Tuesday, 18 June 2013

HA4 - Task 6


Polygon Count and File Size

Two measurements of a objects file size is the polygon and vertex count, for instance, a character in the games that we play may stretch from 400 -500 polygons to over 35,000+ polygons.

Polygons and Triangles

 When a game concept artist starts to talk about the number of polygons in a model weather it is a character or a building, they are actually talking about the triangle count. Games mainly well, almost always use triangles and not polygons as in today’s world nearly every single graphic hardware is built to accelerate the rendering of triangles.

Although polygons do have a very useful purpose in the gaming industry as a model made of mainly four sided polygon (called a quad) these will definitely work well with edge-loop selection and some transform methods that will speed up the modeling, this will make it easier to review how the model flows. Most Artists mainly preserve the polygons in their models as long as possible.

Triangle Count and Vertex Count

The vertex count is initially a lot more important for the performance and memory rather than the triangle count but for many reasons artists use the triangle count because it’s more common for them to use as the performance measurement. A way to look at it on the most basic level is that the triangle and the vertex count can be comparable if every single triangle is connected to one another for instance: 1 triangle uses 3 vertices, 2 triangles use 4 vertices and 3 triangles use 5 vertices and so on.
                   
The overuse of the smoothing groups and too many material assignments lead to a much bigger vertex count, this will put stress on the transforming stages for the model, basically slowing the performance, it can also increase the memory cost for the mesh because there are lots of vertices to store.


Rendering Time

When rendering is the final process of creating the actual 2D/3D or animation from the prepared scene. This is slightly similar to taking a photo after the scene is setup is finished

Many rendering methods have been developed and they can be specialized, these can range from the clearly non-realistic wireframe rendering over a polygon-based rendering this can lead into more advanced techniques such as: ray tracing and radiosity. The time It takes to render can take up from seconds to days for either a solitary image or a number of frames.


There are two different types of rendering ‘real-time’ and ‘Non real-time.

Real-time


Real-time is for rendering interactive media such as games and simulations which can be calculated and displayed in real time which the frames per second can range approximately from 20 to 120. The goal in real time is basically to show as much information as the eye can process in a fraction of a second, for instance, one frame.


Non Real-time
Non Real-time is for non-interactive media, for example using in feature films and videos, they are rendered at a much slower rate. Non-real time rendering basically enables the advantage of limited processing power in order to maintain a high quality image. Times when rendering for every single frame can vary from a few seconds to several days for very complex scenes. However rendered frames can be stored on a hard disk then it can be transferred to other media sources such as motion picture or optical disk. These frames can then be displayed at very high frame rates, typically 24, 25 or 30 fps (frames per second) this can achieve the illusion of something moving.  



No comments:

Post a Comment