Press Kit Wiki

Hurtworld Update #71


I finished one of the muzzle brake thingummies and a standard magazine, looking tasty (if you ask me) I also got onto testing the new Colour Mask Shader, which has been upgraded and now allows for a l0t more options. We can now mask with varying levels of opacity whereas before we had to mask 100% anywhere the designs would sit. This allows us to get a blend of mask colours that the player can change and the ‘set in stone’ Albedo (used to be called diffuse)

We have just this afternoon added functionality that will allow us to override the PBR properties under the mask. This will allow the player to change their Colour Masks from say Matte paint to Glossy or even Metallic. Whether we let people change these at their leisure will be a choice based on aesthetics and performance.

MaskApplication MagTipSkins_0003_Layer 1 MagTipSkins_0002_Layer 2 MagTipSkins_0001_Layer 3 MagTipSkins_0000_Layer 4


Hey folks. This week I’ve been working more on the crafting workflow, and looking into two new mechanics – repairing and scrapping items. Many items of course have durability, so after a while they’ll break. If you take it to the crafting machine that made the item (yourself if it was handcrafted) you’ll be able to consume some of the blueprint’s uses to repair the item back to full health. Doing this will require a new resource, scrap, which will be found in a lot of places around the world. You’ll also be able to turn some items into scrap, which will hopefully incentivise players to pick up loot while at the same time freeing up inventory space so it isn’t filled with a bunch of stuff you don’t want to use.


A lot of stuff influences how these values work – for instance, a degraded item will turn into less scrap than one that is brand new. Rarer items turn into more scrap and require more scrap to repair fully. At this point you can use any amount of scrap to repair your items somewhat, so even if you don’t have enough to get an item back to full health you can prolong it breaking a bit. We’re also intending on making repairing an item more efficient than crafting new ones, to create a sense of attachment to blueprints. Your gun is your gun, and while you can make other copies to give to your mates, it’s better to keep and repair it.

I also changed how Item Slots work again, improving their extensibility and ability to be customised for specific purposes. One example is the two new types of slot I’ve created for the repairing UI – the proxy slot, which can point to an item but not actually store it, and a collection slot, which easily shows all types of a certain item in a slot UI for showing how much scrap you have available to repair with.


This week I’ve mostly been working on the hotfix. The way all our construction items are setup (both machines and construction hammer objects) a construction volume is set around them where we check against overlapping objects to determine if the object is able to be placed. For the update I separated the volume checks against players from the other checks against terrain, constructions and other objects.
This allows us to have a larger check volume against players that totally covers the entire collider of the object. This isn’t possible when checking against terrains, constructions etc. as we often allow them to overlap. For instance you can see the new storage chest validation, if we checked all layers against the outer volume you could never place it on a slope, only where it is perfectly flat.


It also turned out that the round pillars you can build with the construction hammer were inheriting a volume check from the slightly smaller square pillars so these have also been updated. If you are using a construction mod that’s setup to override the standard construction set and it contains pillars, it will need to be rebuilt!

Vehicles have had an entry check added that’s essentially the same line of sight check that they run on exit, this stops players entering vehicles through walls so you shouldn’t have to worry about parking on the edge of your garage where the interaction trigger can occasionally stick out. Players will no longer be able to use bugged vehicles as a bridge to get inside rocks and bases.
Also on the vehicle front, crashes have been changed so if the vehicle has an attached body simulation (where we connect a dummy rider to the vehicle via a joint, both the kanga and goat are setup this way), when crashing the crash vehicle is spawned at the location of the attached body rather than spawning at a preset location with a quick sight check. This solves a bug where sometimes the sight check would unexpectedly pass when it should’ve failed during multi-vehicle collisions, occasionally allowing players to crash through walls.

I also spent some time setting up vehicles in itemv2 ready for our new coloring system, now that we can change colors on a per item basis we will change vehicles so each panel can be individually painted. The backend technical side of things is working now for vehicles, up next is working out how this will fit into gameplay (ie. how to color something, what it costs, where to get these resources etc)


This week I’ve been pushing forward with survival content, filling all our new awesome systems with a prototype progression for our upcoming survival test and have to say its coming together really nicely.

I started the week with a big wishlist of stat modifiers for equipment and am making my way through evaluating feasibility / implementing them as I go. To support things like clothing giving damage bonuses etc I extended the system I spoke about last week to enable attack effects to be derived from an entity stat on the attacking player. To achieve this, we create a second damage stat on the attacking player, one that can be additively contributed to by all equipped items via direct damage numbers or percentage multipliers, then a weapon will specify that it uses the result value of that stat as the attack effect amount. Another example of this is a mining helmet increasing your mine power while you have a pickaxe equipped.

This is our current wishlist for equipment stats:


Not everything here will make it into the game, but I will do my best over the next week to implement as many as possible.

To support these new stats I rewrote a few tooltip panels for equipment to show their effects more clearly and support the new stat based effects.

I also spent about an hour building a system to procedurally name items, its mega basic and has no correlation to stat affixes, but at least makes it possible to tell 2 generated items apart. Eventually icons will actually render differently for different items to make sorting stuff easier.


