Necessary… Things

So, cameras. Well, wait, let’s start here: 

So, yes! I have actually started working on gameplay instead of the dark void of graphical features and visual tweaks and fixes and optimizations and etc. etc. etc. That’s the altar you see in that there pixel image; it’s what defines your starting point in the world, serves as a beacon for other players to come and hunt you down, and also is the one thing in the game world you must protect at all costs. Because if people can’t worship your godliness then, I mean, really what’s the point?

I’m also now, officially, a resident of Washington, DC. I can see the Washington Monument from my apartment. There’s a big ol’ white house about half-mile down the road from me. And people honk the horns of their cars a lot. A lot. 

Anyway, back to things you care about: I’ve also revisited the pipeline for destruction on the whole. I’ve had to (at least temporarily, but likely, at least, until post-release) put destructible terrain on the backburner. I told my good friends over at VoxelFarm that I just couldn’t keep working on both the UE4 integration/shaders in parallel with the rest of the work on Sacrilege that I had to do and, because they’re great, they were totally understanding and cool and even refunded me my previous month’s payment. I really hope to revisit that tech at some point in the future as I’m fairly certain that, unless Epic has been cooking up something dramatically unexpected, it’s going to be the best alternative for destructible terrain in Unreal Engine 4 for quite some time. But this lets me focus on more meaningful destruction and a more robust simulation as far as the actual game systems go, so I think it’s a good call for the time being.

I’ve also pretty much ruled out the destructible mesh editor within UE4 as it appears to be nowhere near as amazing as I remember it being when I first encountered it. I kept feeding it meshes and it kept generating this really flat, planar, uninteresting, and unbelievable pieces of fractured geometry and that… That’s just no good. So, I did what I normally do what I encounter an art problem: I opened up Houdini.

And I was able to rig up that simulation in about a half-hour. And upon importing into Unreal Engine 4:

So, that’s all well and good, right? Well… There are some issues. For one thing, there’s an issue whenever I place the fractured mesh in-game where, somehow, the physics simulation gets corrupted to such an extreme point that it actually generates an expanding black hole of graphical artifacts. Which is pretty impressive, but also not incredibly sustainable. And then there’s… This issue:

So, I have some issues with that pipeline to work out. After posting on the sidefx forums, the developers of the Houdini Engine UE4 plugin said that destructible mesh support directly from Houdini was something that they’ve had a lot of interest in and would likely look into very soon. That’s not exactly a timeline or something I can count on, but it’s something! In the mean time, my experiments will continue, but having a fully destructible building isn’t exactly a blocking task by any means.

Oh, and I’ve officially plonked down the money that I would otherwise put towards things like eating in favor of a CoherentGT license from Coherent Labs. They’ve been an absolute treasure to work with (and they also had an interest in me doing things like eating), so we worked out a pretty great deal that lets me eat meager things while still having their top-of-the-line tech and support. I plan on writing at length about their tech/UE4 integration in the future as, well, it’s pretty fantastic. And it allows even a plebe like me design a responsive, dynamic, and good-looking interface, which is saying something if you’ve ever seen me do interface design before. 

Beyond all that, I’ve been experimenting a lot with world scale lately. I accidentally lowered the cloud altitude to a somewhat, well, low height off the ground and then just had fun zooming around in the clouds for more time than I should publicly admit.

This actually led me to two things:

  1. Oh, I have cloud shadows? Time to figure out why I didn’t know this until now. (So, now I have cloud shadows when the clouds aren’t chilling just above the treeline). 
  2. My land mass is entirely too big for a reasonable player count if I want the game to last in my target 15-30 minute range. With the editor camera at 80% its maximum speed, it was taking me a solid 5-7 minutes to go from one edge of the world to the other. And this map (max), my vertical slice map, is being targeted for 3-6 players. So, I reduced the size of the terrain to about 60-65% of its previous size, regained some GPU cycles, made the world still feel large without it feeling overwhelming and bland, and got a generally better idea of the kind of terrain scale I should be targeting.

I’m sure I started this with some other ramblings to cover, but I’ve since forgotten them. I’m still aiming for a vertical slice in mid-April, but I think calling that “optimistic” is a laughable understatement. So, let’s target mid-to-late May.

Oh, that’s right. I started talking about cameras at the beginning of this thing. Well, cameras are the worst. Just the worst.

And I’ll try to update this thing more than once a month from now on. 

3 Comments

  1. Anony mouse on March 21, 2016 at 2:32 PM

    I’d like a vertical slice of cake, please.

    • Trent Polack on March 21, 2016 at 2:48 PM

      if i made cake, you probably shouldn’t eat it. i burn oatmeal on a routine basis.

      oatmeal.

  2. monkey on August 19, 2016 at 9:17 PM

    Another chocie to edit the destructible meshes is use nvidia PhysxLab https://developer.nvidia.com/apex-destruction-physxlab-tutorials

Leave a Reply