Wednesday 9 January 2013

Part 6- Blog Summary


In the Second semester my goal was to improve my Maya skills and heighten my understanding of photography, texturing, 3d modelling, rendering and many other features and techniques.  The aim for this third and final semester was for me to tackle the UDK engine for the second time to make it a playable level with more art ideas, then to carry out the design of the level.

The First thing I had to do was to try to import everything into UDK from Maya. There were of course problems such as the exporting of the meshes collision frame and the finding out that the mesh needed lightmaps. These Lightmaps took me a while to figure out and I still haven’t got them perfect even now at the end of semester 3. The features that are available in UDK were completely new to me as an artist but I persevered and got the vast majority imported into UDK. I still have a few problems with it which I hope to solve soon, mainly the terrain and landscape tools. I moulded the terrain the way I wanted and during texturing it completely glitched the textures whilst having this problem I found out that the terrain tool isn’t used anymore, my solution was to start from scratch and so I redesigned the layout of my level. Yet again there was a problem with the landscape tool that wasn’t cutting the hole in the landscape for my underground base, so I figure had to away around it through a cinematic and by teleportation.

One of my major problems I encountered was that I made the poly speed trees too high. From my artistic side they looked fantastic but in the editor it was a big problem. It took me a long time to figure out the reason why my level wouldn’t build the lighting, it was because the lightmass builder didn’t agree with the trees therefore I failed every time to build them and so I had to then make the trees in Maya by hand.

Even though I had these problems that I was struggling to understand I pushed ahead,  finding temporary answers them as I continued. From that point on I stopped listening to my artist side and started to knuckle down on the design of the underground level. The important assets such as the levers and movers were modelled first so I could layout the level quickly. Design ideas started to flow back to me as I produced the dissertation on game theory , I took all things into consideration and used colour theory for  improving the feel and atmosphere of the game, I guided players by using visuals and lights to show the player were to go. The time saving asset building I found came in really useful when planning the layout. 

The major problem in my project is my artistic and design creativity constantly being at odds with each other. The artist side usually wins because that’s my nature. I love see things that are aesthetically pleasing. This did affect my workflow heavily so my design side of me kicked in, but it was only at the end of the project even so it did enable me to make little change here and there to help problem solve. This also affected me in the final production lighting build because I was extremely keen to have a hole in the landscape but I messed up on the components which produced a terrible gridded shadow map, this momentarily sent me into a panic but thankful I found and rectified the problem before the deadline day.

Overall I have learnt so much about UDK especially concerning its animation and light mapping the landscape tool, many Kismet sequences and triggers, the matinee, movie cinematic and animation effects, volumes, lightmass and many more tools and features the result is that I have learnt to me more aware of the workings of engines  and their practicality. The journey has been inspirational and it will benefit me in my future projects and employment. So From this experience I have taken away a much greater  knowledge of the difficulty to comprehend and use together both sides of game theory and aesthetics.

Part 5- UDK Final Renders

(ALL TEXTURES were taken with my own camera (my 3.5 gb texture library) except the sky dome and speed tree leaves which I modified.)


Monday 7 January 2013

Part 4



1. Two weeks before deadline I had to start my extra level under the temple just to give it more gameplay and character. In that short amount of time I learnt more about Kismet, cinematic, sound, music, particle effects, volumes and much more. I also allowed my level be tested two days before deadline day which found about a few things to change like the round puzzle door. Instead of interacting with the door you press the alter icon instead. The problem was overlapping the trigger volumes on top of each other and it was difficult to interact with.(Unlit Screenshots and Mesh Screen shots) 


2.

3.

4.


5.

6.

7.

8.

9.

10.

11.
 12.

13.

14. I had many problems with UDK but this was the most annoying problem I encountered. But I managed to fix it by turning off the landscapes indirect light function in its properties so that it will self shadow but not cast shadow. I had to improvise and add meshes to cast shadow were the landscape would have shadowed.

Tuesday 18 December 2012

Part 3

1.
Learning how to do landscape materials.



2. There were a few things wrong with the landscape tools like for example the visibility tool wouldn't cut holes in the landscape so instead I thought that I could have a cinematic instead to take you through after solving the puzzle.

3. Painting the landscape and experimenting what materiel goes well.

4. Now it was time for me to lay down all the stairs for the temple path but I had to create them in Maya first and export them over.

5. My level need somewhere interesting to start so I thought of a door you can go through to enter the game. (489 Tris)

6. Carefully measured the door to confirm that it fit the giant frame. I kept the theme of my temple puzzle. The round puzzle door has star signs and so this door has the sun and moon as a lock.(Maya render,670 tris)

7. At this point I need some type of view point to look at my environment so I quickly modelled and textured this into UDK. At the same time it textured the door and the frame.(Maya Render)

8.  Odd tweaks here and there. Textures on my pillar need sorting and I needed to construct many flights of stairs down from the entrance with side blocks.

a.
b.



c.
9. Placing of the pipes was a hard task to do. The 5 sided pipe wasn't perfectly symmetrical so I had to be clever and find a way to show my pipes lined up with each other.

10. When I was forming the rocks for the environment I slowly started to realise the potential of normal mapping as I knew it was difficult for me to figure out the NDO software. It also didn't help that I was doing all this through a dark laptop screen which had no screen back lighting.

a.

b.

11. Rock placement. I studied cliffs to get a feel where I should place the rocks and found that a near enough solid straight cliff face does look a lot better then a really rough cliff side.


12. This is the time when I started to work with the foliage tool using my own foliage meshes.

13. In half a day I quickly learnt speed tree and put the trees it into my level BUT regrettable because I now had these trees in the Lightmass kept on failing to build! So I had to remove them and manually create my trees in Maya.


a.

b.

14. Deadline was a month away so I had to get more assets into my level for the bottom half and I only managed to get half of these textured and exported into UDK. 


15. Examples of extra assets.



16. At this point some of the shadows on my object were getting messed up (shadow bleeding) So I did some tests and studied light map tutorials carefully. I found all my light mapping textures were wrong but I didn't have enough time to fix so I winged it and fixed the important meshes and that day I learnt allot more about light Mapping.