@klknv
Thanks! This is great feedback, your render look really nice btw!
The Lens Sim Setup uses an orthographic camera to work that has to have a scale of 1.0. Pressing Disable Lenses will convert it back to a standard perspective camera that represents the lens as closely as possible(it will also remove the constraint). When disabled, navigation methods like First Person View will function as normal. I think this is the best way to work around the orthographic camera. It’s not perfect, so if you have any suggestions please let me know.
I just wanted to address this before the anamorph setup.
I’m not a photographer, so my philosophy around the focal length of each lens is to utilize the largest possible image area for that lens. I also think this addon is created for extra lens artifacts, so having a larger sensor will maximize these effects. Therefore, I’ve set the Best Fit sensor size to the largest usable image area. This is of course not realistic, hence the varying focal lengths you get from each lens compared to the actual mm you see on the name.
I don’t really know what frame sizes each lens should use, so I encourage people to experiment to their likings. And advance users like you can always pick a realistic sensor size to their liking
I haven’t found a good way to do this, and I think I can explain what the issue here is if you want an exact anamorphic setup. I will probably repeat some of the things you said so that I can make sense of the process.
Blenders camera frame will always be inside this square box, regardless of how the render size or squeeze factor on the blender camera is set to. Setting the Sensor size will determine the size of the actual sensor fitted into this box. So setting a sensor size of 24mm will look like this:
So here comes the problem, if I got it right you want to render with the correct pixel squeeze and at the same time see it normally in the viewport.
We start with the expected result:
With a 2.40:1 ratio and 1920x800 pixel size as the final result we should divide 1920 / 2.40 to get the raw sensor pixel size: 800x800. By not desqueezing the image we get the exact result we want in the render. The sensor size matches and if unsqueezed by 2.4 we get the wanted result.
Now the setup:
To get the correct unsqueezed results in the viewport, but the correct squeezed result in the render. This is what we can do: We have to unsqueeze the image and multiply the Sensor Width by 2.40 to get the correct unsequenced image width inside our image square. Then we set the render pixel aspect to 2.40 so that the height is compensated and the wanted rendered image is being shown in the viewport and rendered correctly.
Did that make sense? I think there is room for improvement here in terms of having some extra buttons or sliders to make the conversion easier to set up.
The built in anamorphs lenses are also varying the squeeze factor to compensate for the focus breathing. With this setup you probably want to override the squeeze factor and and do the desqueezing yourself in comp.
I haven’t thought about a good way to do this, so feel free to let me know what you think
I wish we could have individual camera render resolution so that we don’t need to override the whole scene for a setup like this.