[UE4]Project Demeter - Medieval Fantasy
polycounter lvl 7
Hello all!
some new environment art I did over the weekend
Inspired by my old village environment from 2 years ago and the same inspirations from that, aka fable, WoW, etc.
Been working in my spare time for a few months now after work on my own little game project using UE4. I thought it would be fun to both learn how to use Unreal Engine's blueprints and scripting (different enough to not feel like work outside of modeling ship docks as my day job) and have a little fun making environment art as well. It would be awesome if I actually finished it as a full game, buuut that's a long goal and probably a year or two down the road, mostly just experimenting and having fun with it for now
Started replacing some of my placeholders for more evolved placeholders? Anyways the game projects always changing as I find out what's working, isn't working, etc. E.G. it first started off basically because I loved mount and blade but hated that you can easily die during sieges, causing you to have to wait on the sidelines for a long time while the NPCs had all of the fun. So first things first I started making team-based AI using UE4's blueprints, with the added ability that you can take control of any NPC at anytime.
Well as it turns out when you can take control of any NPC, fear of death becomes moot so I might have to rework that idea to be a lineage-only ability or something that's unlocked later. Hence I'm toying with the idea of adding in a dialogue/quest system too, sort of similar to Zelda or Fable, with the ai wars/simulation/rts thing running in the background and becoming slowly unlocked as you progress through the story. Who knows! For gods sake I'm still wondering if the world has magic or not, probably will cause who doesn't love magic beyond not making it OP? Quests will definitely be text based if implemented, but I like the idea of using a Sims/Spore language when someones talking, as well as background "life" for npcs just talking to each other. Setting up a quest system should be fun to make for me as well as I enjoy writing, and it would help give the "player" a sense of progress or achievement as I find having the AI doing too much during the game starts to give off the feeling of "oh, so this is cool to watch the AI do what I want them too as a programming project, buuuut if the AI does all of the war-making decisions what does that mean for a player besides being a spectator..."
Coding-wise it seems RTS-wise I've hit a road block where I can get the ai running up to 300-400 NPCs fully battling. Started off at < 100 NPCs causing lag with alot less AI, so that's progress coding wise for me after I rewrote it a few times as I learned some more! (Roadblock might pertain to the way the charactermovement component and skeletal meshes are coded in the engine according to the debugger which is currently above my knowledge level, so I'll probably have to revert to an off-screen simulation and/or respawn system like most other games vs entire sim running all the time, but i think with that I should be able to do something fun. IE Armies have "active" size caps depending on what villages they control, they can however have reinforcements while defending by spawning out of barracks, or if attacking coming out of a siege camp tent set up prior to invasion. Might be fun to route off the reinforcements or charge out to try and destroy the enemy camps vs just waiting for all of the enemy's reinforcement counters to dwindle down to 0.
Environment-wise I started off with simple abstract low-poly stuff just to get some assets in there for the game to progress, but over the weekend decided f-dat, I want to have fun with textures! So here's a couple of older prototype videos and some environment stuff from the weekend. Most likely I will be using world machine to generate the basic landmass, which is what this level test started out as as a test for it. Then I'll set up the main faction castles (3-5 factions max, I like odd numbers as then if the AI gangs up on each other there's always an odd man out), some villages inbetween, and will probably start making a cave snapping set too to prototype quests similar to how skyrim had a shitload of caves. Trees were made with speedtree if you couldn't tell, and I do want to have ta day/night cycle implemented eventually though that just distracts me too much from programming as I always get caught up in focusing on the lighting too much for simple tests! As in, it looks great during the day and night, but god damn it I hate how it looks at 12:00 or inbetween when the sun is down and moon is coming up (straight down shadows at 12, and no shadows inbetween, ahhh it kills me even if that happens in reallife!!!!)
Still am debating on what to do character wise, though these proto-knights are going to have to go with the new style. They especially will need arms and legs at the very least for animating things like sitting or eating
Mind the long and terrible videos, they were made to show to people at the office and close friends, not actual cinematography
first video ever when I got basic team combat working
https://www.youtube.com/watch?v=Fu-q64of4gI
second where I got way too excited about implementing apex flags.
https://www.youtube.com/watch?v=NlBPbHShCx4
Not shown in above but now the AI also will search for areas with flags and try to capture them if they don't currently own them (dynamic arrayed so all I have to do is place one entity to tell the ai it's a new village/target, yay!), and added in some castle horns that blaaasts when an enemy is sighted or the flag has changed teams, inspired by lord of the rings movies of course.
some new environment art I did over the weekend
Inspired by my old village environment from 2 years ago and the same inspirations from that, aka fable, WoW, etc.
Been working in my spare time for a few months now after work on my own little game project using UE4. I thought it would be fun to both learn how to use Unreal Engine's blueprints and scripting (different enough to not feel like work outside of modeling ship docks as my day job) and have a little fun making environment art as well. It would be awesome if I actually finished it as a full game, buuut that's a long goal and probably a year or two down the road, mostly just experimenting and having fun with it for now
Started replacing some of my placeholders for more evolved placeholders? Anyways the game projects always changing as I find out what's working, isn't working, etc. E.G. it first started off basically because I loved mount and blade but hated that you can easily die during sieges, causing you to have to wait on the sidelines for a long time while the NPCs had all of the fun. So first things first I started making team-based AI using UE4's blueprints, with the added ability that you can take control of any NPC at anytime.
Well as it turns out when you can take control of any NPC, fear of death becomes moot so I might have to rework that idea to be a lineage-only ability or something that's unlocked later. Hence I'm toying with the idea of adding in a dialogue/quest system too, sort of similar to Zelda or Fable, with the ai wars/simulation/rts thing running in the background and becoming slowly unlocked as you progress through the story. Who knows! For gods sake I'm still wondering if the world has magic or not, probably will cause who doesn't love magic beyond not making it OP?
