My goal with Venture the Void was to gradually replace each part of the game in-place, working towards overhauling the visuals as a first step.
You can read about this in more depth, here:
I had thought that after an initial organizational-refactoring burst, I could work periodically in 2 week chunks, alternating with other projects and working on what I wanted to.
I've spent the last 4 months or so of development time (itself just bits of time I piece together between other responsibilities) exclusively on this initial burst.
The good: this is largely done now. I did a ton of cleanup and solved a lot of hard problems. Big chunks of this work was interesting and fun.
However, I think this plan is too time consuming to be worth doing.
It would work— but not fast enough, so for now I'm shelving and-or cancelling this.
I want to describe the problems in detail. As usual this is for my own benefit.
Workflow is too slow because the code-- essentially all of it-- is C++. Since Venture the Void, I've designed around faster workflow. This means putting more and more code into Lua, keeping the C++ part of the engine comparatively minimal.
Paradise Never has hotloading of most of it's code-- I can tweak UI behaviour and appearance, object behaviour, creature logic, npc logic, dialogue, etc. without re-starting the game. I can also reload shaders to tweak visuals.
The C++ part of the engine, becuase it's comparatively lightweight, is also relatively fast to build.
In VtV there is no good way to move parts of the code base away from C++ without a lot of focused effort.
VtV is not such a terrible experience, actually. Playing it now, I think it works better than I ever gave it credit for.
But it's not really the kind of game I would make today. I would do away with the shooty parts and make it purely exploration/mystery-focused. I'm interested in making a "plug in" system for quests that players could write additional modules for, for instance in Lua. But those goals wouldn't be achievable until much later in the project, after I had fixed the visuals and finished vast amounts of other refactoring/reorganization.
Since I would be going through monumental effort but ending up with this same gameplay, but just improved visuals, it would prove ultimately a bit unatisfying to me.
Redoing the UI would be necessary at some point for me to feel happy about the game, but it's a mountain of work and not fun. Likewise, supporting controllers and solving accessibility problems would be mountains, though smaller.
As much as possible I want to create work that I can release regularly, now. Even if improvements are tiny, it's satisfying to push things out.
A VtV visual overhaul would itself need to be mostly in place before I would want to push anything out. Partial updates would cause the game to lose any semblance of visual cohesion since my goal is to change it quite drastically.
If I decided to, one day, it would make much more sense to pull those parts of VtV into a new project, and build up from there, releasing regularly. This would inform the design as well.
Finally, I should be willing to let the game stand as it was originally created. Why reform it into something else, anyways?
My original reason for this is simply that it would always be a complete game, even if step by step it turned into something else. Secondary to that, I do like a challenge and particularly this kind of work reorganizing/cleaning up code I do find enjoyable. To a point.
It's super important to abandon things when they aren't working out, even (especially) when it's a hard lesson. The time I would spend on VtV would destroy so much other interesting creative work before it could even come to life.
So-- as far as the complete reworking of the game into something else, consider it canceled. I've got more interesting things I could be doing.
Obviously I will keep the game up on itch.io.
I am also nicely set up to push new builds to itch.io, so will continue bug fixes as much as reasonably possible.
Two other things are possible in the future.
First, I might put a bit more work in at some point in order to expose the world generation code to players. This would be one good outcome of the work I did over the past months.
Second, it's not totally out of the question that I could tidy the rendering a bit-- but without changing the art "style"-- to improve performance and compatibility.
Right now though I'll turn my attention back to Paradise Never and RAPALA, which are both feeling might appealing right now ;)