Resistance

[schmeerlap]

A flight of Ki84 Hayate (allied code name Frank) meet resistance from a boy atop a mountain. An imaginary event in the spirit of the man confronting tanks in Tiananmen Square. The aircraft model is by Anders Lejczak (Bazze). The boy is extracted from a photo and integrated into the scene using Photoshop. The prop blur is also painted (smudge tool) in Photoshop. Below the single object instance of the Ki84 model is a pop of the same; to give the illusion of them being at slightly different heights I have varied their size.

John

[Tangled-Universe]

It's a cool scene but there's something odd regarding scales and depth.
The planes which intersect the clouds look huge because the clouds appear to be at a much lower altitude. The planes seem to be closer to the camera but still intersect the clouds which is conflicting.

[mhaze]

Nice idea, great rocks - agree there's something odd about the scales.

Reminds of climbing in N.wales before they brought in rules about how low planes could fly, I was suprised by planes passing under me and 50ft away from the rocks, nearly fell off couple of time

[schmeerlap]

It's a cool scene but there's something odd regarding scales and depth.
The planes which intersect the clouds look huge because the clouds appear to be at a much lower altitude. The planes seem to be closer to the camera but still intersect the clouds which is conflicting.

The render camera is at an altitude of 6300.
The near altocumulus clouds (which the near / topmost plane is flying into) are at an altitude of 6000 with a depth of 500. I believe this means their altitude extremities are 6250 and 5750.
The main body of altocumulus cloud is also at an altitude of 6000 but with a depth of 3500. So they will have a high point of 7750 and a low point of 4250.
The near / topmost plane is also at an altitude of 6000 (300 below the height of the render camera). So it is big because it is relatively near the camera.
The near clouds (with the near plane flying into them) are not, as you say "at a much lower altitude". The main body of clouds (which have been shoved away from the camera with a distance shader) because of their far greater depth may appear to be lower.
The other planes (a pop) are at an altitude of 5000. Below the near alto clouds, but still short of the main body of deeper alto clouds.
Because I am seeing the near clouds and plane as near the camera (which they are) I am not perceiving the scale problem you seem to be seeing.

John

[aymenk2003]

Good image and a good scenario ...the cloud is wonderful

[Matt]

The density of the clouds intersecting the planes and the mountain near the boy suggest that the boy and the planes are much too large. Or the clouds are too dense. That is where the scale problem arises for me.

The clouds look OK for the numbers you mentioned, so it's probably a perspective/scale issue with the planes and the boy. I would try to get closer to the peak, closer to one of the foreground clouds, reduce the size of the planes (and bring them closer to the camera to maintain the same size in the image). That way they will fit with the clouds better.

Have you measure the size of the plane sitting on the ground?

Matt

[Matt]

The render camera is at an altitude of 6300.
The other planes (a pop) are at an altitude of 5000.

That puts the other planes at 1.3km below the camera, and more like 4km if you take horizontal distance into account. They look pretty big for planes that are kilometres away. That's the main problem.

(It's only the density of the clouds that throws it off for me, though, so adjusting the foreground cloud density might be the simpler way to make the scale work better.)

Matt

[Tangled-Universe]

It's a cool scene but there's something odd regarding scales and depth.
The planes which intersect the clouds look huge because the clouds appear to be at a much lower altitude. The planes seem to be closer to the camera but still intersect the clouds which is conflicting.

The render camera is at an altitude of 6300.
The near altocumulus clouds (which the near / topmost plane is flying into) are at an altitude of 6000 with a depth of 500. I believe this means their altitude extremities are 6250 and 5750.
The main body of altocumulus cloud is also at an altitude of 6000 but with a depth of 3500. So they will have a high point of 7750 and a low point of 4250.
The near / topmost plane is also at an altitude of 6000 (300 below the height of the render camera). So it is big because it is relatively near the camera.
The near clouds (with the near plane flying into them) are not, as you say "at a much lower altitude". The main body of clouds (which have been shoved away from the camera with a distance shader) because of their far greater depth may appear to be lower.
The other planes (a pop) are at an altitude of 5000. Below the near alto clouds, but still short of the main body of deeper alto clouds.
Because I am seeing the near clouds and plane as near the camera (which they are) I am not perceiving the scale problem you seem to be seeing.

John

I don't want to sound anal or rude or whatever but the numbers may be correct in theory but as you can see that doesn't mean it also looks correct eventually.
The reason I think you don't see it is because you look at it with the "logic" eye since you know the exact numbers etc.
I certainly would try Matt's suggestions for improvement.

[schmeerlap]

"There is a general perception that statistical knowledge is all-too-frequently intentionally misused by finding ways to interpret only the data that are favorable to the presenter."
I got this from Wikipedia. And it fits this knuckle-headed scoundrel like a glove.
The more I peer at my flawed scene, the more those near clouds and the Jap fighter skimming them look wrong.
Constructive suggestions from you both, Martin and Matt, are taken on board, and will be applied to rectify things.
It is sometimes necessary to be "anal and rude" to recalcitrant boneheads. How else are they (am I) going to learn?

John