Hey folks,
I kinda crashed out yesterday during the time I had planned to get some work done. Most of the time I spent with my wife and son, but during his nap and later in the evening I'd been planning on working. Slept instead, which was probably for the best. I've been burning it pretty hard.
This whole process is so much fun and is exhilarating with everyone being so excited about what is going on and so helpful in figuring things out quickly, but it's also mentally exhausting. There's the ever-present anxiety of not getting things done in a timely-enough fashion to launch the kickstarter before the last of the existing funds leaks out of the bathtub so to speak, etc. We're not rushing anything, but trying to find the middle ground between being fast and being both faithful to the existing game while fresh is... well, that part can be stressful.
Obviously I'm super behind on the graphical load tester thing. Blue and I have had some long talks about a number of things relating more to far zoom mode, trying to figure out a balance between performance and clarity there. I think that we may wind up using something approaching my prior graphics pipeline that I created that's used in Starward Rogue for that layer of the game, but we'll see. That would be a triple-hybrid graphical pipeline for the game (a version of my own dynamic batching for sprites layer, the unity gui for that part, and then a more traditional unity layer for 3D), but ultimately I think that will help us to get the best performance at all layers. We shall see.
With Release Raptor, I spent a couple of months just focused on performance at all levels, from physics to simulation to graphics. I want to do that same thing here, and I will have PLENTY of time to do so if the kickstarter succeeds, since those areas plus general gui functionality will be my main programming focus on this project. But I guess my stress comes from trying to show you something that is representative of what the final graphics will look like, but without giving you "bullshots" of things that look too good to be practical in the end.
Ultimately there might be quite a few gradations to what we do in general, I'm not sure. Basically some "cinematic" modes where you can see some (comparably) super detailed stuff and record what is happening, but it's not remotely at that detail most of the time. I think that would be fun for times when you've already basically set up your ships in such a way that they are just blowing the enemy to bits, and you want to watch that happen "from the trenches." That seems like a really satisfying place to watch from, and nice and immersive. With that sort of approach I'd have a pretty huge "budget" in terms of GPU power in particular, because I can use all sorts of tricks to make that scale as needed for your hardware.
That stuff has been weighing a lot on my mind more than anything to do with the actual gameplay, to be honest, because in order to do a compelling kickstarter video I need to have some "flash in the pan" exciting graphics that show what a difference this is from the prior game. I'm just so worried about giving something misleading by accident, though... or on the other end not really stretching our legs enough (so to speak) and showing the sort of thing that we can REALLY do if given a proper GPU budget for those cinematic moments.
I guess one way to handle that would be to design some of those things as cutscenes that are definitely 100% in the game either way you cut it, but are on the main menu basically. So it's realtime graphics and we get to show off what we can do when we fully open the throttle up, so to speak. And you get to see that sort of thing in the game itself in the form of cutscenes for things like story and end-of-campaign stuff (that will be such a huge improvement in general, too, eh? Compared to "you win" in a really cruddy font).
Anyhow, I am alive and am going to be back working on the design document again.
