Hi. I need to render an image at about 30000 pixels. My camera is set, and I am wondering if I can somehow split my camera into 3 or 4 cameras to make the process a bit easier. It’s almost like having fixed render regions.
These are the aspects I would like to achieve with your answer:
Blender crashed when I tried to stop the render after I spotted a mistake. Closing the render at these high resolutions is not a good idea.
I would like to render the whole image once and then work on a particular part of it that I need to update. No need to generate the entire render for each iteration.
Is it possible to have the camera I set divided into four so that each sub-camera’s outcomes will fit PERFECTLY into one big image in photoshop?
Thank you, I hope you can help me with this…I am starting to feel I am in deep trouble.
I think that you need to split the whole image into some tiles and after that compositing on 2D software like PS or Gimp.
To create the split, a quick research give methis addon.
I don’t know if it works today.
I think that something similart was posted also here.
This script is eight years old. I am not sure this will work. I can try. I do not know why, I thought there were more updated solutions or maybe a plugin that splits a camera into sub cameras.
This might work because I could set 4 cameras with four different render regions that altogether compose my image. I will make a test before creating the whole F wall.
Thank you
I am using the render border plugin now. This saves the project!
Thank you all for the insights and the excellent recommendation. This works so well and easy to understand. I will double-check if the stitching outcome works well. I imagine it’s precise.
Oh , no…I can not divide it into different cameras cause the “render border” allows only one setup per scene. If I move the render region of one camera, then move all the other render regions.
Edit: Ok I found out. The crop sizes can be keyframed, so one camera will frame different crops at different keyframes.
This isn’t that hard to do with just setting render regions (ctrl+b) for each quarter and render it out. Make sure they overlap a bit. Then in PS you just put each image in the right corner and you’re done.
Recently I had to render a project in six parts because of the size (4m x 3m @ 300ppi) and I used this method.
wow, that looks great too. Do you know if you can choose which tiles to render? From the description, I can see it subdivides the image, but I did not find if I could pick which one to generate. I do not want to render the whole image everytime I press “render”
Yes, you can start at specified/selected tile and after it’s done, simply quit rendering (Esc).
Only issue I found (and still investigating) is, it can’t handle custom OCIO configurations.
I have a problem when I use SRR addon, it will automaticlly output open EXR files no matter which format I selected before. But when I open exr files in Photoshop, the color of the images are not the same
as in Blender viewport. Do you know how could I adjust the color of exr files? Or could I output jpg or png files by using SRR addon?
