The other day I was thinking on the way to work:
Today everything is going physical... Everyone is trying to reproduce realworld, in eather the rendering, the motion or modeling. Maxwell is a good example, physics based renderer where you can set true world physics properties of shader and etc. Max's reactor engine which fantasticly simulates cloth, rigids and etc. Motion captures of human actions/behavior expressions and etc. Technology has really progressed for the last 5 years.
But, where does modeling stand today? same place it was 5 years ago. The basics of creating models are the same. Build the cage, smooth it, make sure it works, build in wrinckles creases etc, then paint your zbrush maps displace your model and go onto rigging. The model ends up limited to what it has been prebuild to. (and takes long time to do)
But I belive that this is not the future way. I think as others, modeling will also go physics based. For example:
You model your base cage, then go into every crease, fold anywhere where there is skin wrinckles and code the physical properties of human (or any skin) so that when there is transformation the program calculates based on true physics the amount of skin folds, wrinckles (based on musculars or whatever) and performs the creases, folds and other physical motions. The progressive skin will then behave like meta mesh or fluid mesh, where it will change and adopt to required amount of creases in a particular area.
I know this is not simple thing, and probably today it is not possible. But down the road 10-20 years from now it will be.
Just thought I share my thoughs with you guys.
Let me know what you think.
Today everything is going physical... Everyone is trying to reproduce realworld, in eather the rendering, the motion or modeling. Maxwell is a good example, physics based renderer where you can set true world physics properties of shader and etc. Max's reactor engine which fantasticly simulates cloth, rigids and etc. Motion captures of human actions/behavior expressions and etc. Technology has really progressed for the last 5 years.
But, where does modeling stand today? same place it was 5 years ago. The basics of creating models are the same. Build the cage, smooth it, make sure it works, build in wrinckles creases etc, then paint your zbrush maps displace your model and go onto rigging. The model ends up limited to what it has been prebuild to. (and takes long time to do)
But I belive that this is not the future way. I think as others, modeling will also go physics based. For example:
You model your base cage, then go into every crease, fold anywhere where there is skin wrinckles and code the physical properties of human (or any skin) so that when there is transformation the program calculates based on true physics the amount of skin folds, wrinckles (based on musculars or whatever) and performs the creases, folds and other physical motions. The progressive skin will then behave like meta mesh or fluid mesh, where it will change and adopt to required amount of creases in a particular area.
I know this is not simple thing, and probably today it is not possible. But down the road 10-20 years from now it will be.
Just thought I share my thoughs with you guys.
Let me know what you think.
Comment