JULY 23RD UPDATE - DEAD MATTER
Hey folks, think it’s time to get back to our regularly scheduled content updates after last week.
We’ve had a lot on the go with development over the last little while, so we’re going to try to touch on the real highlights and answer a few questions about some major things that didn’t work very well in 0.6.0 and how we’re fixing it. We’re still working on a couple of major things, so this update will be lighter than we had hoped, however the next couple of updates should be a fair bit heftier as a result of this.
LEVEL DESIGN
Over the last few weeks, LD has been seeing some pretty solid inroads into addressing some of the more minute issues we had with the world that the CA shipped with. Of these, there’s two key ones we wanted to touch on. Firstly, something that never really sat well with us was the lack of meaningful industrial sites in the world. In reality, the Bow Valley is peppered with various industrial sites such as quarries, and yet our version of the world gives no indication of this. With the world we’re working on for 0.7.0, we’re making efforts to ensure that that these sites are finally present.
These sites, beyond adding some much-needed variety in the kinds of locations that players can explore, will also provide a great source of materials for working on bases you’ve setup. In addition, these larger spaces offer more options for combat against both infected and other survivors - be them humans or NPCs - and so should make a huge difference in the combat environments present in 0.7.0
Another item we’ve been attempting to address is the presence of different kinds of loot sources in the world - namely, corpses.
While the asset we’re using here is still something of a WIP, and we can’t guarantee that these will be 100% ready for 0.7.0, we’ve been aiming to provide a touch more of a feeling of death and decay in the places that warrant it. This is something both our artists and our community have wanted since before the launch of the CA, so it feels great to be able to finally start moving forward with this kind of content in the game.
INFECTED AI
Over the last while, the AI for our infected have been finally seeing the start of their refactor. There’s still quite a bit to be done with them, however good headway has already been made on beginning the process of fixing the issues they had and bringing them up to standard. We’re still not quite happy enough with their state to provide a first look at them in action, but we wanted to quickly touch on a few things that we’ve been looking at with them over the last bit.
Firstly, of note is the changes we’ve been making to infected spawning in the world. The system present in 0.6.0 was - for all intents and purposes - trash, and has been scrapped completely. When it was written, the system relied on a grid-based system to abstract the simulation of a pre-defined population of infected being present in the ‘tiles’ around a player, allowing the game world to draw infected from these surrounding tiles to the player’s location in response to player actions. Very few infected spawned outside of situations where a player took action to draw their attention, meaning that once players did attract the infected, suddenly you found hundreds of infected coming at you, even in the middle of nowhere.
This wasn’t going to be an issue, but the spawning system for the infected was extremely broken. The system used was supposed to be setup to allow for infected populations to be defined for the surrounding tiles. This would mean that areas that wouldn’t have many infected around would throw few (or even none,) in the player’s direction, while urban areas could have a larger population.
Unfortunately, this system never worked, and was coded in such an arcane and often backwards manner that it left us more or less unable to spend the time reverse engineering it just to figure out how to fix it, especially when the system was never what we were aiming for in the first place.
In addition to the above issue was that this system would always aim, no matter what, to spawn the infected behind the player character. This feels like rubbish and really contributes to the infected spawning feeling broken and boring. This was ostensibly done to prevent the player from seeing the infected spawning in (a side-effect of the grid-based abstraction of the infected presence in the surrounding tiles,) but in reality this did little to address the issues present with our infected, and just made them feel worse to deal with.
Alongside these issues, auditing the code for our AI has revealed a ton of issues that make it clear that nearly all of our Infected AI code that was used in 0.6.0 is borderline unusable due to the number of bugs and major issues that were impacting performance. As a result, about 90% of the code used in that iteration of our infected has been thrown out.
The good news, however, is that we’ve nearly completely replaced them. Over the last couple of weeks we’ve been working to build a good-quality replacement for the infected, and it’s at a point where we’re almost ready to give a little focused showcase to them. We’ll likely be doing this in the next couple of weeks, so stay tuned for that - it’ll contain a lot more useful information than this little post mortem.
WRAPPING UP
We had initially been hoping to do a deeper dive into the new Infected AI this week, which is why we pushed back the date for this weekly - the hope had been that we’d have everything squared away on them by then, and be able to provide a proper showcase. Unfortunately, this ended up not being the case, so we had to go with a couple of alternate items. That said, we’ll be able to likely provide a more meaningful update next week that should contain a lot more interesting information, and we should also have a few more interesting things alongside we can touch on.
In the meantime, we still want to thank our backers community for their support. Stay tuned for the next update - this one’ll likely be a doozy.
