Hi all,
In dec 16 I started a similar thread:
I seemed like we all agreed that
- SU cannot use multi threading because it is basically a modelling software, not a renderer.
- “SU is a polygon modeler based on lines and faces defined by facettes whereas NURBS modelers in the CAD area as e.g. Revit/Inventor etc. are based on smooth cuves and surfaces defined by (much less) curves.” This also makes a prestanda difference.
But I still can’t help thinking “What-if” …
So, if we focus deep into the SU system’s workflow here, isn’t it like this:
There a only TWO basic activities: A.Changing the model (for instance moving a chair in a room) and B.Changing camera and rendersettings (position,zoom,pan,fog etc)
In real time A and B are split up into micro-steps after which a re-RENDER is performed. So it is the render that makes a very big model lag so much after every little change, not the change itself.
This makes me think that if SU incorporated a multithreading render engine, this time critical render-step could be boosted very,very powerful. What if?
Cstrom