I’ve been working overtime the last week to get as mush as possible finished before the delivery. Texturing, lighting and making corrective tweaks to the models to better get them to mesh has been very time consuming. I underestimated the amount of time it took just to place the meshes and build the scene inside the game engine.
When I first started this project, my goal was three buildings, a crashsite, canal with bridge and some clutter in the courtyard. It sounded quite reasonable back then. It has resulted in over 80 unique meshes and that number would’ve been larger had this been a real company and game project. Time have meant that some of those meshes will not be in the scene as it is simply impossible for me alone to create textures and shading maps for so many objects. The three buildings, crash site and canal are still there, they’ll just not be as interesting to look at as I first imagined.
Looking back, this project would’ve been more optimal for two team members. Being just one person, I would have been better off focusing on a single building and trying to make that look good. Instead I will unfortunately end up with several mediocre-looking buildings. I really have myself to blame as I did not build my meshes in Maya the way that would’ve been best for the pipeline and was forced to go back and redo most of them. This ended up costing me alot of time.
This past week I’ve completed the rehaul of all the models, they’ve been tweaked, imported and placed into a level in UDK. Texturing is well underway aswell. Since I remade just about all meshes I had to create new UV-sets for them. All this ended up very time consuming and I’ve fallen a bit behind on my schedule. I will make the delivery for milestone three, just not with as much detail in the textures as I would’ve wanted.
After milestone 2 I went back and revised all my models, making them more fit for UDK. I added details to the mesh instead of having to over-rely on textures. Many models like for example pillars were halved to provide for greater detail in the textures.
This change, while need puts me a bit behind schedule, though it’s still within acceptable range.
As of this milestone texture is underway, as well as the various maps.
Looking back at things for this milestone a few things are clear:
-The timeplan have held up reasonably well, with no real adjustments necessary.
-I should have “stressed tested” the graphics engine to more closely determine the optimal number of polygons.
-I might have been to zealous in keeping the polycount down.
All Maya meshes have been completed, as well as all the UV-mapping. Texturing is underway, as is the Mudbox sculpting for the pieces that needs that.
Modelling in Maya is just about complete now. As of writing this I calculate that I will be finished with this stage on Sunday. All that is left are parts of one building and some tweaking to the crash site in the courtyard. UV-mapping is also well underway and is currently 1/3rd complete. Next week I will complete the mapping and get halfway through the basic texturing. I will also start exporting some of the meshes into Mudbox to sculpt some of their parts.
This week I’ve just about concluded my research. Creating collision for my meshes was one of the last obstacles as the tutorials for it on UDK’s homepage were rather outdated. Fortunately it’s been taken care of with the help of others in the class.
I began modelling the main building this week and it’s proceeding along nicely. I calculate that the Maya part of it (more than 90% of the model) will be completed no later than monday next week. At this rate I will have more time to create more and finer details for the models. I ended up doing some UV-mapping as well, mainly to make it easier for myself later as many parts of the buildings will be duplicates, such as the brick rows between windows.
The model functions properly in my quick tests in UDK, so so far it looks like I won’t have to change anything because of technical difficulties.
All in all the week have gone nicely, I’m on schedule and haven’t had any unforeseen or major problems. Looking forward to completing this row of modeling.
I’ve created a very simple mock-up to test the layout and composition of the buildings I want. The three building types will be arranged as to form a courtyard of sorts, with a canal running right through it.
One change that I’ve decided on since this was made is that the plane, rather than just standing there on the courtyard, will have crashed instead. This will make it look more like a scene where things are happening, rather than just a lifeless showing of 3d models.
This is my reference board for the models I’m going to do. Inspiration from the early 19th century classicism of the buildings in Gendarmenmarkt Square in Berlin. The plane is of the kind that the Red Baron flew in the First World War. All this will allow me to work with detailed textures and normal/specular maps alot, which is one of the things I had in mind with this project.
So, first post. This week I’ve completed my project plan and done research into the game engine I will be using, namely the Unreal Engine. I’ve finished sketches for the game assets I will be creating and I have tested importing pipeline into UDK with some rough mock-ups.