2
$\begingroup$

I love adaptive subdivision. For those that don't know, its a really cool feature that lowers subdivisions to meaningless values as an object increases its distance from the camera and increases it as the object gets closer. This saves up to 90% RAM usage and can slash rendering time in half!

While this feature is cool and all, I use dynamic topology a lot while sculpting, and I was wondering if anyone could explain to me the difference on both a mesh/code/render level?

For reference (i.e. the research I've done):

Adaptive subdivision

Micropolygon displacement

Dyntopo

Improve this question
$\endgroup$
0

2 Answers 2

Reset to default
8
$\begingroup$

The difference between Adaptive subdivision and Dynamic topology is the way the surface gets subdivided. Both triangulate the mesh but use different algorithms.

Adaptive subdivision:

enter image description here

  • each polygon is divided into a number of micro-polygons based on camera distance and screen pixel size
  • these micro-polygons are displaced at render-time with texture
  • the subdivision algorithm makes it easy to increase/decrease the density
  • resulting triangles are nice (a quad results in smaller quads that form 2 triangles each)

Dynamic topology:

enter image description here

  • the division algorithm divides the mesh based on brush size while sculpting the mesh
  • the created vertices are created at their final destination
  • decreasing/increasing resolution is done by collapsing/subdividing edges locally with brushes and globally in tool panel under Dyntopo settings
  • local subdivision and collapsing of edges usually create nasty geometry with long thin triangles and triangle fans.
Improve this answer
$\endgroup$
1
  • $\begingroup$ I agree Jerryno may have the better answer to the question. $\endgroup$
    – hawkenfox
    Commented Mar 18, 2017 at 17:22
2
$\begingroup$

Adaptive Subdivision A operation to optimize what is happening in a real-time environment like in a game engine. What is happening there is that game probably have a lot of high-resolution models or high numbers of models going on during execution. So Adaptive Subdivision is a way to cut down all the polygons from showing up all the time. If it's not important or far away don't show them. Like pycoder said, it's a level of detail (LOD) system. Design to display real-time environment efficiently.

For the adaptive option on the open subdiv tool however, it's also included as a tool for real-time surface tessellation at cpu/gpu render time. Especially in an animated sequence where your camera will be moving closer or further away from the objects in the scene.

Dynamic Topo It's more like a tool for 3d content creation. In the past, they have multi subdivision to do the job. What happens is that we end up with a sculpt that has lots of polygons, regardless if there are any necessary details on the surface. The whole model is ramped up with polygon every other higher subdivision steps.

Dynamic Topo solves that problem by adding only necessary polygons density to necessary parts where there are actually details or sculpted area that really needed those extra polygons to display them. Resulting in a more efficient workflow, and saves you more resources, money and time.

Conclusion They both does one thing in common, that is to make rendering 3d object on screen more efficient. The difference is adaptive subdivision is purely a real-time environment display mechanism and the other a tool for 3d artist.

I am afraid at the coding level, I know too little to help with that. But that's all I know about the 2 topic.

Improve this answer
$\endgroup$
2
  • 1
    $\begingroup$ Please add a comment if you downvoted. So I can see how I can help to answer your queries further. $\endgroup$
    – hawkenfox
    Commented Mar 18, 2017 at 5:10
  • $\begingroup$ Hi hawkenfox, I think the answer misleads the reader. Both methods have real-time performance, adaptive subdiv is not limited to environments or game-engines. It is also not culling and does not cut down polygons from showing up - it increases polygon count. You answered the difference of usage more than how they really differ. $\endgroup$
    – Jaroslav Jerryno Novotny
    Commented Mar 18, 2017 at 12:32

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .