Hi guys! I sent this to @ValeroStudio already, but for the sake of documentation and visibility, I’m going to copy and paste the response I sent in case anyone else encounters this in the future:
The issue with shadows and face me’s boils down to how shadows are created and handled in Enscape. When either Hardware-accelerated Ray-Tracing or software ray-tracing (called “Ray-Traced Sun Shadows for captures” in 4.8) are turned off, Enscape doesn’t retrace shadows with each camera movement, meaning shadows are stuck in place. This is fine for most applications of Enscape because shadows usually only change if you add geometry, which automatically triggers retracing. But Face Me components don’t actually change geometry with each camera movement, which is why they’re so great, but is also why the shadow thing happens! This results in the shadows of a Face Me getting stuck at the original pivot point, which casts the weird shadow on the component’s axis point.
However, when both ray-tracing features are turned on, the shadows are redrawn with each camera movement, and shadows follow the component instead. I’ve made a short screen recording that you can view here of both behaviors using SU25 and Enscape 4.8.
Even with both ray-tracing options turned on, the shadow can still appear at the pivot point with specific camera angles. To lessen the chance of this happening, you can move the pivot point to the side, as seen here (please excuse my apparent inability to snap axis points…). However, this can make it difficult to place objects, so I recommend utilizing free-standing lines to serve as crosspoints instead.
Now, V-Ray handles this well by default because, as a ray-tracing engine, it actively re-traces shadows with each camera movement. Ray-tracing engines trade rendering time for powerful photorealism, while Enscape’s real-time engine trades true photorealism for time and performance. Here’s a comparison of Face Me’s with V-Ray and Enscape’s default settings. Notice how V-Ray pauses for a split second to configure moving the face while Enscape does it instantly? This is the engine difference at work: V-Ray takes time to process while Enscape just barrels through. (Here’s the comparison of them with all ray-tracing turned on, if you’re interested)
(NOTE: you might need to switch to the “Old Graphics Engine” in Sketchup 2024-2025 for Face Me’s to work properly in V-Ray)
Discrepancies between face-me positions are also caused by the fact that the face edge isn’t anchored to an actual point. Enscape can only guess where the face will fall, sometimes failing. V-Ray is better at predicting this due to the engine again: it has the time to figure this out thoughtfully.
Sketchup is the only program we support with a feature like this (you can add it via scripts in Rhino, but it runs into the same issue), and Sketchup’s code doesn’t differentiate Face Me components in a way that would reliably target only them to prevent this 100% of the time. Because Sketchup is the only one with this and Sketchup’s unique code requiring us to spend time excluding other pieces, we determined that the effort and scale it would take to improve our handling of Face Me’s just isn’t an economical use of our resources.
I hope this information is useful! Please let me know if there’s anything else I can answer!