#cg toolkit

Assessment 2 Final submission

The series of assets developed for the project included 4 dynamic buildings and a series of modular ground pieces.

Each building is individually constructed of between 7 and 31 individual pieces that upon placing it in the 3d environment automatically randomizes itself to create a unique structure, though most of them retained the same silhouette, some modifications significantly changed the design of the building, making it significantly different.

The construction scripts in each of the blueprints were relatively simple though did in some occasions become quite large and awkward to manage. On reflection I would have preferred to split the buildings up further in order to split the silhouette up further and allow each structure to be more unique in its variations like the following building.

The Tower created for the project was an interesting piece though the design did not translate very well from the concept art. Like the other designs I would prefer to break up the design further next time and add further elements to the design.

The modular ground pieces, though fitting together well and functioning as required appear too static, and require more elements to break it up. Next time I would create a number of objects such as garden areas to fill the space and make the environment more interesting.

Many of the buildings also would have benefited from smaller details such as air-conditioning machines or electrical boxes, these smaller details are less noticeable, though still important, and could also be made to utilize the same randomizing technique that the rest of the buildings use.

This week we looked at introducing character and character animation into unreal engine. To begin this week i had a lot of trouble trying to get the ARTv1 from the epic-games market place to work. I looked around to see what could be causing the issue with it but had no luck, i was especially confused as i was following along with the video tutorials. though today(monday 14th) i realised that i had been putting the usersetup.py in the wrong script fold. i had been putting it in the location on the drive maya is installed on rather than the my documents file path. once i had fixed this error of mine i set to work on understanding how to use ARTv1 in maya and moving between maya and unreal.

though from this weeks task i realised that all the models that are used by the art tool are rather basic they still have massive potential for use in general animation and in unreal. Like with the only needing to create one skelton/mesh and then applying multiple animation to the same skeleton and the same the other way.

I think that using the ART tool early on in the production process is best for a pipeline as it allows animators to work with a base model and start build an animation backlog that can be altered later if need be. but it also allows animators to see how the character and animation will interact with the world that they create.

Time breakdown

Troubleshooting - 7-8 hours

working-3-4 hours

experimenting- 1-2 hours

Reflection

Before starting this project/assignment, my knowledge was limited when it came to working in unreal engine. I had only imported and created a scene using models outside of unreal. though with having completed the majority of the grey boxing for the scene I was very satisfied with the result that I was able to produce whilst still learning and adjusting to the building in unreal.

I found the process of grey boxing fun, you are able to achieve good results in a lot less time than traditional modelling. which is nice for creating a rough layout of a scene or environment, but I also enjoy the simple appeal of grey boxing and its style. I like the fact that you could grey box a scene and test out camera movements with ease, unlike other 3D programs where the task can be long and tiring. 

Through some trial and error at the beginning of the project, I found that getting the scale set up right at the start makes life easier in the long run. I did this by creating a cylinder that was the average height of a person (170cm) to then compare to the structures that I was creating.

The render time using unreal is great I had no trouble rendering out the final images sequence in 10-20 seconds which made seeing results or changes in the final version good. But also real times’ ability to always run smooth and not get caught up in frame drops really is nice compared to sometimes in other 3D packages. 

Break down of the week

5-6 hours build and adjusting to Unreal

2-3 hours looking at Epic games out on sequencer and various other videos.