Hello Hivers,
I’ve been thinking for quite a while about in what small way I could contribute here on Hive regularly.
Then all of a sudden @atomcollector drew my attention to the new game Idle Raiders by praising it on Discord.. I’ve been playing it for almost two weeks now and I’m addicted to leveling up my little chibi fighters and throwing them into battles of all kinds (World Bosses, Territory Wars and Dungeons -> the main vehicle to accumulate RAID, the token to upgrade your heroes).
@Mainstream, the creator of Idle Raiders, is a solo developer - and even though I haven’t asked him, I’m sure he uses a number of AI tools to be as productive as he is. He pushes small updates almost daily, and the whole game looks very professional with a clean design.
That brought me back to an old idea of mine that I kept postponing because I lacked some of the key skills involved: drawing and programming :o).
Yesterday I read the latest blog post by @acidyo, who asked all of us - everyone who posts here regularly or irregularly - what value we bring to the platform.
Each of us has to answer that for ourselves. In my case, I tried many things and dropped them again because they didn’t feel “right” or because I didn’t manage to stay consistent with one topic. Despite long breaks, though, I’ve always remained loyal to Hive as a reader and player/investor of various games, and I see it as my online home. One I hope will grow more successful and attract a larger following over the years.
So now I’m trying a small project that I’d like to introduce here to do my part.
A New Project – Fantasy RPG
Let’s be honest. Many of us harbor ideas for great novels, games, TV series, or films, yet we never dared to dream we’d get anywhere near a finished product, or even begin to develop and execute one of those ideas.
As far as I'm concerned, I want to finally make use of today’s age of wonders (despite plenty of skepticism and a bit of fear about the medium- and long-term future) and, with the help of AI tools, work step by step toward a small prototype of an old idea of mine:
A fantasy role-playing game.
As far as I can remember I’ve played JRPGs, adventure RPGs, all kinds of fantasy games, and I consume a lot from this genre in general.
Many of those scenes, archetypes, dialogues, myths, etc. stick with you and blend into new ideas.
The time has come where any one of us can open a small-scale production studio for comparatively little money and try our hand at creativity in ways our predecessors or older contemporaries couldn’t have imagined.
It’s entirely possible I’m being too optimistic, and that even with today’s sophisticated AI techniques it still isn’t feasible for someone like me - who has little programming experience and, beyond HTML and CSS, doesn’t really have the understanding to build a game backend.
But I’m going to try, and I’ll share my attempts, as well as my failures, here with you.
At the same time, I’ll present the respective AI tools I use and what they can do not only to achieve my goal but the overal purporse and usage of those tools. That’s the part I hope will help some of you (keyword “added value”).
Of course I’d also love to read comments over the coming months from people who might have ideas I haven’t thought of yet.
Harnessing the Power of AI
So how do I start?
I figured the easiest way is to use the best-known AI and - from there, with the of said AI - explore other tools.
ChatGPT is now in its fifth version, and even though I’ve read somewhere, there are competitors with better reasoning or programming abilities, I don’t think I’m making a mistake by choosing OpenAI’s AI, even if only for a month. The €20 for a Pro account was worth the experiment to me either way.
A great new feature of ChatGPT is Projects, which let you connect disparate chats into a single project, either from the getgo or later on, so context memory is improved.
From initial impressions, ChatGPT still forgets context when a chat gets very long or, as it happened in my case, when you split chats and then merge them into a project, but compared to ChatGPT-4 it’s definitely a step forward.
It could also be that my prompts are often too vague and not precise enough, or that I don’t immediately push back when the AI generates an off answer. I have to keep reminding myself I’m not talking to a human who truly “understands” what I’m saying.
Plus, people forget things as well :o).
Overall, though, I’m pretty satisfied and I’m understanding better day by day how to feed the AI to get better results.
My best approach so far has been to let it take the lead and, when entering a new topic, ask it to treat me like a total beginner (which I often am) and give me a broad overview with tips on how to go further and deeper. Even with helping to better prompt.
If you go to the “Personalization”-section unter Settings and explain exactly what you expect and how much, or how little, “fluff” you want in the answers, you can improve results again.
For example, I asked it not to spit out constant niceties and to get to the point instead. That has helped enormously across chats and projects.
The Core Idea
There’s one area where I didn’t want help, and probably won’t take until the end of this unusual collaborative process:
-> the creative act of inventing my story.
I have a story kernel that certainly won’t become a new Lord of the Rings or Game of Thrones, but it could be “catchy” enough to carry a small startup project.
I don’t want to go too deep yet, but the protagonist will travel through the world in first-person perspective.
So there won’t be a player avatar that you can control as in World of Warcraft.
I imagine the combat system will be similar to the old JRPGs - Suikoden, Shining Force, Final Fantasy, and the like.
Whether alone or in a party (there will be companions you can collect), you fight turn-based battles against enemies.
There will be experience points, various stats (which I’m still working out), as well as loot drops and real progression that can be accelerated by story-based and side quests.
The story itself isn’t directly embedded in Norse mythology, but its atmosphere strongly echoes Nordic-Germanic vibes. I’m German myself and I’m more familiar with German and Germanic (In German, "deutsch" and "germanisch" are distinct from another) mythology - as well as German names and word roots - than with their English or Latin counterparts. So a story in that mythic sphere suggests itself.
The hero starts in his home village and, in a small semi-tutorial, will complete some early quests in the safety and idyll of the starting area before traveling into the surrounding region to face his first real adventures.
I’m currently fleshing out the first region. Former or current World of Warcraft players will likely get the picture if I say its size will be comparable to Elwynn Forest.
What I loved most about WoW were the many side quests that not only taught new recruits how to level and defeat foes, but also introduced the world and its inhabitants. You immediately felt at home and strangely connected to the world you - whether as rogue, warrior, priest, or paladin - wanted to save.
Even if I can’t reach WoW’s density of story, I still want to create the illusion of a populated world, and I want to put a real effort into the individual NPCs and quest lines so that players don’t just chase XP, but actually immerse themselves in the world of Branningen (that’s the working name for the first region).
Future Blog Posts
These are my initial thoughts on the project, and even though I’m already a bit further along, I’ll leave it here for now.
I don’t yet know how often I’ll post. I’d need to see the future to know how fast I’ll progress, but a weekly update with all the dead ends, fails, and hopefully a few small wins should be doable.
I’m genuinely curious whether it’s possible for a total noob to get their own game off the ground.
- Pictures created with ChatGPT5 Pro