Hello,
the right sphere and the one in the middle are imported from blender as .obj. The left sphere is an object from Terragen.
The question now is, why the sphere in the middle shows a correct transparency and the right one does not?
What is the difference between the noise of the function (Voronoi 3D diff scalar) and the noise from the Powerfractalv3?
Michael
Michael i tried many things then tried to search the forum.
I should have tried that first probably (you too ;) ).
I found an old post where Matt said this:
"There is a bug in 2.1.18.1 which requires objects to have UVs if they are ray traced. If you render without ray tracing, your objects do not need UVs."
http://www.planetside.co.uk/forums/index.php/topic,8557.msg91691.html#msg91691
Probably not yet fixed.But it is still easy to use the powerfractal node when you UV map the object as you see.
In Terragen terms what you are showing is related to Opacity rather than Transparency.
Matt explains some details about this in this post http://www.planetside.co.uk/forums/index.php/topic,17508.0.html (http://www.planetside.co.uk/forums/index.php/topic,17508.0.html)
Quote from: fleetwood on February 25, 2014, 10:46:53 PM
In Terragen terms what you are showing is related to Opacity rather than Transparency.
Matt explains some details about this in this post http://www.planetside.co.uk/forums/index.php/topic,17508.0.html (http://www.planetside.co.uk/forums/index.php/topic,17508.0.html)
Have you tried the file?
He might be using transparency instead opacity etc.but the problem here is that it doesn't work with the "powerfractal node".
It works only if you don't use "raytrace objects" and-or use UV mapped object.
Yes, I looked at the TGD file. In each sphere example, input is being routed directly to the Opacity Function of a Default Shader.
Quote from: fleetwood on February 26, 2014, 12:37:34 AM
Yes, I looked at the TGD file. In each sphere example, input is being routed directly to the Opacity Function of a Default Shader.
Yes you are right,but the problem is that they are the same object and not different ones.
The only difference is that one of them uses the powerfractal node.
The problem you linked is not the same.
When you switch the "Power fractal shader v3 02" node to the object in the middle and the "colour adjust shader 01" to the right object you get the problem this time on the object in the middle.
Or when you use an object that does have an UV map you can render it with ray trace objects on.
By the way when i wrote
"He might be using transparency instead opacity etc.but the problem here is that it doesn't work with the "powerfractal node"."
i was mentioning that he might be using the terms of "Tranparancy" and "Opacity" wrong .
I wasn't speaking about the opacity or transparancy input itself.
Maybe you were referring to that?
Thanks for the suggestions.
Now it looks even better.
The transform input shader does the job :)
... and now with transparency.
Thanks again for your help!
Michael
... sorry, I forgot the total transparency :-\
Kadri --- Yes the terminology issue was exactly my meaning. Always very hard to get things hashed out without using common set of labels.
Michael --- Now on the full transparency you're seeing the "black specks" problem on the left sphere. This problem has been around for a while. There was an earlier version of Terragen 2 where transparency worked, not just using the Water Shader, but also worked when turned on in Transparency option of the Reflective Shader, but it became broken in later releases of T2 >:(
On the earlier renders, the shadow of the Terragen displaceable sphere on the left always remains the shadow of a full sphere.
This little example is from that early T2 version before transparency got broken.
Nice that you got it working.
But i would still use an UV mapped object to get correct shadows Michael.
Not sure if there is another workaround for that.
Matt does have it on his to-do list to look at the water rendering issue and see if it can be fixed.
- Oshyan
Hello,
due to your recommendation I used the UV mapped object, but did not use the transparncy option.
Here now the final image:
[attach=1]
Michael
(But I still wonder what is the difference between the noise of the function (Voronoi 3D diff scalar) and the noise from the Powerfractalv3?)
Quote from: Kadri on February 25, 2014, 07:53:37 PM
I found an old post where Matt said this:
"There is a bug in 2.1.18.1 which requires objects to have UVs if they are ray traced. If you render without ray tracing, your objects do not need UVs."
http://www.planetside.co.uk/forums/index.php/topic,8557.msg91691.html#msg91691
Probably not yet fixed.But it is still easy to use the powerfractal node when you UV map the object as you see.
I thought I had fixed it in the very public release, v2.2. But it looks like the bug remained for opacity calculations only.
I've fixed this now so it should work in the next TG3 release.
Matt
Quote from: russe166 on February 27, 2014, 11:55:50 AM
(But I still wonder what is the difference between the noise of the function (Voronoi 3D diff scalar) and the noise from the Powerfractalv3?)
The bug only affects texture coordinates (UVs or UVWs). The Powerfractal uses texture coordinates (as do most shaders, although some are configurable). In your project you have a Get Position node plugged into the Voronoi. Get Position is not affected by the bug. If you had plugged in a Get Position in Texture (which I recommend for anything should be a texture), it would suffer from the same bug as the Power Fractal.
This bug will be fixed in the next TG3 release.
Matt
Quote from: Matt on March 03, 2014, 10:13:56 PM
Quote from: Kadri on February 25, 2014, 07:53:37 PM
I found an old post where Matt said this:
"There is a bug in 2.1.18.1 which requires objects to have UVs if they are ray traced. If you render without ray tracing, your objects do not need UVs."
http://www.planetside.co.uk/forums/index.php/topic,8557.msg91691.html#msg91691
Probably not yet fixed.But it is still easy to use the powerfractal node when you UV map the object as you see.
I thought I had fixed it in the very public release, v2.2. But it looks like the bug remained for opacity calculations only.
I've fixed this now so it should work in the next TG3 release.
Matt
Thanks Matt!
Thanks Matt for the answer. Now I understand much better.
Michael