booleaned the new toes to the feet resculpts, and used linked objects for the claws to speed up the edits later.
This media is not supported in your browser
VIEW IN TELEGRAM
forgot to fix the pacing haha. (too much camera shake though...)
This media is not supported in your browser
VIEW IN TELEGRAM
Breakdown of how that could work.
one pass of a subdivison surface to get the initial mesh density, giving the next modifier enough to work with.
Pair the height maps with a displace modifier to control the height and intensity to bring in the main displacements.
Run another subdivision modifier afterwards to even out the forms and bring up the mesh density to render quality.
Base Mesh: 14K Faces / 27.6K Tris
First Pass: 47.2K Faces / 94K Tris
After Modifiers: 135K Faces / 269.3K Tris
Not every efficient to uniformly subdivide the mesh even once to say the least. The base mesh would be importable under the polygon limits of something like VR Chat ( 20K faces max)
No plans to do that yet, just an example.
But... for non-realtime 3d Renders or targeting future hardware, it could work fine, or relying on tessellation an existing engine.
one pass of a subdivison surface to get the initial mesh density, giving the next modifier enough to work with.
Pair the height maps with a displace modifier to control the height and intensity to bring in the main displacements.
Run another subdivision modifier afterwards to even out the forms and bring up the mesh density to render quality.
Base Mesh: 14K Faces / 27.6K Tris
First Pass: 47.2K Faces / 94K Tris
After Modifiers: 135K Faces / 269.3K Tris
Not every efficient to uniformly subdivide the mesh even once to say the least. The base mesh would be importable under the polygon limits of something like VR Chat ( 20K faces max)
No plans to do that yet, just an example.
But... for non-realtime 3d Renders or targeting future hardware, it could work fine, or relying on tessellation an existing engine.