Trying to create a wire frame object in 3D

Question

Hi Folks, I am trying to create two objects: something called a gravity well and a wire framed globe in Photoshop 3D CS6. The examples are here: http://www.carreonphotography.com/3D/   #'s 1/5 and 5/5. I tried using a gradient map to create a depth map but the form is not editable if I understand PS 3D correctly. Only the scale, light, texture and perspective.
The surace of the gravity well needs to be a  white wire grid pattern over black with. The globe is basically the same. Can this be accomplished on Photoshop CS6 3D or do I have to farm this out to a 3D guy.
I have included samples of my objectives: #1 and 5 plus screen shots of what I ended up with, the form of which was uneditable based on what I understand. I also don't know how I would create the wire frame once I have the form.
I deally I would use the final images as png's on a 2D suface which I would later wrap onto a 3D surface.
Any thoughs please!
Thanks folks in advance:
Ed Carreon
www.carreonphotography.com

Answer 1

Sanpanza wrote:
How do I eliminate the artifacts along the edges of the frontal vertices of the well?
I think I know what you're referring to, but to make sure we're on the same page, repost the image with red ellipses/boxes surrounding the artifacts which trouble you.
As recommended by someone on the forums I started with a 32 bit grid map but knocked it down to 16 bits to speed up the 3d engine. Was my thinking correct? Can I use a gray scale image instead so PS CS6 doesn't get bogged down?
I used an 8-bit grayscale image for the grid and I rendered to 16-bit output (i.e. my main document's mode was 16-bit RGB). Rendering to floating point 32-bit output images can have advantages when using illuminants with realistic intensities but it usually requires subsequent tone mapping. I don't think your illustrative image will benefit from that, so I recommend you render to 16-bit.
There is absolutely no reason to use a 32-bit per channel image for that grid opacity texture. The grid has only black and white and is entirely horizontal and vertical lines, so you should be able to get away with a 1-bit per pixel texture, but use an 8-bit grayscale texture because I know that works in Photoshop
The output of the final document is 5000 X 5000 pixels but the gravity well is but small element in the background so I have made it 3000 by 3000. If the 3D engjine is using vector math to draw the lines, does size even matter?
The polygons making up the surfaces that are rendered will be calculated with vector maths but the size of the opacity texture should be large enough to prevent it becoming fuzzy due to upscaling to cover the surface area when rendering an image of a given size. Remember that a texture can be tiled (up to ten times in each direction) to repeat over a surface (using the UV Mapping Properties), so your grid doesn't need to define every "square" of the well.
Can't figure out how to move a point light from the bottom so it matches the second example. I like the way this artist handles transparency in his grid. Can you tell me now to move the light. The interface is different than in CS5 tutorials.
Click on a point light with the Move Tool to activate its manipulator then drag on the heads of the red, green and blue arrows. You might want to switch view to each of the top, right and front views to aid alignment. You can store the current 3D view before changing then return to it. There's a secondary view window that can be enabled, which would aid with alignment, but it doesn't show lights for some reason.
Is it possible to select all the shape layers containing horizontal lines and change their width simultaneously? I am doing so one layer at a time.
Do you mean to scale the width of the lines without scaling their offset?