Week 8 Tuesday 2018-05-22
We are all feeling pretty good about the status of the game! Some of us were busy with other classes this week though, so we didn't make as much progress as earlier weeks. I imagine our schedules are only going to get busier.
My goals for this week were finishing up the skybox, tweaking the way shaders accept texture parameters, then finishing up metallic/roughness implementation, and getting Assimp to accept fbx model files.
The skybox now works perfectly (after some hackiness to get the sides to appear right-side-up and properly connected), shaders can accept an arbitrary number of textures (though they'll throw errors if you don't pass enough), all shader classes now have access to the skybox's cubemap (but don't do anything with it yet), and Assimp now accepts fbx model files.
I've put off metallic/roughness until shadow-mapping and dynamic light sources are implemented by Nick, because otherwise I'll be duplicating his work. I've just barely started the class framework for terrain looping, so that's completely unimplemented. I'm not too worried about this second thing not being done yet because it was one of several goals added during the week.
Next week I'm going to implement terrain looping and metallic/roughness implementation. I may also start on animation depending on how these other parts come along.
I learned that some of the packages on NuGet (such as Assimp) are five years old and don't match current documentation. I had to completely switch out the NuGet package for stuff downloaded from their GitHub to get fbx files to load.
We're making decent progress and I'm largely satisfied, though I'm mildly worried/frustrated because I need Nick to finish shadow mapping, but I haven't heard from him how much progress he's made (so I don't know how long I'm going to be waiting).
This week my goal was add the item-obtaining feature. When enemy vehicles are destroyed, they drop items which can be picked up by the player vehicle.
I successfully finished this feature. Enemy vehicles drop items upon destruction. The number of items can be changed in the future depending on size of enemy vehicle. Items can be picked up by the player vehicle upon contact and the item will appear on board on the tile that collided with the item.
I finished all my goals this week.
I think I will be picking up another gameplay feature or possibly moving back to sound as our sound producer produces more sounds.
I was able to apply what I learned last week to this week's feature. It was fun to build off what I learned and doing so helped solidify my understanding.
I'm feeling good about our progress.
Shadows and lighting
Nothing, 163 project attacked
163 happened, and so did a bout of fatigue and low motivation.
I plan on actually being a productive member of the group, unlike how I have been. There are only a couple weeks left, so I want to give it a strong push
I'm really bad at time management
Senioritis is hitting hard and I'm getting overwhelmed, but I need to step it up and be more active.
Enemy ships react to the shape of the player vehicle, and also to other enemy ships.
I didn't accomplish much this week, I was kinda busy with other stuff, and I'm feeling tired and stressed due to school and other stuff piling up at the end of the quarter, including several events I have to prepare for.
I feel like I keep re-learning this lesson, but... If it seems like there's a mountain of work in front of me, if I try to keep it all in my head at once, it seems insurmountable, and so I end up not doing much work at all. But if take it task by task, and fully focus my attention only on the current task, things feel doable, and I can make progress.
Also, I should worry less about making things “perfect”, since theoretically there's an infinite amout of work possible to do on projects like this, so I should just work efficiently in the time allotted.
Medium. I think our game is doing pretty good, I'm just kinda tired / stressed.
Start texturing everything.
We got FBXs in. I have all but the face done of the player model which took quite a bit longer than expected (more detail in part 3). No texturing :(
I spent a lot of time on the player model (likely a poor distribution of time, but I'll make up for it toward the end of the next week) and learned more about making a low resolution version of a model. I also have/had a paper to write and had to finish a rather difficult computer vision assignment which were bound to take up time (and some of next week as well).
Finish the player model, texture most things (blocks, then the rest), that'll probably take most of the end of the week when I have time to work on it.
That I know nothing about proper mesh topology and that it's a whole field of study in and of itself for animation, and that I should go for good enough and move on.
The game looks awesome and is playable! I'll feel better when I start making progress on the visuals, right now I'm gonna let my lack of satisfaction push me to make more time for working on models at the end of the week/weekend.
Fix the transform bug, then fix the multiple delete messages. If I have time, optimize the components in a similar way to the game objects, so I don't need to send over render data every frame.
I changed some Transform class functions so I copy over the position/rotation/scale directly between objects instead of the local matrix, then decomposing it. I might change the update message to hold the position/rotation/scale instead of the local matrix as well.
Not sure if the multiple create/delete bug is still there. I think it has to do with creating and deleting objects in the same frame? More testing is needed.
I think the networking portion is solid enough. There are obvious optimizations I can make, but at this point, gameplay is probably more important.
Nothing really.
I think our progress is pretty good. I hope we will be able to get more graphics stuff implemented soon.