Friday 9 November 2012

Week 7 Extended! Material Solutions and Overall Evaluation

i have extended this blog due to forgetting to do the mandatory subjects told by us on the VLE. So in this extra Blog i will evaluate overall the three completed models and discuss the Mapping methods that'd be required for the Models.

And Mine
Theirs
via http://good-wallpapers.com/pictures/3339/imperial_star_destroyer.jpg
We begin with the first Model, the Star Destroyer. Undoubtedly the most difficult of the three, the model itself shaped out very well. As you can see from the comparisons, the shape is almost identical - not including some of the smaller and finer details upon it. Comparing the attention to detail i do admit i could have been less strict with what details to use, and thereby less polygon-strict, i also believe that i should've worked more on the sides of the ship, as there are almost hexagon shaped gaps within the central line on the sides of the ship. I do also admit the mistake of using Boolean - a mistake i sadly made a lot - which had put me in more predicaments than aid. I think what did confuse me a bit was that the references weren't all the same exact scale, as the side reference had to be stretched a few pixels to fit the 'reference box' made by the planes. Regarding the animation, i feel i did a pretty good job, although there is the one close-up shot which could've used the detail on the top half of the ship, the main point of the Star Destroyer is to be destroyed - however, i do feel that i might have to scale the ship to be larger or the Tie Fighters to be Smaller.
See the Hexagon Gaps?
     With the use of Materials, i think the Star Destroyer could be mapped quite simply using unwrapping technique and box, sphere and Planar mapping. The only trouble with mapping will probably be the back, the engines and the lower half of the raised platforms. However, the texture itself would play a vital part in the detail of the ship, as much of it can be relied upon it to get the smaller details aboard it. I may also have used more of the Outline and Inset tools had i learned them before and probably used more of the Chamfer and Bevel tools and the Local Normal option while Extruding.
If i were to redo this model with the knowledge that i know now, i would probably work more on smaller detailing and using different primitives more often (like tubes). In addition, I probably would've started on a Modified Box rather than a Pyramid to get the base shape of the Star Destroyer. As a first large-based model which required an efficient polygon count, i would say that although it isn't the greatest of models, it is still a good and moderately detailed one.


Their Tie
via http://images3.wikia.nocookie.net/__cb20080313170859/starwars/images/0/03/Tiefighterfull.jpg
My Tie
Moving on to the Tie Fighter, the ship itself is considered an easy and popular one, mainly because only half needs to be done and the other mirrored by the Mirror Tool. As a model, its shape is almost perfect - aside from the front window missing. In comparison to the Tie Fighter itself and its involvement in the animation, the little detail isn't much of a worry - as the only Tie Fighter to get close up is both spinning out of control and instantly flies off-screen. Although i minimised the Boolean in this model i do admit that i could've been more detailed on the wings, however i do also find that given the lack of an Outer Wing reference, whereby i had to find my own one, it was clear that there was little detail there anyway. Overall it was considered an easy build, but at the same time i am happy with the way it turned out; even if the use of a sphere increased the Polygon count.
Regarding Materials, a simple Sphere map would do the base given the correct material is given, the Wing arms itself will have to be mapped by Unwrapping, where many of the details will be placed on other 2D Shapes to be mapped onto the faces of the primitives. The Wings would require either a Planar map on both sides or a Box Map with a texture that shows both sides of the wing almost side to side.
To evaluate the model, i do admit there are some details that could've been added alongside the blasters but i also believe that i excluded them mainly because they could've been placed there via texture. In using them for the animation, i might have to modify the size and make copies as 'Extras' in the final assignment.

Mine
And finally we reach to the Engine Room. The Engine room was difficult, problematic and creative as it didn't have a reference. Although i regret doing most of it by eye and overusing the Boolean and Extrude method, i do believe it was a great effort provided that i only had one image to go by. In the use of the Animation, the engine room itself is immobile and only seen for a few seconds, meaning overall detail is not particularly needed.
Theirs
At first glance, i would say that many of the parts in the ship would be rendered nearly impossible to texture, however, the Unwrapping shouldn't be too large given that there are many large polygons in the model. In addition, i also believe that this doesn't need as much detailed texture as the other two and could rely on Splatter textures or possibly just lighting effects (which we should touch on in the next blog).
To conclude the evaluation on the Engine Room, it is probably the poorest of the three; at a high Polygon Count and many visible mistakes made on it, it is evident that there is much to improve about it. However, if i were to do this again, i probably wouldn't try so hard to show different and new techniques, even if i have discovered them by accident.

To evaluate all three models overall, i think there was much that could've been done to avoid such mistakes, however as this is a first try for many of us, i do believe that the effort put into the task and regulations at hand are greater than the mistakes made. As it is said on a Dreamworks Animated film "To Error is Human", and it is good that we both make these mistakes and overall learn from them in future models.

No comments:

Post a Comment