Quests will definitely be text based if implemented, but I like the idea of using a Sims/Spore language when someones talking, as well as background "life" for npcs just talking to each other. Setting up a quest system should be fun to make for me as well as I enjoy writing, and it would help give the "player" a sense of progress or achievement as I find having the AI doing too much during the game starts to give off the feeling of "oh, so this is cool to watch the AI do what I want them too as a programming project, buuuut if the AI does all of the war-making decisions what does that mean for a player besides being a spectator..."Coding-wise it seems RTS-wise I've hit a road block where I can get the ai running up to 300-400 NPCs fully battling. Started off at < 100 NPCs causing lag with alot less AI, so that's progress coding wise for me after I rewrote it a few times as I learned some more! (Roadblock might pertain to the way the charactermovement component and skeletal meshes are coded in the engine according to the debugger which is currently above my knowledge level, so I'll probably have to revert to an off-screen simulation and/or respawn system like most other games vs entire sim running all the time, but i think with that I should be able to do something fun. IE Armies have "active" size caps depending on what villages they control, they can however have reinforcements while defending by spawning out of barracks, or if attacking coming out of a siege camp tent set up prior to invasion. Might be fun to route off the reinforcements or charge out to try and destroy the enemy camps vs just waiting for all of the enemy's reinforcement counters to dwindle down to 0.
Environment-wise I started off with simple abstract low-poly stuff just to get some assets in there for the game to progress, but over the weekend decided f-dat, I want to have fun with textures! So here's a couple of older prototype videos and some environment stuff from the weekend. Most likely I will be using world machine to generate the basic landmass, which is what this level test started out as as a test for it. Then I'll set up the main faction castles (3-5 factions max, I like odd numbers as then if the AI gangs up on each other there's always an odd man out), some villages inbetween, and will probably start making a cave snapping set too to prototype quests similar to how skyrim had a shitload of caves. Trees were made with speedtree if you couldn't tell, and I do want to have ta day/night cycle implemented eventually though that just distracts me too much from programming as I always get caught up in focusing on the lighting too much for simple tests! As in, it looks great during the day and night, but god damn it I hate how it looks at 12:00 or inbetween when the sun is down and moon is coming up (straight down shadows at 12, and no shadows inbetween, ahhh it kills me even if that happens in reallife!!!!)
Still am debating on what to do character wise, though these proto-knights are going to have to go with the new style. They especially will need arms and legs at the very least for animating things like sitting or eating
Mind the long and terrible videos, they were made to show to people at the office and close friends, not actual cinematography
first video ever when I got basic team combat working
https://www.youtube.com/watch?v=Fu-q64of4gIsecond where I got way too excited about implementing apex flags.
https://www.youtube.com/watch?v=NlBPbHShCx4Not shown in above but now the AI also will search for areas with flags and try to capture them if they don't currently own them (dynamic arrayed so all I have to do is place one entity to tell the ai it's a new village/target, yay!), and added in some castle horns that blaaasts when an enemy is sighted or the flag has changed teams, inspired by lord of the rings movies of course.
Replies
Last Image is a bonus of the first quest I wrote to test the games systems, currently can use an NPC to get an associated quest dialogue with options, which will set the "QuestStage" integer, based off a plugin I purchased from the marketplace to jumpstart that areas development, and expanded upon it to work with my own functions on quest progress, tracking, and events. (kind of a mixture between fallout/older RPGs that had text dialogue with no voice but more options and also looking towards Zelda as inspiration.) Basically set-up wise, each quest has stages set by the dialogue or ingame events, which updates both the dialogue and the associated quest blueprint to change things like if already spoken to and accepted, use new dialogue, or spawn in enemies/unlock doors/cause enemies to attack, etc.) Was kind of fun going back over my ai and revising it so now the basic character class will attack regardless if on team red or team blue if HatesPlayer is set to true
Currently I think I'm leaning towards quests can indeed be "failed," although killing the quest giver technically completes it as well
^^First test upgrading the prototype model.
^^Conflicting directions, not blending well
Decided maybe push towards a more traditional sculpted/textured look. Led to the thought that the more textured and detailed I make one asset, the more time it's going to take to do a female version, different faces, hair styles, clothing and armor, making sure the environments hold up, animations etc.. Basically just multiplying the work involved to complete the game by a large factor as one person. Also the notion that there comes with it the uncanny valley, having a style like this requires more work to be done in variety in order to prevent it being too obvious that the base meshes and looks of the characters come from a limited pool of assets.
^^End of day was resolved going back to the drawing board after pondering for a bit, leading me to settle on subtracting him to a more abstracted look. Is the most current vision of the game, though I will be going over to refine both character and environment (will have to take out a few edge loops of the smoothed house that was textured to fit with the harder buildings as a small example. Whether terrain will have textures or not is to be determined for things like roads and grass, but will have to be minimal if any. I noticed godus used some simple texture elements here and there, such as a couple of roof tiles or flowers blended in which might work quite well.
^^ said example of the first "prototype quest." Yay for taking a generic slay the bloody bandits in a dungoen quest and trying to write it so it's (hopefully) somewhat interesting.
Last bonus: messing around with making possible music for the game when I'm too burnt out modeling after work:
Was working on coming up with humans seen above^^ as getting to the point of making variations while working on animations and basically said fuck that, it'll take years to make a bunch of them fully animated for male and female with different clothing and armor.
Humans are hereby scratched, now everything is medieval robots. More fun to make and animate, quicker turnaround, and allows me to create new types of monsters, environments, and effects that would otherwise feel out of place in a human game. Also, yay free synth voice acting!
Robot shakespeare, different plays that occur in the game depending on the time? Why the freak not.