Hello everybody. First post here ?

My question is pretty much in the title. I have a strong idea for an arena battler that's close to my heart. I've been writing and polishing the design document and the maths behind it in my free time for years now and am very convinced that it would make a both accessible and mechanically interesting game. Now I want to take the next leap and get into programming, which is also the biggest hurdle because I have next to no experience (ca. 100 hours in C#), though I'm not completely new to PC game development (I designed levels for 1st person shooters and maps for RTS games, but up to now I always had quality premade tools to work with).

I don't want to develop it as a mod in order not to give up any control over it, so I thought about coding it in Unity. I'll have some help with art assets, music and sound design from friends, but in general (especially design-, engine- and programming-wise) it's a one-person-project.

So you see I'm pretty serious about it and I expect this to take many more years which is OK because, well, it's a hobby. But I don't want to be unreasonable in my expectations. I know I will have to develop SERIOUS programming skills almost from scratch and that the workload will be high, so I'd like to ask some of you with real experience what I'm getting myself into. Is a project like that possible alone, and in what ballpark would I be in terms of workload? Like, in what time scales (in terms of total hours) do I have to think here, by and large?

In the end I'm much more interested in the mechanical side of it, so if it ended up pretty bare-bones in terms of production values, that would not be that much of a loss for me, even if a nice presentation is preferable. As of now I'm aiming for a Warcraft 3-kind-of level of detail. Server-based online multiplayer, one map with about a dozen interconnected arenas. Turn-based tactical multiplayer combat between up to 10 players and a roster of about individual 30 fighters to choose from that can be further customized with equipment. A certain degree of destructible terrain and some physical effects like collision dynamics would be needed, too, for the concept to work.

Aarix said:
I expect this to take many more years which is OK because, well, it's a hobby. But I don't want to be unreasonable in my expectations. I know I will have to develop SERIOUS programming skills almost from scratch and that the workload will be high, so I'd like to ask some of you with real experience what I'm getting myself into. Is a project like that possible alone, and in what ballpark would I be in terms of workload? Like, in what time scales (in terms of total hours) do I have to think here, by and large?

A team of 30 might take 7 years to make such a game from scratch. 210 man-years is a bit much for one person to take on.

Let's assume I'm overestimating and assuming triple-A quality. Maybe it can be done in 3 years. 90 man-years is still a lot.

Can you make something that looks like it came out of a game jam, mostly works okay, and might be fun? Sure. The design document for that will be about one paragraph long, and the art will be reused objects from asset stores.

As Tom pointed out, even moderately small professionally-built arena battle games are generally many work-years for the environment, work-months to work-years for each interactive object, work-decades for programming, work-months to work-years for audio, work-years for characters, work-decades for animations, work-years for effects, and work-decades (at least) for QA. That's not something anyone can do alone.

If you dropped your design requirements tremendously, had other sources for most assets, and leveraged all kinds of external tech I can imagine an extremely minimal project getting polished into a commercial product within a few years of full time effort by a solo developer if you already knew basically what you were doing. That's absolutely not something you could do as a hobby side project, but instead taking 40+ each week for years. I've seen some solo projects do exactly that. It's not an easy thing.

This may surprise you, but that's exactly the kind of feedback I was looking for ☺️ To get a grip on what I'd be getting myself into.

You have to understand that for me it's mainly about seeing the kind of mechanics I developed in action and having people able to play them together. That's the core and the only really important thing about it and everything else can be theoretically abstracted away as necessary, so intuitively there should be some way to get there. Hence presentation being absolutely the least important aspect of it all. I come from a tabletop skirmish-game background and would even implement the mechanics as an excel spreadsheet and have people play it out with miniatures if that was what it took. That would be the most barebones approach imaginable. So what I'll have to do next is find a project outline between that extreme and the full-blown version described in my first post, one that would keep a single person busy for years but would still be realistically achievable.

Start small, cheap and quick.

Knock up a very basic single player game against a dreadful AI with placeholder graphics to bring the “spreadsheet” system to life.

Then iterate from there. If you generate some interest you may get some help from people who can code but not game design.

You might be okay with the game jam model. Be prepared to buy a bunch of pre-made assets, much like buying a bunch of pre-made miniatures and mismatched cards. The larger your rules the more work you have, but if the goal is “I made something!” it can take many work-years off the time line.

Most of the mechanics will (by necessity) be what other people made, not what you made, but it is often easier (still not easy) to bind together other people's work. For your larger design you may still be looking for work years, but at least it would be an approachable number for one person.

Lots of good advice here :D

I was thinking about three milestone goals for prototyping.

First milestone would be "I've implemented all the existing spreadsheets into a software that does all the calculations" (pretty much like a virtual tabletop, just with more fancy maths behind the rolls). That would be enough to run tests if the mechanics work in practice or only on paper (lots of skill chains between fighters and incremental effects there; hence too complicated for tabletop gaming unless you have masochistic inclinations).

If that works out, the second one would be "I attached that software to a bare-bones 3D-engine that is able to track the spatial relationships of multiple characters and is able to feed that info into above calculations", so that the program can calculate things like area of effect or distance-based damage. That would be great to see.

The third one would be "Multiple people can access the 3D-engine at the same time over the internet" (so that it's basically a super-ugly but functional virtual tabletop).

And I don't even want to look further into the future because this alone will no doubt keep me busy for long enough :)

Sounds achievable enough? If so, what's the scale in terms of man hours I could expect these milestones to take?