This name override system also makes it possible for people to name their own items. I would need to bang together a new machine to allow it in game, but all the infrastructure is there now.


After landing on a style we liked, the character was taken to it’s game ready state where we realized the style did not sit right with the low polygon limit. After a few more style iterations we decided taking the character back to basics would be the best bet, so I’ve essentially done an anatomically correct version of the original character with updated proportions and face while still staying true to the look and feel of the original character, hopefully this will be a familiar face to you all. Due to the styling challenges we ran into, the female character has been put on hold for another week, allowing us to work on making the character look great in game, a process which transfers to all other characters.


Hurtworld Update #70


This week was mostly taken up by the project restructure, going through everything and fixing broken references n all that fun stuff. The good news is I am now ready to start pumping out content!

While building the procedural pickaxe generator, I came up against a limitation of the Mutator framework that made complex stats and buffs on items tricky. To solve this I added a new feature to our state machine layer that allows blackboard variables (data that the equipment simulation is driven off) to be bound to entity stat values.

Last weeks devblog shows an item generator that explicitly sets the pickaxe swing speed. But what happens when multiple things want to affect this?

A good example of the use for this is melee attack speed. We want multiple items you have equipped, like gloves, weapon, weapon attachments etc to be able to give bonus melee attack speed, and things like debuffs to be able to reduce your attack speed by a percentage, we then want to take the result of those stacked calculations and push it into our state machine so things like our attack swing simulation and animation runs slightly slower / faster. There heaps of different sources of entity stat effects in the game already, and utilizing this system to drive equip simulation gives us a heap of flexibility. This will extend to things like reload speed, fire rate etc.

This week I’m going deep on content, which should give us something playable in survival soon.

Introducing TehSplatt / Character Rework

Tehsplatt has taken over from Gavku and is doing a pass on the main character art. Our main objectives with this character pass is to solve a few issues with the proportions of specific body parts. While he is at it, we’re exploring ways to make the character fit in with the surroundings and creatures a little better, while not looking strange holding realistic weapons.

Remember… Artists are not programmers, they can’t fix bugs or implement features, and have very little baring on when releases are ready.


Hey, so I’m the new guy. This week was spent redoing the main characters and coming up with a style that can be applied consistently to old and upcoming art. After many iterations this is the style we have landed on, we feel it gives the character a realistic feel anatomically while still staying true to the original cartoon style. These are the first drafts of characters in this style and they still need a lot of work. Today was spent sculpting the female player base in the same style and will most likely be able to be shown in the next dev blog.



I battled the infamous new computer monster this week. My old PC was getting a bit tired so I built a new beefy machine that is humming along. The old computer’s 16GB of RAM was maxed out and I was getting slowed down a lot in 3D Coat.

I did manage to get one of the Stocks completed for the AR15. The gun is starting to come together.

Stock_Routine_0003_Layer 1 Stock_Routine_0002_Layer 2 Stock_Routine_0001_Layer 3 Stock_Routine_0000_Layer 4


This week I’ve been helping out Spencer and Cowtrix fix up issues due to splitting our SDK into different projects. While this took up most of our time this week it was definitely worth it as we’ve reduced our mod build times from around 10 minutes to under 10 seconds. This makes rebuilding while tweaking things way less painful and this will also be an even bigger help to modders who aren’t able to run the Hurtworld client from the Unity editor and make changes at runtime.

We’ve also received more reports of rock bases so I’ve been investigating the problem again. I discovered another rare edge case where our inside rock check would fail and players were able to place an ownership stake inside a rock. I’ve refined our method to deal with this case and now that our rock check seems to be getting robust and is performant we can afford to add it to some more items. We’ll be releasing a small patch sometime later this week to fix up the rock check and add it to the rest of the machines (lockers, chests, drills etc.). We’ll also be adding rock checks to the player that will kill any player found inside a rock.
Hopefully these changes plus a few other bug fixes to stop players getting inside the rocks in the first place will get this issue finally sorted!


Hey folks. This week I’ve been implementing some changes to the crafting window, and implementing the changes to crafting we discussed last week. I’ve also been fixing the machines and various assets up from the SDK split. It’s been great working in this new workflow where mod builds take a tiny fraction of what they used to. Machines are a bit neater in code structure than they were before as well.

The new crafting workflow is coming together nicely, and while the UI still has a bit of tweaking to go it already works a bit nicer than the old version. You can equip and destroy handcrafted blueprints, and I’m currently figuring out what the best way to do repairs are without over-complicating things. The recipe list doesn’t have categories anymore, as categorization is likely to happen on a specialised machine level now instead of cramming everything into the one workbench. There’s also a nice little search bar, we’ll see if it ends up useful enough to actually be worth it.


Handcrafted recipes will actually work just like an extra “hotbar” of restricted inventory slots that the handcrafting menu can go and find. You’ll be able to manage that by context clicking certain blueprints and then clicking “Learn”. At this point you have 12 blueprint slots, although this might change. If they are all full, and you want to learn a new blueprint, you’ll have to open up the handcraft menu and delete one. 



