Ah it’s been a while, what’s been going on?
Android using OpenGL3 compat
So as of Godot 4.0 beta 7, you can re-enable opengl3 as a rendering engine for Godot builds. Great! This is actually the first time I think I’ve had the game running on my phone at all.
The good news is it looks about what I’d expect, except buttons are a little on the small side. But it’s very very slow (esp refreshing the display) so this seems to be a mix of opengl3 being very rough at this point in the Godot 4 cycle and simply the fact my phone ain’t my PC when it comes to the lifting.
The opengl3 engine isn’t really expected to be any good in 4.0 and it’s not expected to be properly good until sometime later in the 4.x cycle so this isn’t unexpected at all.
The other apsect which didn’t work at all is GUI control input from touch. I have “Emulate mouse from touch” enabled which should be all that’s required but either there’s some other major problem stemming from the not fully cooked opengl3 renderer, or I have a real problem with input in Android.
Anyway, hopefully the last remaining Vulkan rendering bug preventing the game running at all on Vulkan on mobile will be fixed soon, and then we shall see what the real causes are!
Dialog and Quests
Both of these systems are ones which I am not particularly familar with in how to approach them. However, the basic dialog frame with player input works. It doesn’t have next-page buttons like it should, but that’s not too difficult to work on.
At the same time, there is rudimentary support for recognising that places in the world can be entered and therefore ask about this. This needs a better indication, as we shouldn’t use the classic quest exclamation mark to mean “enterable”, like shops are not quest points right?
The quest plan itself is roughly bullt in terms of what classes will do
what. There’s a QuestPlan
which contains the whole set of steps for
this quest, and where we are up to, and it holds at start point and
rewards and description for the overall quest. This also stashes an
array of QuestObjective
and each of these are our requirements to
meet. Go talk to people, get objects, kill stuff.. those are steps.
The data model looks okay, but it needs a lot of filling out of the actual quest generation and objective handling. But baby steps. These are the third and second most difficult parts of the game I have to implement. (The most difficult is combat!)
GUI work
I’ve also been fixing up the UI here and there, and trying to make it look a lot more game-like. I’m still using the main Godot GUI components and it turns out that I can theme them much more than I thought was actually realistic to do so. Now I have much more richly-themeed components I can use instead of layering a bunch of controls and 2D elements to fake it.
Wrapping up
Progress is very slow (slower than it’s ever been), but I’m chipping
away at the code as best I can, while the beta releases of Godot truck
on. Apart from some colour lookup issues, and FileAccess
API rework
that landed in beta 5 and beta 3 respectively, it’s been remarkably
smooth sailing in the beta releases so far, so I’m pretty happy about
how things have been going.
Cherry-O!