Progress is currently going slow, but steady. The following problems are keeping me occupied:
- The rook environment is getting closer to where I want it to be, but the overall environment is still slightly chaotic, because it can only be created through foliage editing and procedural creation of assets. This chaos actually adds a lot to the overall atmosphere. It looks like a graveyard, but it's not. It looks like it has a history ... but it's more like a dream, like a mirage. Things don't feel out of place, but they feel ... unsettling. It feels like you're moving through hostile territory, but the visuals don't look hostile. However problems arise from overlapping assets and the unequal distribution of items and hideouts. The less chaotic the environment is, the less it carries the atmosphere of .... surrealism and the less it feels threatening. Keeping that balance of chaos and purposeful asset placement is difficult.
- FPS issues and loading issues occur, but they are not consistent. They are related to the creation of the fences. The fences appear and disappear depending on whether the player looks into their direction. This adds to the overall feeling of being "manipulated", but it also eats up resources. These fences need further optimization, but I don't know yet where else I can optimize. In theory I could replace the bars with a simple shape plane and create the illusion of fences with a masked fence material (like it was done back in the 2000s N64 era). But from what I can see the FPS drop is happening because of background calculations related to the player position and camera focus (the code that determines whether a fence appears or disappears). This is strange, because this calculation is only done once every 6 seconds and the interval is also slightly randomized so the calculation isn't done across the board as that would tank FPS .... so basically I'm seeing FPS drops and loading issues, that by all means shouldn't happen. Dynamic shadows are obviously also optimized across the board and only assets that severely affect lighting create dynamic shadows. So the GPU is definitely not the issue here.
- To keep the maze interesting, items and light beings need to be placed. Players can find locked chests, hideouts and light beings, but it's not enough. In the other segments players could encounter crate groups for additional items. Here I believe that items should be found inside fountains and wells. So for this purpose a new BP needs to be created and distributed among the other foliage actors.
- By far the biggest issue I'm facing is lore related. I would like the main rook tower to have a similar art style as the graveyard section. But this again is very resource intensive (overdraw). The rook tower contains portals that transport the player up or down depending on his movement paths. The power of the rooks lies in distorting reality and creating barriers for the player. These barriers and monuments should all be of the same type of architecture to demonstrate that they are closely tied to the rooks. The rooks display sadistic behaviour patterns. They are like spiders that lure their prey towards them, but also get their prey to weave their own cocoons. Their modus operandi is reality distortion, projection of power from far away and careful guidance of the victim into a prepared trap. They are playful and childlike in their approach. Obviously this is a lot to communicate through asset placement and environment design alone. Ideally the player should feel that something is ... "off". The player should wonder about the structures, their chaotic placement ... the player should feel uneasy. The rooks created this section in order to draw a being's interest while at the same time slowing it down and tiring it out. The maze tower is built in a similar fashion. It draws you in, tires you out and when you reach the top they deal their final blow. In the beginning I wanted the statues to point towards the rook tower. But what if they point towards the exit of the graveyard, in a final attempt to help the player escape?
- I lowered the height of the fences across the board, which greatly improves the overall environment, but it also leaves players vulnerable to softlock themselves. If a curious player jumps into an enclosed section, there is a chance that the player is too close to all 4 fence sides (which prevents the appearance code from executing). In those instances players would not be able to leave and remain there for eternity. I have not found a solution for this. Although I wonder whether I do need this tolerance distance and whether I could potentially remove it now that the fences are optimized.
- The sarcasm of the rooks becomes apparent when you consider, that all beings which eventually reached the top of the rook tower and then were turned into rooks themselves are now statues in this endless graveyard. The graveyard consists of fake monuments of the dead, while at the same time the statues themselves are the actual gravestones of light beings that fell prey to them. So in a sense: Yes it is a graveyard. But the gravestones are not what you think they are. This switch between what you think you see and what is actually taking place perfectly describes the manipulative nature of the rooks. They enjoy outsmarting their prey and fooling them into believing a narrative, that is constructed by the players themselves. And it gets worse: If the statues all point towards the "exit" of the graveyard ... are they even aware, that the exit leads back to the tower? I'd imagine their final gesture was an attempt at helping future victims by pointing towards (what they believe is) the only way out. This again hints at the manipulative nature of the rooks, since even beings who went through the entire ordeal still fall prey to the illusions that the rooks created and inadvertently help them catch future victims.
- Finally there is the main thing preventing me from moving faster: My internet connection. I have 0.5.3 ready for release and the goal is to direct everyone on every social media channel towards this version. Once they play it and get sucked in I've won a community member and supporter. I'd also love to post more videos and short edits. I'd love to create a trailer and upload it in the highest quality. But all of this has to wait until I get the fiber connection set up. Because each upload takes days, which slows down the entire pipeline for bug fixes tremendously.
To sum up: Not much I can do besides experimenting, adding new blueprints and assets and playtesting to check for softlocks, bugs and whether the environment is stable.
Keine Kommentare:
Kommentar veröffentlichen