WEEKLY NARRATIVE
After attempting a few means of implementing Volumetric Lighting on my own I returned to the internet to try some open source packages. I found two that seemed up tot he task:
https://github.com/Unity-Technologies/VolumetricLighting - this is apparently a package of some of the volumetric light effects from unity's Unity 5 Demo "Adam". It's a little easier to implement but dependent on the type of Renderer and GPU you're using. https://github.com/SlightlyMad/VolumetricLights - and this was produced by a guy named SlightlyMad, and is a little more complicated and limited, but appeared (at first) to be more cross-platform friendly.
I was able to get both working in their own demo scenes & projects, but neither to work in the REMemberance unity project.
It took a lot of tweaking and experimenting with inputs and settings to get the lights in the first package to be visibly volumetric at all, but I got it working. Unfortunately that was only after a lot of hopeless debugging that ended up leading me to what seemed to be a bug in Unity's build settings, in which metal support is turned on when the associated checkbox is turned off.
Anyways, Here's the result:
It's definitely not what I was hoping for, but after spying the code behind this effect in both of the open source implementations I found, I'm thinking it's probably not the best idea to spend more time tweaking this. If it turns out that this doesn't hurt performance too much I think it will end up looking this way in game.
CONTENT WITH HOURS
- 6 hrs Volumetric Lighting
WORKFLOW EXAMPLES
See above?
POSITIVE OUTCOMES
- Finally able to visualize enemy's vision area.
NEGATIVE OUTCOMES
- I'm still not satisfied with the appearance of this volumetric spotlight but for time's sake, stuck with it.
- I wasn't able to dedicate as much time as I'd have liked to this.
NEXT TASKS
Rem is still in need of a visual effect for dashing. I'd like to plan more, but time isn't looking like that will be possible.
TOTAL HOURS LOGGED THIS WEEK: 6 hours
POST MORTEM
Remembrance has been an interesting project, and I find it hard to believe it's only been about 22 weeks of development. The variety of work i've done for this project is astounding. I managed to touch almost every aspect of game development during this project's lifespan. I've rigged, animated, textured and modeled. I've learned a ton about post processing, Occlusion Culling, Light Baking, Bread Baking, NavAgents, Mesh Instancing, Ray&Sphere Casting, Audio Mapping, Particle Systems, Trail Renderers, and so, so, so, so, so much about shaders.
I couldn't be more glad about how much I've been able to learn on this project - but it wasn't all technical skills. Managing a team of twenty students is pretty difficult, and even more so when all of them want to influence the design of the game, and the team's trajectory. Throughout the process we each were constantly describing different understandings about our plans. We couldn't all attend every meeting, so decisions were constantly being made and unmade with everyone being notified. This is somewhat inevitable, and is mitigated by a smaller team of decision makers, but for what felt like the longer part of these two terms we all were excited about the potential this game had. We even spent the first two-three weeks of this second term entirely in a concepting phase. (This ended up feeling like a mistake, especially when most of our ideas never made it into the game.)
We had very big plans, and in the second phase, when the size of our development team doubles, we felt encouraged to make those plans even bigger. We would continually say "We have six modelers, and eight programmers! We can get all this done." - but alas, not a one of us was already proficient in any of the skills we needed to do it. And even worse than that, a time of this size is nearly impossible to organize.
We could have survived the added organizational stress, though, but the design decisions we made ended up leading us to try to improve the game by adding too many abilities and new mechanics, and too much content. Though at the time and during our discussions expanding the player's toolset to deal with evasion situations in the game seemed like the smartest idea, in the end we ended up spending far too much time scraping by to develop the most basic versions of these abilities on schedule, and thereby ran out of time to polish any of them to a point that felt good.
Throughout the project's lifetime, our playtesting feedback constantly resulted in players complaining about the character controller, about the camera, about jerky movement, and about a lack of clarity about the game's mechanics. And we all had great ideas to fix these things - but among a denser cloud of concerns we were overcommitted to developing more content and abilities; those problems were never solved, despite us having plenty of good ideas about it.
That said, I think the primary reason for that was that the game's original mechanical concepts were weak, and very poorly defined. We ended up discarding them early, and spent the rest of our time trying to improve an idea from every angle at once that had become confused, and mostly lost.
This isn't something that can be avoided very easily at all, and even if we had realized this early (we had) we wouldn't have been quite free to turn back and work on a different game.
No matter the end result though, learning any skill at all - especially project management, documentation, team communication, or learning anything technical at all, is inevitably a process of trial and error, and experience. Working on REMemberance, I think all of us acquired a tremendous amount of experience that we might have had from only on the best internships.
PS - though I wish I could have spent more time in the last several weeks of this project on visual effects, there were several weeks during which i managed to do a whole lot. After this project lead me to begin with graphics (In order make the Xray portal effect), I ended up going to spend hundreds of hours of free time teaching myself about shaders, rendering, particle effects, and procedural animation. I ended up steering all of my projects in that direction, teaching myself droves of linear algebra lessons, and going through dozens of hours of animation and rigging tutorials. My personal projects look cool. I feel cool. I am so glad I started learning about graphics and effects because of that Xray portal, and It's certainly the career path i now intend.
I hereby rate REMemberance 8.6/10.
No comments:
Post a Comment