would that allow cloud cover to extend further without significantly impacting performance? (Assuming it doesn't do something like this already.)

 

That's why I asked Murmur, if in his test, from ground level, you can see a "colour" that extends out to the horizon, and if so, what is it (not the colour, the texture)? We can all see there is something there (grey mass?), but what is it, what's being drawn? Like you say, perhaps whatever it is could be replaced with a 2d cloud that extends to the horizon, for performance reasons. They don't have to look spectacular or be of high quality, just extend out to the horizon for that overcast "look".

 

It's overcast here today and the "clouds" sky are visually symmetrical, from above me to as far as I can see, to the horizon. Can that not be reproduced in a simulator?

 

Glen

[Murmur 3,941]

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.

 

Yeah I agree! I think a bug report could be filed.

 

That's why I asked Murmur, if in his test, from ground level, you can see a "colour" that extends out to the horizon, and if so, what is it (not the colour, the texture)? We can all see there is something there (grey mass?), but what is it, what's being drawn? Like you say, perhaps whatever it is could be replaced with a 2d cloud that extends to the horizon, for performance reasons. They don't have to look spectacular or be of high quality, just extend out to the horizon for that overcast "look".

 

Here's my placeholder textures used for overcast. First pic, max visibility, second pic, 10 miles visibility.

 

 

 

 

[Muskoka 104]

So (thanks Murmur), it would appear that Efrain is correct. Is it a bug per say, or a conscious decision based on performance reasons?

 

Glen

 

Edit: Thought we had discussed all this before, and doing a quick forum search found multiple threads discussing the exact same issues early last spring. When XPX first came out we had the 2 slider system for cloud adjustments, I don't recall suffering these visual boundaries / anomalies when those options were available to the user. I'm sure you can see where I'm going with this, it's no different today than the discussions we had 12 months ago on the same subject.

[Murmur 3,941]

From Ben Supnik ( http://developer.x-p...stency-and-hdr/ ) 6 months ago:

 

Murmur says:

July 9, 2012 at 12:10 am

 

Slightly off-topic, quick question:

Will there be further developments of the cloud engine during the V10 run? I’m not referring to the “low-level” cloud rendering, but to the “higher level” generation of cloud shapes, patterns, distribution. The cloud engine is currently not capable of representing vertically developed clouds (towering cumuli, cumulonimbi, etc.).

If instead the current cloud engine is to be considered definitive for the V10 run, will it be possible to have access/override clouds creation via plug-ins?

 

Ben Supnik says:

July 10, 2012 at 2:43 pm

 

I don’t know if it’s final. We are not considering a cloud creation API. We may someday allow custom weather “scripting” via modified METAR files, but there’s a lot of other things that are higher priority.

 

In other words, the only way for a plugin to improve clouds depiction (apart textures) would be to completely override X-Plane cloud system and create their own cloud system/depiction... Not easy...

[BimmerCop 332]

From Ben Supnik ( http://developer.x-p...stency-and-hdr/ ) 6 months ago:

 

Ouch... :-/ oh well, since I can fly whatever I want, I'll just fly in CAVOK conditions hahahaha

[jcomm 5,890]

:-/ not good to know :-( Well.... At least the effects can be tuned, as Austin has demonstrated at least twice regarding turbulence and shear...

 

Overall, and although not being volumetric, I still prefer by far the FSX cloud system, specially when used with good add-ons that replace the default textures / etc... FSX with a good weather injector and cloud enhancer looks great, and MS FLIGHT as well, while X-Plane can't compare in this area, IMHO...

[Murmur 3,941]

In several aspects, X-Plane 9 was better at depicting clouds:

 

 

 

 

 

X-Plane 10 clearly needs improvements to the cloud system.

 

 

Marco

[Rations 102]

In several aspects, X-Plane 9 was better at depicting clouds:

From those screen shots, I agree. While the clouds are not as sharp, they look more free, natural, and unique.

[Murmur 3,941]

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

 

Well, I think it's a little different actually. "Cirrus" and "few cumulus" (the latter roughly corresponding to "cumulus humilis" in X-P) have basically no thickness in X-Plane, so the 2000' is a "virtual" thickness that forces separations between different cloud layers.

 

"Scattered cumulus" should roughly correspond to "cumulus congestus" in the X-Plane simplified system, and "Overcast cumulus" should correspond to "Towering cumulus" and "Cumulonimbus" (even if the final depiction often is not very similar to the real life, as we know). So it makes sense to give a minimum thickness to these types of clouds.

 

On the other hand, I assume "stratus" clouds in real life can be less thick than 2000'?

 

:-/ not good to know :-( Well.... At least the effects can be tuned, as Austin has demonstrated at least twice regarding turbulence and shear...

 

Overall, and although not being volumetric, I still prefer by far the FSX cloud system, specially when used with good add-ons that replace the default textures / etc... FSX with a good weather injector and cloud enhancer looks great, and MS FLIGHT as well, while X-Plane can't compare in this area, IMHO...

 

Well, sometimes X-Plane, as well, can do nice things with default clouds: :smile:

 

 

 

 

[Geofa 108]

Marco-thanks for your investigations and zeal to make ths sim it's best.

P.s. I was in Italy for a couple weeks last summer-what a wonderful place you live!

[Murmur 3,941]

P.s. I was in Italy for a couple weeks last summer-what a wonderful place you live!

 

We say: Italy is a wonderful place to live... except it's full of Italians. :lol: :lol:

 

Where have you been? Florence? Rome? In the years I befriended a lot of american tourists riding on the bus together :-)