unnamed Mountain (possible NWDA Entry)

[Goms]

After 4.5 hours @4x2.6 GHz i may present you: the (yet) unnamed mountain.
My Idea was to try to get a realistic shape of a granite mountain.
The Surface is not done so far and also the metaclouds need to be tweaked a little bit.
I hope i'll get the image done until saturday afternoon.
An ideas to improve this?  

regards, Goms

edit: argh, wrong board. can maybe some admin move this to the image area.....?
sry

[deepdish]

[EoinArmstrong]

Great PoV and contrasting colours!

[Goms]

The contrast is from a little bit of manual HDR-mapping.
The original had no contrast  in the areas with snow cover.
In the next step i will adjust the coverage of the "icy"-parts on the rock with favor depression and give the rock itself more color variation into a light beige.
What i'm not sure about yet is if i will add some high clouds. I think it would seem a little bit strange compared to reality?

[FrankB]

Hi Goms,

I am *very* happy to learn that you are going to participate

Here are my thought on how this could be improved: the light blue color on the granite looks wrong. Also it seems that you are using a texture for the granite - it looks a bit washed out. I reckon you would get crsiper result if you make the texture procedurally. Then you could also render at a higher resolution and benefit from that extra details, whereas the texture will look even more washed out.
I think that your recent procedural surface for the micro cracks could be a good texturing component for the granite (although we may not be able to see all the details, and you would probably have to adjust the colors of it somewhat to get that granite look).
However, I had good result in slightly mixing a texture into a predominantly procedurally textured scene. That's something you may want to try.
Also try intersect underlying and favour depressions for the snow coverage.

I love the simple sky with the strong blue gradient and the light cloud - which could probably be a little whiter.

Cheers,
Frank

[schmeerlap]

I'm not sure what a "metacloud" is, but I'd just keyed in "it could be a little whiter" when I got the "further messages while you were writing" warning. I see the "metacloud" as snow driven off the mountain by the wind, and that is probably your intention.  I wouldn't add any more clouds in the background as that would detract from the contrast. Looking forward to the finished scene.
(Planetside might wish to add that excellent jawdropper smiley posted by deepdish to their array of smileys.)

John

[Goms]

@Frank: the blue color should be ice on the rocks, but as i wrote i will have to improve it. The favor depression function should give the right results when combined with some reflections. I've tried to use intersect underlying for the snow, but it didn't work very good. To look realistic the patch size needs to be "very" huge, which leads to some problems in combination with smoothing. But i will work something out.
The texture is in fact procedural, but has no displacement yet. I will add some more contrast and color variation as cracks and some strata too.

@John: The Idea of meta-clouds is from this thread: http://forums.planetside.co.uk/index.php?topic=3691.msg50489#msg50489.
And you're right, it should be snow driven from the top. I'll make it brighter and see if i find a seed and a good warper to get a better structure.

Also i want to add a few rocks and maybe a climbing person if i have the time to model it. Next update will probably follow this evening.
Thank for your comments so far!

[FrankB]

Is that the merchant from monkey island I see as your avatar here

[Goms]

Yep, Stan.

[FrankB]

@Frank:...
The texture is in fact procedural, but has no displacement yet.

Hi,

ok, so I thought the granite and light blue color did belong together as a photo texture, hence looked washed out, but I see how it is now.

Thanks,
Frank

[Goms]

I'm just about to render another "preview". The displacement intersection still won't work like i want it to, but i also don't want to use more than one compute terrain node. I will see in around 4 hours.

[Henry Blewer]

I am really impressed by two things. Mainly this render. It really is fantastic. I agree with Frank about the blue coloring. I have seen real rock formations that are blue, but a lighter color. This color would look good as a component of a marble texture.

The second, less topical, thing is the jaw dropping smilie.

Looking forward to more refinements of this render!

[Tangled-Universe]

Very nice work, impressive Looking forward to see where this goes. Frank already provided you some good pointers.
Personally I'd remove the thin snow which reaches to the top of the mountain. I'd keep it to one snow-border/edge. Just personal preferences, know what it's worth!
I like the deep blue sky and perhaps you could try to keep it also abap (as blue as possible:)) near the horizon, think that would be a tad more realistic.
Also, there are some topics about how to make clouds follow terrain, perhaps you might find it interesting to try.
Good luck, you still have 3 weeks, so hopefully you can improve it even further

I'm just about to render another "preview". The displacement intersection still won't work like i want it to, but i also don't want to use more than one compute terrain node. I will see in around 4 hours.

I understand why you don't want to use too many compute terrains/normals because it slows down the render, but as a matter of fact if you build up your network "logically" you won't need an extra compute terrain/normal to let the displacement intersection work properly.
Simply put, it often works best if you use those kind of shaders as last in the chain.
My snow and canyon (sand in canyon) use this principle with success.

Martin

[Goms]

Well, i have only until Saturday; Sunday morning i wall leave to Korsika for two weeks. 
The Problem is i have to write a 15-20 pages paper about medieval... well, some kind of boring stuff which only is interesting for historians.
The Problem with the compute Terrain node is that i use a intersect underlaying function twice.
One for the icy parts as favor depression, which shall give very exact results, and than one for the displacement intersection, which shall give only bigger snow-fields.
For Terrain-following clouds there is not time, i guess. but i will think about it.

[Henry Blewer]

The paper wouldn't be about trade treaties in the Hanselactic League would it? I did one in college about it.