Simple shape

[TheBadger]

Its interesting.
If I enlarge the ring, by checking move textures with planet, in the invisible planet. The ring will eventually be in the middle of the connected planet. If you look at it going by where the planets bonding box is. But its just an illusion or something. Nothing really changes in the render. except size, position stays the same.

A bug, Kadri? It may be. My first thought was I broke the tree somehow when I was playing around.

[Kadri]

Change the order of the atmosphere +cloud ring node tree then change it to the original ones more.
The unattached atmosphere will be in it's normal position ones again after this.
If the problem is this of course .
This bug might be fixed by now maybe.

Without seeing the file i can not say much more !

[TheBadger]

It works, thanks Kadri.
[attach=1]

The problem was in the altitude constraints of the surface layer. I had thought to try here too. But when I played with it I got a cloud in a perfect sphere. And that made me think that indeed the problem was here. But when I tried to adjust it my clouds just disappeared, and I gave up on that as the source of the problem.

Kadri, how did you figure the correct number value? I really don't know. And I don't think I would have ever found it. YOu must have made a calculation bases on some other parameter, but what?? Or you guessed? Heck of a guess, -5 million and -7 million, in meters no less.

I have some tweaking to do as you can see from the image, lots really. But the mechanics of the scene are all working now!

Thanks!

[jo]

Hi Michael,

I would say he used the radius of the earth in metres (6,378,100) as the basis. 5-7 million spans that. It's negative to offset things from the origin to the centre of the planet i.e. from the top to the middle of the planet. That's my guess anyway :-).

Regards,

Jo

[Dune]

Do you still have the same problem as I had in the distance shader+cam setup I made for you? The problem that the clouds are only visible in front of the 'invisible' planet, I mean. I noticed it then, and I can see something like that happening in your last image.

[Kadri]

It is not so hard as it sounds.
By unchecking the lower limit it was clear that the problem was there.
The number looked small as Jo said, so i basically put one zero then one more zero and the ring was nearly where it should be.
After that you only adjust the upper limit to the lower one.

The worst thing to do is changing more then 1 option at a time.
And by the nature of TG3 don't hesitate to put big numbers in any place.
By exaggerating you can see where the problem is and adjust it faster afterwards.

[TheBadger]

Thanks Jo and Kadri.

Ulco,
It is an interesting problem. In the image above I simply ticked the "translate textures with planet" parameter in the planets main node window. And then increased the size of the planet until the ring was the size I wanted. This is the simplest fastest way I saw of making the radius of the ring bigger. But you see it has some problems.

Here is a render of the ring at the default size of Ogre's node tree. Kadri rendered this as a test, the cloud is visible in this shot except for where the planets shadow should be, I think. But when enlarged the way I said, that shadow grows unrealistically.
[attach=1]

I am going to have to return the invisible planet to its regular size, and enlarge the ring the hard way (all setting individually) to see if the problem is fixed.

The file you gave me is a different situation I was guessing. Wasn't where the ring apeared dependent on the camera, and the cameras field of view?

Going to work on it tonight, as soon as its quite around here. My kids are on some kind of rampage or something right now.

[TheBadger]

see. The invisible planet at default scale.
[attach=1]

Anyone care to define this?

[TheBadger]

K, It works completely perfectly.

[attach=1]

Cloud depth needs to be increased in relation to the size of increase to the Simple Shape shader. However it is not a 1:1 ratio.

The result will be that the missing spots fill in. But you will get a "O" that is really fat at the sides and narrow at the top and bottom. So I had to add a second simple shape and "merge shader". And then mess with the blending "inverts" of other nodes.

Now I get to do the fun stuff! Make the clouds look the way I want them to.

I did render a test from the surface of the planet of the cloud ring in the sky, its a rather pretty effect.


K, now I think I really am out of the darkness this time.   

[Dune]

Looking good so far, Michael. Problems are there to solve, and I'm glad you get where you want. And, yes, I meant the cam setup. I wonder though, how high your cloud needs to be on this galactic scale, and high good the quality needs be and thus render time will be....

[TheBadger]

Hi,

I have run into yet another strange issue.

I have the cloud ring and the rock ring. Both are perfect Circles when at 0,0,0. However, when I give them the tilt we have been talking about, the rock Circle becomes an oval. The settings in the Simple shapes are identical.
[attach=1]

I have been having some problem pin pointing the parameter that effects this because of some crashes. Does anyone reconize this problem? I need both to be a Circle or an oval. Don't really care which.


@Ulco,
As it is now it adds a good bit to the render time to the time without the cloud ring. But the cloud is super big and thick. And Im just leaving it that way because its really easy to see while I work on everything. Ill change it a lot before its done and render time should be better then too.
Although, just the blue cloud ring in the sky is kinda cool from the ground.

It does need to be as big a radius and distance from the planet that it is too. I don't think faking it would have the same look.

[Dune]

Are you sure they're in the same plane?

[TheBadger]

Meaning?

[Dune]

Well, if the rocks are seen on an angle different than the cloud it would look like an oval, if the view is not perpendicular.

Or maybe your SSS is not a circle, by accident?

[TheBadger]

Thats the thing, Dune. They are perfect when they are at 0,0,0. But when I add the tilt and rotation (-15,0,20) Then the rock circle deforms to an oval, but there is no issue with the clouds.

I just checked and rechecked. Only the edge width of some of the SSS where different. But I fixed that and there was still no improvement.

If its normal for the deformation to occur under these circumstances, than why does the cloud circle maintain a perfect form?
There is no difference between the sss in the clouds and in the rocks. I copied the rock sss to use on the clouds.

The only thing I can think of is that I have a fuzzy zone some place. But that does not explain why an oval, only why the rocks can seed outside of the sss boundary.

I hope being vexed is not bad for you kidneys or something. Because Im vexed again.

On a good note I have not had one single crash all night, which I don't understand at all. Last night I could not work for ten min, and tonight no problems. I have no explanation. Vexed I say!