Este fue un proyecto de una semana, el objetivo era forzar el uso de n-gons(caras de mas de cuatro lados) y ver hasta donde se podía llegar y con cuanta fluidez antes de que saltaran errores.
This was a one week exercise, the idea was to force the use of n-gons(faces with more than four sides) and see how far can a go before errors appear.
Environment HDRI by = Greg Zaal http://adaptivesamples.com
High res --> Ver aquí / See Here
En la siguiente imagen podemos ver algunos artefactos a los costados del modelo en la parte azul, eso se debe a que las cuatro puntas del segmento(el segmento tiene muchos vertices, menciono cuatro puntas para definir la forma) no se encuentran en un plano perfecto, lo que da al momento de la subdivision automática esos quiebres en el reflejo, esto es mucho mas evidente si el reflejo no tiene un poco de glossy.
In the next image we can see some artifacts on the sides of the model in the blue area.
This is due to the segment`s borders were not perfectly align on the plane and that made weird edges that breaks the reflection, this is lot more noticeable if the reflection has not a bit of glossiness
In the next image we can see some artifacts on the sides of the model in the blue area.
This is due to the segment`s borders were not perfectly align on the plane and that made weird edges that breaks the reflection, this is lot more noticeable if the reflection has not a bit of glossiness
Why didn't i fix it? Would'n be easy to fix by flattening the faces? Because the model accomplish its objective, i was able to drive a mid complex model to a point were it fails. This is something that can be use as a reference, something to be consider into future projects were to visualice the results keeping in mind the technique that its going to be used to get it.
No hay comentarios:
Publicar un comentario