The demo for Instruments of Destruction will be available for about 2 more weeks, and will be removed from the Steam page on October 25th. Apologies to anyone reading this in the future who wanted to play, but I need to focus on the full Early Access version.
The demo has served its purpose well, and the usability of the game has improved a lot because it, especially the building process. One usability issue that wasn't addressed is the Flip Vehicle button. People who find it have no problem with it (aside from the bugs), but those who don't get very frustrated with the inability to look under the vehicle.
It's tempting to just stick to my "vision" for the game and only allow a specific camera angle for the game. I love the tilt-shift aesthetic and visuals, but there's no reason the game can't support other camera types as well. The main camera mode will still be isometric by default, but a full free-look camera will be available too. The visuals of the game hold up well to zooming in, partly because there are no textures, and the destruction looks pretty good up close too. So I don't have a good excuse to stick to only isometric visuals. The in-game models aren't necessarily high-detail, but neither are they true low-poly. I actually want to add more geometry detail, so it's possible the final game will have a few more polys (not including the stuff below) than what you see in the demo.
Quick aside: There is now a Discord for the game that I invite everyone to join! [Special thanks to Kevlar for getting this set up]
Because there's a lot to do, it will be a while until the next major update on the game after this one. It's possible I won't post another update until the release date trailer is ready, and I wanted to give an overview of what will be changed/added before the Early Access release.
Over the last few days, I've been working on a GPU-based particle system. I know people like the particle effects that are in the game already, but I want something way better than what's in there now. For some reason, Unity's built-in particle system is the slowest system in the game the majority of the time, and just the smoke from tires is killing the framerate far more often than anything to do with destruction. With the new particle system, not only will the game run faster, but it will be able to do so with 100x more particles.

Particle test with 2 million particles running at 2,000 FPS
If you're wondering why more Unity games don't use GPU-based particles, it's because until very recently you had to custom-code them using compute shaders, and compute shaders are difficult in general. The newest Unity renderers do support GPU particles, but I can't use them without re-writing ALL of the other shaders in the game, and I'd still have the learn the new systems, and they might still have other quirks/issues that I can't control. This path seemed faster overall despite the higher difficulty, and I'd be learning how to use compute shaders at long last.
I'll also be using my new compute shader skills to write a custom water system. The current water looks good to me, but it is causing compatibility issues on older cards, and I can't reasonably expand the water out to the horizon. I'm hopeful to have water impacts generate waves and such. I don't anticipate having floating debris (such as wood pieces), but you never know.
Rope rendering is also pretty slow right now, and could be sped up a ton by switching to compute shaders, but that may wait until after the EA release. It depends on how much ropes affect framerates after the particle/water changes. My overall goal for the game is to have it run locked at 120FPS on my machine (i7-10900k with 2070 Super) except for rare crazy-destruction cases.
Beyond those changes, I'll be adding a sky--which might break my no-smooth-shading rule--polishing some parts, and adding more stuff like islands/objectives/expert mode/etc. I expect to hit at least 70 polished parts for the initial EA release, with a new part or two for every bi-weekly update after that. That stuff is all relatively easy and fun to do (and proven to work), so I don't anticpate problems on that side of the game.
Anyway, I'll be around to check the forums a couple times a week, trying out some different capsule images for the game, and I may post a minor update or two before the release date trailer. In the meantime, thanks for all the feedback and enthusiasm for the demo!
Luke
