Hello,
Unreal Version: 4.11 Preview 7
I got to test this earlier and here are the results packed in a GIF:
As you can see sunlight “Movable” looks best and most accurate, “Static” is OK but useless in my case, “Stationary” being the most important looks inaccurate and has some flaws in finding shadow areas and/or working properly with the shader in softening out those shadows and absorption in this case. I do have a subsurface material here along with constant tessellation (without camera distance LOD for the test and baking). I tried various lightmap sizes but they all looked identical with just more strange sharpened shadows.
Also please note how Static and stationary Darken the intersection edges of the snow and the objects harshly (upper right and on stairs), while movable light displays softer more correct feel.
Any thoughts on this?
Thank you in advance.
Following up with this post, 4.11 has been released. I tested this again and the issue is still there. Any feedback on this or expected fix time?
Thank you.
Hello ,
Could you provide me with a few screenshots of you material’s set up so I can get some more context?
There could be a number of reasons why each mobility type for the Directional Light looks different. If you have not changed any of the settings for when the light is set to ‘Stationary’ compared to ‘Movable’, that could be a big part.
Remember that Stationary lights integrate static and dynamic lighting, so there will be indirect light bounces, differences in shadow appearance through Cascaded Shadow Maps, and other various settings.
If you can provide me with enough concrete images and steps so I can reproduce this in a blank project on my end, we can determine if it is in fact a bug, or just some missing steps. Let me know if you have further questions or need additional assistance.
Cheers,
I could provide some files if you could provide an email address to send it privately.
Thank you,
For security reasons, I cannot issue you my email address, but you can send me your files via a private message on the forums. You can find me by searching my name. It will show I have the Epic Staff tag.
Thanks,
Thanks , just sent you a PM.
Hey ,
So there are a few reasons as to why you are seeing such radical differences between the lighting mobility types. For starters, I simplified your material so it is just the ‘Base Color’ input, ‘World Displacement’, and ‘Tessellation’ being made available. I then changed it over from the ‘Subsurface’ shading model to the ‘Default Lit’ model. I did all of this to eliminate any outside variables that might cause differences in lighting. Subsurface, as you probably know, needs the opacity input and accounts for scattering within the objects surface. We definitely want to eliminate that for the test.
Secondly, I fixed your ‘World Settings’ and set them back to the default. You had some environment color which is going to effect the appearance of your Direct/Indirect Lighting when running Lightmass. The third and final settings I modified before running the comparison test, was to make sure the ‘Cascaded Shadow Maps’ settings matched when switching your Directional light between ‘Stationary’ and ‘Movable’, as well as unchecking the ‘Override Lightmap Resolution’ of the static mesh. Below are my results after making the mentioned changes.
Static Lighting
Stationary Lighting
Movable Light
As you can see, the ‘Stationary’ and ‘Movable’ outcomes are nearly identical. The Static lighting outcome is expected as it is going to take some tweaking to get the similar results. The end result being, this is not a bug, but just a need to modify your settings and values between lighting mobility types.
If you have questions about my process, please let me know.
Cheers,
I can’t thank you enough for taking the time and helping me out on this one! I believe I threw you on a tangent when the simplest issue was the “Cascaded shadow” parameters left at 0. After adjusting this, things worked just fine! even with the subsurface shader as is. Many thanks for your support!
P.S. (It is the stationary lights I care about most so it is fine if Static doesn’t match as well).
Glad I could help. Yeah, I noticed the impact immediately when I modified the ‘Cascaded Shadow Maps’ values of your Directional Light. I wanted to kinda walk you through the troubleshooting process I needed to take in order to determine the culprit, by eliminated any outside variables that could be missing with the final outcome 
