I am currently working on a few tutorials on dynamic shattering of geometry.
Now I have come across an issue: I have a simulation where a ball hits a sheet of glass and breaks it. As the shattering occurred before the ball touched the glass, I increased the substeps on the dop network from 1 to 2. Now it works okay.
In the tutorial I am doing, they then increased the substeps to 5 which worked in the tutorial but not when I recreate it.
In their case, the sheet still breaks upon impact of the ball. in my case, it breaks almost as soon as the sim starts (with the ball being nowhere close).
I assumed that increasing substeps has an impact on simulation precision, not its timing…
Can anyone shed some light on this? (I have attached the project file).
Dynamic shattering substep question2123 3 1
jeff wagner goes over this in detail in this masterclass
- Quick Links