Devblog 158

I kind of skipped ahead with my plans and added something that is sure to shake the game up a bit. We now have “Scrap” in Rust. If you destroy a barrel you’ll get one scrap, and looting boxes will yield 5-10 scrap based on their type. Think of it as a sweat equity resource: you’ll know exactly how much effort each piece of scrap represents. It also happens to be very valuable. You can use it in the Research Table to craft components. For those unfamiliar with the Research Table, it is a deployable you can once again craft for 1000 Metal Fragments, 25 High Quality Metal, and 5 Sheet Metal. This will allow you to turn scrap into components. Each piece of scrap you add to the Research Table will increase the probability of successfully finding or creating the target item from the heap of scrap. Success or Failure will destroy all of the scrap contained in the Research Table, so the risk is yours to take. Each type of component has a different amount of scrap required to reach 100% success rate. It is important to note that you must have at least one of the components you wish to find to begin searching the scrap. You can’t just pull them out of thin air. Alongside with their traditional yielded parts, recycling components will yield 10% of their scrap requirement in scrap. This feature will hopefully address a bunch of issues. Firstly, it has an actual, measurable value associated with it: you know how difficult a quantity of scrap was to produce. Secondly, it has an intrinsic value: you can use it to craft components. And thirdly, it is accessible to everyone. In my mind this makes it a good candidate for a common currency in the game, and I hope to see people charging for scrap in their vending machines. Because each piece of scrap is associated with a certain amount of sweat equity, it allows you to bypass some of the RNGness of the game. You know if you hit X amount of barrels and you really want it and have been unlucky, you can get that gear you wanted. I’m sure lots of balance changes will need to be done, but I really hope you guys enjoy the feature and give it a chance for the next few weeks.

I spent some time this week improving animal AI again. One thing you’ll notice is you won’t find animals that are randomly stuck any longer. This was a problem with the flee logic and is now resolved. In addition, animal flee behavior is more interesting. They used to just pick a position and flee to it; now they constantly evaluate the best position to flee to. If you’re chasing an animal and you start flanking to the left, it will try and move further away from you by turning right. It’ll also be evaluating dangers so that if you are chasing the animal into a pack of wolves it will most likely turn around and consider you to be the lesser of two evils. I think all the little changes like this will add up and produce a robust and interesting feeling AI.

Let’s talk about raiding. Raiding has been in a pretty bad state for a really long time now and it’s probably fair to say that we have been arguing about solutions without actually acting on them for far too long. This ends now.

Raid Towers

The first major change is that you can now place all building blocks of the twig building tier inside other player’s cupboard radius without being authorized. This obviously opens the door for raid towers, dramatically changing the building and raiding meta of the game. To some this may seem like a drastic step and I’m sure many will immediately take to Twitter to proclaim that we’ve now finally ruined the game in its entirety. Our aim with this is for raiding and building to involve more creativity. Raiding shouldn’t happen on the exact path the architect of the building envisioned when planning his defense. It shouldn’t be a fixed equation of sulfur and explosives to get to the loot. It shouldn’t be binary in the way that either a base is destroyed in its entirety with all loot being stolen or it fails completely with no loot being lost. It should involve planning, risk and reward. You should be vulnerable during the execution of a raid, even if the owner of a building is not online. It should be rewarded to recognize that one of the loot rooms in a building is exposed and can be raided rather easily by a small group of players or even a single person. People should build multiple loot rooms with various kinds of defenses and maybe even a secondary base to fall back to. Raiding shouldn’t only be accessible after farming massive amounts of resources. But a successful raid also shouldn’t mean the complete and utter destruction of a person’s life. It should be a setback, the intensity of which depending on both the layout of a base and the creativity of the attacker. I believe we need various means of executing a raid to get to this point, and raid towers are an important step to experiment with this new reality.

Ladders

In addition to making raid towers first class citizens I’ve enabled ladder placement inside the cupboard radius. Ladders are supposed to be used as an alternative for boosting or be used in the construction of a raid tower. Ladders now have more forgiving placement rules, even allowing you to attach them to the edge of floors. They can also be climbed from both sides and be traversed around to switch sides. Climbing a building with ladders should involve some risk but also be a relatively cheap way to exploit weaknesses in a base design.

