This content has been marked as final. Show 6 replies
Is that screenshot from max's export preview? If so, try opening the actual w3d - I think the problem is just in the export window, and not in the actual w3d.
I just tested with a 2048x1024 texture, I saw the same 'noise' as you in max's preview window, but when opening the w3d in director it looked fine.
hope this helps,
couple more points:
* I tried 4096x2048, also had noise in the max preview, but displayed fine in the resulting w3d.
* non power-of-two sized textures will be automatically resized to the nearest power of two texture (might always be downwards actually), so it's not worth bothering with sizes like 1536X768 or 3072X1536, because they won't actually be displayed at that resolution anyway.
Sorry, one more thought
If you want to generate the images for the 6 sides of a cubic VR, just create a camera and adjust its FOV to 90 degrees, place it in the centre of your scene, and use the standard render option to render a square image (power of two dimensions, remember) in each of the 6 directions. Make sure 'angle snap' is on, and just rotate your cam to get each of the views.
Thank you for taking the time to help me out! It's been very frusterating, to say the least....
Fantastic advice Ben and I will try it right now, to see if it works! I' will post again soon to let you know..
As for the cube part, do you know how to use the Reflect/Refract map option to render out cube panoramas? I know it exists, but am having trouble using it...mainly, what object I am supposed to "select" when it tells me to pick an object.
Basically, I just want a camera in the 3d room and have it automatically generate the 6 graphics of the room for me. Just not sure how to have it generate those cube maps... Do I select the "camera," a "wall"..?
You might also want to have a look at my SPi-V engine, a Shockwave3d
based panorama viewer. It reads normal images, as opposed to a w3d model
with images wrapped to it, and it accepts arbitrary image sizes. The
images are chopped up into 'powers of 2' chunks before they are
displayed, to ensure maximum image quality.
The SPi-V movie can be embedded into Director projects as an LDM.
Documentation is available here:
Not sure how to do it automatically, never tried it. But it's pretty simple to do it manually using the method I described - that's how I've always done it. (camera with FOV 90 degrees, render square images, and manually turn the camera to each of the 6 sides).