problem with orthographic camera

Use this forum for general user support and related questions.
Fox
Posts: 318
Joined: Sat Mar 31, 2018 11:17 am

Re: problem with orthographic camera

Post by Fox » Sun Feb 10, 2019 10:54 pm

I get this result in Blender and 2.1 beta4 after i apply rota and scale for all the objects other than sun and camera.
Fox test.jpg

mick
Posts: 80
Joined: Mon May 21, 2018 7:57 pm

Re: problem with orthographic camera

Post by mick » Sun Feb 10, 2019 11:40 pm

Dade wrote:
Sun Feb 10, 2019 10:41 pm
I have yet to check where exactly is the problem but the root is in "scene.camera.screenwindow" for sure.
I don't understand why screenwindow should cause this.

The issue is that light intensity is missing in directly irradiated area compared to shadow area.

Here another example:
Untitled.png
Below the thick glass is dark shadow: good
There is more light below the thin glass, with the gloss from the sun: good (but actually there shouldn't be that gloss)
The other areas are darker than the area below the thin glass, and no gloss from the sun: incorrect

Here again how it should look like (rendered with distant perspective camera):
Untitled2.png

Fox
Posts: 318
Joined: Sat Mar 31, 2018 11:17 am

Re: problem with orthographic camera

Post by Fox » Mon Feb 11, 2019 1:04 am

Seems like orthographic camera scale of 10 is infinity times brighter, scale 1 is much better and perspective camera is perfect.
Switch the auto linear off first.

mick
Posts: 80
Joined: Mon May 21, 2018 7:57 pm

Re: problem with orthographic camera

Post by mick » Mon Feb 11, 2019 8:36 am

Just a quick test with manual linear 0.00001
Untitled.png
Shadow below thin glass is still much brighter than directly irradiated area.

I don't think that it has to do with the tone mapper. Maybe ray sampling???

CodeHD
Posts: 182
Joined: Tue Dec 11, 2018 12:38 pm

Re: problem with orthographic camera

Post by CodeHD » Tue Feb 12, 2019 11:33 am

We already had the topic of glass behaviour recently in another thread:

viewtopic.php?f=4&t=793

I also posted an example of a working lens system with a detector made of matte translucent plate and an ortho-camera there. Check out these two posts for images and the blend-file:

viewtopic.php?f=4&t=793&start=10#p8335
viewtopic.php?f=4&t=793&start=20#p8384

Not sure if it is exactly what you are going for here (didn't have time to check out your file) but since you said you need to simulate optical lenses I though it might be helpful...

mick
Posts: 80
Joined: Mon May 21, 2018 7:57 pm

Re: problem with orthographic camera

Post by mick » Tue Feb 12, 2019 1:04 pm

Thanks, I'll have to check later.

Just for curiosity, why do you measure from a translucent plate. This is my real world setup. But for Blender/LuxCoreRender I can just put the orthographic camera in front of an opaque plate, right? All other objects are camera invisible, and the camera anyway.

CodeHD
Posts: 182
Joined: Tue Dec 11, 2018 12:38 pm

Re: problem with orthographic camera

Post by CodeHD » Tue Feb 12, 2019 3:55 pm

The "matte translucent plate + ortho camera behin the plate" setup is one workaround to mimic a CCD/CMOS-detector instead of a camera.

I found that the matte translucent plate needs to be thin, but still have a finite thickness. In my case, if it was thinner than 1e-5, there would be strange behaviour. If it is thick, it blurs the image.

It should be possible as well to have a "matte plate + ortho camera in front of it".
My idea - which I didn't play around with further - was the additional settings matte translucent gives me, potentially combining it with glass so that your light path goes through the sequence "air - glass surface - glass body - matte translucent surface - air - camera". This could be a way to simulate specular reflections off the detector surface for simulating ghosts.

Post Reply