If I render perspectives with "normal" cameras I get images which are congruent pixel by pixel. That means, if I render a region and paste it in Photoshop into the complete image there are no seams, no shift, you can not see where the insert is. If I use a camera with orthographic projection with every new rendering the image is a little bit bigger or smaller and is a little bit more left or right. This is very annoying if you have to do a lot of changes where no time for a complete new rendering is available. Does anybody else made similar observations. What do I wrong. The camera is locked. The resolution is the same, "show safe frames" is on. If I switch through different views and then back to my camera with orthographic projection all seems fine (tested it by overlaying an editor window fitting the outlines of my building). But the renderings have these small differences. How can I obtain identical sized renderings?
Solved! Go to Solution.
Solved by tony.su. Go to Solution.
I cann't reproduce your issue. Because I'm not sure how did you locked the camera.
When you use orthographic camera, lock move, rotate, scale, lens, fov is not enough. Tatget distance will be not changed. But in some case, we could change this setting by accident.
Hello Tony,
thank you for your tips. I found out what the FOV changes. It is my 3D-Mouse, Space Navigator from 3D-Connection. It is so sensitive against vibrations of my desk that it changes the FOV a little. So little that I didn´t see changes in the viewport. Only in the renderings.
I have locked movement and rotation of the camera under Hierarchy --> Link Info. But I didn´t find a way to lock the FOV of the camera. Can you help me? Thanks a lot!