Turrets

To make turrets play a more essential role in base defense they can now see and fire through twig building blocks. This is mostly to counteract the fact that people can now place twig building blocks inside the cupboard radius, but I think turrets will automatically start to play a bigger role in base defense now that defenders have to protect against a wider range of potential attacks.

Twig Decay

The twig building tier now decays six times as fast. After an initial delay of one hour until the decay starts, twig structures will decay completely within another hour, making it two hours of decay time overall. This is obviously meant to quickly decay away raid towers after a raid was either successful or failed.

The Future Of Building

I also tweaked some elements of building with this update. I reduced the health of low walls by 50% and added wall sockets to the top and bottom end of foundation steps. This makes them less useful to block the construction of raid towers. Over the course of the month I want to think about what we can do to open up more options and allow for more creativity in base design. We’re currently talking about the possibility of half height snapping of foundations and half height walls. We’d love to hear your thoughts on what you think the future of building and base defense should look like. To summarize, I’m not claiming to have found the final solution to fix everything about raiding overnight. However, this is meant to be a big step in order to shift the game on the right path going forward. I think the ideas behind what we’re trying to do with this are solid and can develop into far more involving raiding and building systems than the path we’re currently on. Let’s see how it plays out, this is early access after all.

This week I finished up on the mailbox concept by Paul! I adjusted it a little bit from the concept by adding some support beams in where I felt it was needed. Take a spin below.

Progress is steady on the Launch Site. Half of the monument is complete, leaving most of the work remaining on the factory and the office building interiors.

I have implemented a caching system for the workshop skins that vastly improves their load times, memory usage and performance. Unity is pretty shit with loading and processing textures in real time, which made skins cause significant stuttering whenever they were loading in. We also couldn’t enable texture compression on them since that would have made the stuttering even worse, which in turn made memory usage and performance suffer. The cache is now created only once, after a skin is downloaded. This does have some overhead, but any future load will be almost instantaneous, completely eliminating any stuttering from the skins after the initial download. This should really help with performance.

More R&D this week. We talked a while back about perhaps having some admire animations for the weapons, so you can enjoy the beauty of your skins close up. Here's a mock up of one.

Ever since the new animals were introduced there was a pretty major bug that some animals were not visible on the client at all while still dealing damage to players around them. I’ve tracked this down to the network group update routines and fixed the offending code.

I've almost finished up the Roadsign Armor rework. The old textures were in a pretty bad shape so I’ve had to scrap them and start again from scratch. They should feel a lot more realistic and less blown-out now. I'm finishing up baking the high poly leg pieces now, ready to move onto skinning and LODding next week. For now here's a comparison between the old and new jackets:

I’ve got the fade in points and control clips I discussed last week finished, and all the changes to the playback system are working now. I’ve been working on some final little UI updates on the editor side, and a code cleanup/refactoring pass now. After those are finished the code side of this will be done! I’ve also been working on tweaks to existing songs, finishing the remake of the classic Rust theme, and moving those songs all over to the new system. I’m hoping to have all that done within the next couple weeks and then this should be ready to go out. I might try to sneak a new song in there too. One thing I’m trying to do with these tweaks is slim the in-game music down a bit so it feels more like it’s part of the ambience when you’re out of combat, and so the drums don’t sound quite as overpowering when you are in combat. Our current music definitely takes over the soundscape a little more than I’d like. Here’s what the editor UI looks like now: The little green clips in the top section are control clips. The ones in this song reduce intensity at those points instead of the slow gradual drop we had before. We’re now dropping intensity at more musically relevant spots now and are able to drop it more rapidly because of that, which makes everything feel a lot more cohesive. The bottom section is where I can set fade in points. The vertical green lines are when the fade will be complete, and the little bit that hangs off to the left on the top of those lines shows the fade length. In this screenshot, you’re looking at a drum clip and can see that it’s only allowed to fade in right before drum hits, and will always be 100% faded in by the time the hit happens. Bringing new tracks in and out feels way smoother and more musical now!