Dear Silica Community,
Have you wondered what makes Silica so unique and what it took to combine the RTS and FPS elements in one game? This is what Dram got to present as a special guest during the Game Access Conference in Brno this summer.
We managed to break down his presentation into 3 smaller blog posts to give some insight for those interested in game design. We'll also dive into what challenges Dram faced when creating Silica and how he managed to overcome them. So let's get right into it!
The main reason why Dram wanted to have an authentic scale in Silica was because he wanted to put across the scale of the world of Baltarus. The scale of the vehicles and building being so large gives an interesting feel. It's a feel that Dram himself recalls seeing in Dune 2 where the vehicles were very large compared to the troops, and he wanted to portray that in Silica.
Large building challenges
There were several challenges when it came to large buildings, one in particular was building placement. Because the terrain is not flat but more "alive", it was difficult to write a system to snap the building on the terrain without having a part floating. To solve this, Dram defined point on the building which had to be beneath ground and points which had to be above ground. Another issue was recalculating the navigation graph around the building when it was placed or when it was destroyed, because vehicles need to avoid the buildings and troops need to be able to walk on the buildings.
This is the fabrication technique - the visuals that players see when a building is being constructed. It was chosen mainly because it was the most technically feasible for the massive team size. It was simple to set up because Dram had these square shaped pre-fabs. That way he could cover the building and therefore do it so they progressively appear, cover the whole area where the building should appear and once the building is complete, they disappear. One of the main inspirations behind this, were 3D printing techniques!
A physics driven world
When Dram was working on Take On Mars, he did a post-mortem on that, where he stated that it is a bad idea to do everything physics driven! Evidently we went back on that, but the reason at the time was mainly because processors were slow and because it adds a ton of issues. However, thanks to the experience with Take On Mars, it was much more easy to go with everything physics driven in Silica. Another reason why physics was the way to go, is because the tires of very large vehicles, for example the Harvester, would be very difficult to simulate without physics and subsequently, their movement would look very incorrect given the fact that they need to roll over obstacles.
Because Silica blends two genres, Dram couldn't take the route of having simplified colliders, for example pipes where you can see through them because they're hollow but the collider would cover the whole thing. Because the FPS element exists there, and these pipes are large, you could theoretically shoot through them because it's what a player would expect from an FPS. Having the bullets would bounce off the side, it would look very strange and in Dram's words "you wouldn't fulfill the expectations of the players of the world from an FPS perspective". Plus, Dram thinks it's more interesting to have these accurate colliders and things bouncing where they're meant to! Given the fact that there are realistic ballistics where bullets can ricochet, there's penetration, it's important to have realistic colliders.
Challenges with colliders
One of the challenges with regards to colliders, is that the more complex the colliders are, the bigger the problem is to update the navmesh. It takes much more time compared to a simple box as a collider. The challenge is to make complex colliders but not too complex because then you run into performance issues. In the case of Silica, the more complex the collider, if the crabs were walking on them, it would take up more performance because there are much more physics calculations required on the actual collisions on the collider.
We'll let you digest this first part of Dram's presentation but if you want to hear the whole interview in depth, you can check out the video!
The next part should be coming out soon so stay tuned and we'll see you on Baltarus!