Ciess Retrospective
Ciess (pronounced like "CS") was the old VR Jam prototype version of Darknet. I've mentioned it a lot since announcing Darknet, but as I focus more on newer stuff, I plan to stop referencing Ciess and spend more time looking forward. Before I do, though, I thought it would be nice to do one final summary of where Darknet stands in relation to its predecessor. So, here's the last word on Ciess!
First off, I'll acknowledge that Ciess did a lot of things right. It was praised for being easy to learn, addictive, and comfortable in VR. All are important traits for a game, and I'm duly proud.
But. But! It also kinda sucked. The randomly-generated levels felt same-y and too long. It started with a long-winded and boring tutorial. Few players cared about the time attack metagame. There wasn't much strategy in choosing which nodes to hack. The hacking puzzles themselves were simplistic, and their optimal strategy was way too consistent and easy to figure out (just shoot into a corner as fast as possible).
Worse, it never looked fun. A lot of the positive feedback said, in some way or another, "it was a lot more enjoyable than I expected from the screenshots". This might not seem important, but I really want my games to be attractive; it shouldn't feel like a chore to give the game a closer look.
So, I started out development of Darknet by deciding what I most wanted to keep from Ciess and what I most wanted to change:
1) The hacking puzzles in Ciess would have to be thrown away. I built an entirely new hacking mechanic to fill that role in the game, which I've already described in greater depth in my first dev blog post.
2) The art would need a serious makeover. I ended up getting help from the immensely talented Michael Heald, and you've seen the results of that help in the screenshots and trailer released so far.
3) The network-layer strategy (e.g. choosing which nodes to hack) would need to get a lot deeper, while hopefully staying easy enough to learn. I've improved this aspect of the game, introducing concepts like ICE and node armor, but very little of that effort has been shown so far. I'll be describing this layer of strategy in greater detail on this blog at some point.
4) The networks themselves would need to become more varied. Random generation is a great tool, but it's hard to do well, and the results all look the same if it's done poorly. I've been working on different ways of adding variety without breaking balance or cohesion, and I think it's a solvable problem, if not a solved one.
5) There would need to be a stronger motivation than "beat your fastest time". Ultimately, I think the player needs to be able to actually lose the game. Right now, that's reflected in a strict time limit, but I'm still considering other systems. There's also a much more advanced metagame layer planned, but I'm not quite ready to talk about that yet.
6) After adding all of this complexity, the game would still need to be easy for new players to jump into. So far, the plan is to start players off with a simple subset of the mechanics and then add in the complexity one mechanic at a time.
I'm pretty pleased with Darknet's outlook at this point. It still shares some of the weaknesses of Ciess, especially the difficulty of quickly showing or teaching the newer, more unconventional game mechanics. (That, by the way, is why the initial reveal has focused so much on the theme.) But overall, I think that building Ciess and seeing its weaknesses ended up giving me enough perspective to build a more solid foundation for Darknet. We'll see if that's true once it's done!
