Renderling quality

Whenever you encounter any issues with Enscape or your subscription please reach out to our dedicated support team directly through the Help Center or by using the Feedback button as detailed here.
  • Interesting -

    With more sampling on Ultra is it overcompensating for the lack of lighting data?

    I also don't understand why

    even though i use ambient Brightness for 100

    it doesn't improve

    and worse when you increase the brightness of the sun


    ambient brightness for 100



    sun brightness for 100%


    I am satisfied with " high":)

  • I have found myself "just" using HIGH instead of ULTRA as well. Regarding those blotches, I get them frequently with surfaces in darker areas of the model that have reflections and bump, even if minor. Even my basic "paint/drywall" texture which has some normal and reflection applied. Sometimes I just remove bump/reflection altogether as the added detail isn't worth the quality destruction. Particularly in these high contrast multiple room environments.


    Probably in a pinch I'd just run the image through Topaz Denoise and try to remove the blotches that way.


    OR put a simple artificial light source behind the camera to compensate for the large dynamic range here.

    • Official Post

    It looks like the room where the camera is located doesn't have any light coming in except through the doorway from the other room?


    In that case the results are not entirely surprising (eventhough I haven't seen such noisy results for quite some time in a capturing):

    Quality 'High' only considers single bounce indirect lighting. In the above scenario that'd mean, that we never actually reach the bright spots of sun light falling into the other room, but rather hit surfaces in the room the camera is located. They will be lit rather homogenously by the asumed ambient brightness -> therefore there's not much variance between indirect lighting samples and therefore not much noise.


    For Quality 'Ultra' multiple light bounces are calculated - this means that some of the rays do have a chance of hitting the bright spots of sun light in the other room, but that probability is most likely still rather low. But when a sample does hit those very bright areas it's going to transport a lot higher radiance then the other samples which don't hit any surface that receives direct sun light. This leads to a high variance between samples and thus unfortunately to more noise - which is filtered, but obviously not rigorously enough for a clean image.


    As has been suggested: It can help to place at least some artificial lights into the rooms, if there's otherwise not much direct light to reduce noise.


    In general we're of course aware of the problem and hope we can improve both on the denoising of the image, but also on the multi-bounce lighting quality in general in the future.