I’d consider breaking up your render into smaller ones, each one being a fraction of the total image.
You could then join the renders in an image software like photoshop or gimp (maybe even blender compositor).
You can do that using border and crop tools in the render panel (or Ctrl+B in camera mode).
Another option is to shift the camera in the camera settings.
There probably are addons that can manage this as well but I don’t know any right now.
I just checked again the amount of pixels, 3.32m = 9411px and 2.66m = 7540px. I don’t know how my colleague got this size. However I suppose this is 72dpi in blender, don’t know yet how to change it.
Hi,
You don’t care if blender sets the output image to 72dpi, just the pixel number matters. Once you got the render, just change the dpi for your printing.
If this dimension is still too big for your computer, you can use a renderfarm.
This doesn’t “jive” with what my printer told me when I was creating a graphic (of a 1930’s era mural) that was going to cover the length of an entire wall. I suggest that you should “talk tech” with whomever you’re going to engage to produce this graphic, and ask them what image-resolution they require and recommend. Follow their instructions exactly and don’t guess what they will be. (I tried to, at first.)
When you’re even ten feet away from an image, you cannot see “the pixels.” What you do see is what’s referred to as the [URL="http://en.wikipedia.org/wiki/Acutance"acutance of the complete image. This is a characteristic of how the image data is prepared and it has little to do with physical resolution.
Also bear in mind that print media (of all sorts) presents something to your eye that is made up of reflected (environment-sourced) light, in the CMYK color-space, whereas all CGI work is initially done with generated light in RGB. These are entirely different worlds that play by entirely different rules. I seriously doubt that you actually need this kind of resolution.