Train delayed (or cancelled)

After my last update, I was on shaky ground with what I was doing with this game:

• The scope is big, and I didn't really feel like I'd thought it through properly
• It's a huge project that is contributing to a stack of very similar games that already exist (with more being released all the time)
• I wasn't convinced that I could figure out a cohesive art style for it that a) looked any good and b) I could ever get finished myself

The art style was a huge one for me: in my head, I was competing with the likes of Stardew, but this is a trap I've fallen into so many times:

I just cannot get games done when they demand high fidelity/detailed artwork and animation. Everything I've ever managed to get done, happened because it allowed me to use simplified/abstract visuals, where the details and animations didn't have to be perfect. 

For some reason I strayed (for the nth time) from this, so I put train game aside and started prototyping. Here's a recap of this year and the random things I've stumbled through making:

Racing game

The first game I prototyped was a topdown wipeout-like futuristic racing game. One motivator for this was the idea of a fast-paced high-BPM soundtrack.

It didn't have a name other than 'racer' but it was kind of neat, and had a decent feeling of speed to it. I also managed to get back to dependable abstract visuals (TM)

One rabbithole I did fall down on this - that worked out, and I'm really happy with - is that I added networked multiplayer to it. I riffed off the Quake 3-style network model, had decent working client-side prediction and everything. 

This was actually a kind of highlight of this year for me, because although I knew I could implement this kind of tech, I'd never sat down and actually done it on one of my own projects (and entirely from scratch).

Client side prediction frames

Eventually I kind of went off the idea just because I didn't feel as though I knew what good racing controls should feel like. It's not really my genre, but it was neat to play around with.

Top-down Diablo 2-like RPG

Then for some reason I got the idea to try and make a very simple hack-n-slash Diablo 2 like - but reusing my network code and focusing on multiplayer co-op. Not because it was a deep or interesting concept, but just because it might be fun, I don't know exactly. 

This actually got pretty advanced - it had enemies that could pathfind and chase you (and working combat), useable items, levels that were constructed from prefab 'blocks' - and of course, networked multiplayer.

This image is two different player screens concatenated together

It started to fizzle out for kind of the same reasons as the train game - although I was using abstract visuals, it was ugly (being actual topdown not isometric) and an art style that would make it appealing would be out of reach for me to produce on my own.

Pathfinding debugging with loads of enemies

I also didn't really know what I was doing with the lore and combat etc. - I tried to get too complicated just to avoid writing story and imagined e.g. procedurally generated storyline and therefore procedurally generated quests etc. But I never actually made that happen. 

'Spy game'

This is actually my current prototype that I'm still messing with - although I'm not sure for how much longer.

This is actually the first and only idea that I would say is properly original - it's not just riffing off of things that are well established. And it manages to retain abstract visuals (although it's still very ugly at this point)

The concept is that the player doesn't control people - but rather the systems that allow the flow of information between them. Intended to be 60s/70s era too - so telephones, teleprinters, radio, reel-to-reel recorders etc. that kind of thing.

Running an intelligence agency, their objective is to use their agents to steal information from an opponent, whilst guarding against any individual agent/entity having 'too much' knowledge of their setup that could fall into the wrong hands (e.g. due to double agents)

It's hard to convey in a screenshot but information in this game is an actual visible thing - the little blue boxes. It flows down the lines of communication between equipment, and into the agents themselves. Every action in the game relies on this - if your information can't reach something, it cannot be controlled. Equally, anything you do can be intercepted, modified etc.

I'm quite taken with this concept but I've still not figured out some of the details of how you would find/interact with enemy agents, and this may be the death of it, frankly. I do have a tendency to hit on a good high concept, but fail to wed it with the spatial layout required for it to work when played against an opponent (vs. just being an interesting simulator to watch)

What now?

No idea. I'm trying to just play with ideas without putting strict barriers on anything. Maybe I'll come back to train game with a fresh vision, maybe it's dead. Not sure. Maybe it will become a top-down racing RPG spy train game. Who knows.

I guess right now, this post just serves as a recap for myself that, while I haven't achieved what I wanted on the 'big project' I started up at the beginning of this year, I have in fact done a bunch of things that were worthwhile