Hurtworld Update #69


I finished off one of the barrels for the AR15 this week, there will also be custom coloured designs to go over these as with the cars. I am waiting to finish the current models with textures before I go and do all the graphic skins.

Really happy with how the PBR materials are turning out on these. The lighting in these screenshots doesn’t do it much justice as it isn’t the runtime game lighting. Now onto the Stock, then probably the Magazine then a Muzzle Brake or Silencer.

AR15BareelKeg_0001_Layer 3 AR15BareelKeg_0000_Layer 4 AR15BareelKeg_0003_Layer 1 AR15BareelKeg_0002_Layer 2


This week I’ve been working on our color customization system again that I explained in blog #65. Whilst the system was mostly working it ran into problems if two models were baked together that shared the same space on the color lookup texture. This was a rare occurrence as we allocate texture space based on slot type (ie. Head, Legs, Chest etc.) but something like a roach or goat wheel uses the same model baked to different wheel bones dependant on the slot it’s equipped in and also when dealing with mods we can’t guarantee the slot has been assigned correctly so I’ve developed a system to detect overlaps between equipped models and assign spare space to a model when necessary, overriding the models custom UV channel to point it at the new space.
I created a quick color randomizer, added some colorable goat wheels and setup a whole bunch of items with new color masks that Gavku made to test the system and the results have been good! Colors are now synchronizing over the network and color texture space conflicts are being resolved properly!

I also spent some time working on vehicles, fixing some bugs around lights and the new item system. I took this opportunity to make some improvements in vehicle parking behaviour, now wheels will update their position if the vehicle is moving at all and also whenever new wheels are attached which means no more wheels sticking through the ground when the vehicle isn’t occupied and when first loading them in.

I also made some improvements to our mesh baking tools again, our custom mesh attachments now store a reference to the model they were created from and you can recreate the mesh attachment in one click from its reference after making changes to the source model. I’ve also added a tool to create mesh attachments based off another mesh attachment, as long as they share the same space (ie. have the same offset to the same bone). This saves the hassle of having to position, rotate and scale each new model relative to the character skeleton.
These tools should help speed up the work flow both for us internally and for modders once we get to release.


Hey folks, this week I’ve left off creature progression and have been looking at the crafting machines, blueprints and general UI improvements.

Crafting is going to change a lot in ItemV2, and we’ve been discussing internally and on the devblog what those changes are going to be and why. A big part of this is finding a blueprint for an item you want, and then choosing to invest in that design by building some machine to build it. I’ve been implementing this machine, that uses a model that you may have seen before but has so far not been used. I’ve also been trying to keep in mind moddability when creating these items, so a lot of it has been changed to be very component driven and smart about hooking itself up. Hopefully we should get to a point where it’s very easy to attach visual effects to a machine that will automatically turn on, scale and react to the what the machine is doing.


Handcrafting will change as well. We see a system where all players have base recipes that they are able to handcraft, like now, and then are able to consume a limited amount of blueprints found in the world to be able to handcraft them. Not every blueprint would be able to be used in this way, but it means that you will be able to repair some items while out in the world, give out and trade items that are uniquely yours, and regear if you die and cannot respawn at your base. The effect of this is to hopefully tie the items that you find and are valuable to you, to your character in a way that persists across deaths, while it still being worthwhile to hunt and kill other players. This has flowed through to changing how a lot of the UI is built, which will allow for much more of how the UI works to be changed by mods. Mods should now be easily able to create their own UIs for custom machines and objects.


This week I implemented the blueprint system that allows any loot table (creature drop, resource node, crate) to generate a unique item, but instead of dropping the usable item it drops a blueprint to be crafted. To achieve this I used the nested item system used for weapon attachments, and wrapped the fully generated item in a blueprint container. This container then has its own loot table to decide how much the item will cost to craft. This can be configured like all other loot tables to increase costs based on rarity or quality of the roll on the item. For example, if you roll a rare pickaxe that give a big attack speed boost that most pickaxes don’t get, it can increase the recipe cost to craft it by 20%.

As cow_trix described a bit above, these blueprints will come in 2 varieties: Handcrafted and Machine crafted. The purpose of this is to give relatively new spawns a couple of item tiers where they can learn blueprints directly onto their character. Meaning before you have established a base, you don’t need to setup machines to craft your items. This makes it a little safer at low levels due to the fact that you can find a blueprint and learn it right away and cannot lose it.

Once you progress a bit further, higher tier items will require a crafting machine to use. Meaning if someone takes over your base, they can also take over your crafting facilities.

These systems are pretty much in place now, just waiting on that magical elusive content.

Build times
Once I started working on content I realized that the workflow for building anything from our content project had gotten so slow nobody could really get anything done. To preview an item in game took 5 mins of waiting. I spent a day splitting our content up into smaller manageable chunks and made our build process run in parallel. Taking content from authoring to build takes roughly 15 seconds now.

Lastly we bid farewell to Mr Gavku, who is taking a break from work to look after his new baby. All the best Gavku!

To re-bolster the Art team we welcome he who shall be known as Tehsplatt, who is a 3D sculpting wizard and once held the world record for Wolfenstien speedrunning. Expect to see a lot of him around the place.

Dev Blog