16 July 2015
I've been taking it easy this week after my couple of weeks away. The team held it together pretty well, apart from the one day late patch/blog a couple of weeks ago - so I can probably start to relax more now.
The female model is in. But not actually. We still have some stuff to work out, like making alternate clothing models for her, but if you're an admin on a server you'll be the female model.
To clear up some confusion, when we it does go live you won't get a choice of whether you're female or male. We're not "taking the choice away" from you. You never had a choice. A man's voice coming out of a woman's body is no more weird than an 8 year-old boy's voice coming out of a man's body.
Our intention has always been to use the Steam Workshop in a similar way that TF2 and CS:GO do. Our original vision of Rust was more of a MMO with a single huge universe that all the servers are connected in, rather than a multi-server game where your progress is different on each. I know this shit makes people nervous, so I'm going to try to answer some common questions head on.
Are you making the game pay to win?
No. The items we're adding are cosmetic, and we're not charging for them. They're awarded to players based on playtime.
How does it work?
You join a server and play for a few hours. You are randomly awarded a red t-shirt in your Steam inventory. You can now craft that red t-shirt. It works globally. You can join another server and craft that red t-shirt.Is this the case for all blueprints?
No. Just some select clothing items right now.
What about armour?
We have stopped short of making any armour blueprints a part of the global system. On one hand it's pay to win because you're unlocking shit with money that gives you an advantage. On the other hand you'd still have to harvest resources to craft it, you stand a chance of having the blueprint drop within your first hour of gameplay anyway, and other players could benefit from it by killing you and taking the armour.
So we haven't made a decision yet - let us know how you feel.Are you selling blueprints?
No. I'm not saying it won't ever happen, but we don't see any benefit to that right now. The only way to get them is by random timed drops, trading or buying from other players. We see this more like a card collecting game where the cards are something you actually want.
The idea is that you can play rust for a while and get a bunch of blueprints. If you get duplicates you'll be able to trade them for blueprints that you don't have or sell them to lazy rich people. And because the marketplace is a common market you can go to the inventory and trade your blueprints, a bunch of CS:GO guns and some TF2 hats for a copy of Bad Rats. That's the future.Will the game have stupid gimmick clothes?
You mean stupider, more gimmicky, right? We'll have to cross that bridge when we come to it. There's no workshop system yet, we're not accepting clothes from the community yet, so it's too early to say. What's more likely to happen is that community contributions would be categorized into "fun" and "serious" - and servers that are set up as serious wouldn't allow fun clothes.Does this mean you've fixed all the bugs and the game is finished, since you have time to piss around with this shit?
Nice. The thing I guess to understand is that Rust is STILL in Early Access and we need to take advantage of that fact. This is the time for experimentation - because once it's not in early access we won't be able to do these kinds of game changing things - and we'll wish we had.This is fucked.
Maybe. Let's give it a try and find out.
Steam Inventory Items
There's a limited amount of items in the inventory system right now. We've got a couple of t-shirts, four long sleeved t-shirts, a pair of jeans and eight jackets. This isn't much, but it's a start. I would expect a moderate player to have been awarded most of those in a month.
It's important that we do this now so we can trickle in more items every week as we go along, and establish the system for when we let you submit your own skins to the Workshop.
- Fixed network issue which meant corpses and dropped items were showing at their spawn positions on the client.
- Thrown spears return to the belt bar when picked up.
- AntiCheat bans are livetweeted to @rusthackreport.
The roadsign jacket is now an ‘over’ type of armor. This means you can wear it on top of shirts and stuff the same as the metal chestplate. I also added the pants version of this item.
Bone & Wood Armor
These two sets of armor are also designed to be worn on top of existing clothing. When I implement armor damage I’m going to make them block much more damage, but fall off after absorbing a few hits.
- Wood armor good protection vs melee & explosions.
- Bone armor has *very* good protection against most melee and is vulnerable to blunt & stabbing attacks.
- Both sets are default blueprints :),
- Some of the armor slightly hovers over the player - will be mitigated next week.
- Some of the icons (namely bone) are WIP.
I spent all week working on the same things I was working on last week (mining system, new event) but didn’t get it done in time, so I switched gears and implemented some items the artists completed this week. Hopefully you all enjoy them!
More work on some mining features and events, hopefully I’ll have something to show by this time next week!
I fixed a bunch of issues with our movement code. There’s still a lot of improvements left to do, so don’t expect it to be perfect just yet. You should encounter significantly less annoying stuff though.
- Improved landing, fall damage, grounded and sliding detection.
- Made jump and sprint blocking after landing feel more natural.
- Fixed various situations that could cause jumping weirdness.
- Added climbing state between grounded and sliding where the player can hold its position and move sideways but not upwards.
- Changed slope speed adjustment to mostly affect sprinting speed.
- Fixed wobbliness/bounciness when trying to move up a step that’s slightly too high to move up.
- Improved ground normal detection.
Holding ALT no longer looks back, instead you can hold ALT and move your mouse to freely look around. Right now the head rotation isn’t networked to other clients because I didn’t want to mess with networking on patch day, but I’m hoping to add that next week.
The natural next step for this system is to add support for head tracking. There’s a thing called TrackIR that allows you to (relatively) cheaply add head tracking to a traditional monitor setup. The way it works is that you turn or move your head and the in-game head of your character moves by an amplified amount. This means you can look around in the game without losing sight of your screen in real life. I added a first implementation of this that allows you to rotate your character’s head around two axes in the same manner the ALT freelook works.
The current implementation should be looked at as a first, somewhat limited prototype of what can be done. I’m planning to add more functionality in the future, and ideally I’d love to get to a point where you have the full 6 degrees of freedom via head tracking plus left and right leaning of the entire body via keyboard or pedals. Obviously all of this could also be used to add VR support and I’m planning to look into ways to render the entire body of your character instead of just floating arms, but that’ll be a bigger task to tackle so don’t sit around waiting for it.
- Removed some deprecated and very limiting foundation placement checks which are now handled by the new systems we added over the last few weeks.
- Fixed decal shader on DX9.
More of the same. I didn’t check off a lot of the bugs on my TODO list this week because the movement fixes and head tracking require a lot of testing and tweaking and therefore ate up most of my time, which means I still got a full list of fixes and polishing to get to. I also postponed the conversion of more decals to the new system because they were broken for some people and I have to verify that the fixes I added this week had the intended effect.
I’ve been debugging crashes on an older Macbook Air sporting an Intel HD 3000. Took me two days to setup and test - while working on other stuff simultaneously - OSX, Windows over Parallels (which ruled out driver issues) and then Windows 7 x64, which was the crash target. The internal SSD drive was too small for our project so that caused a few problems, and running it from an external HDD is slooooow.
To any of you having problems running on this hardware, please be assured that we’re very close to be able to do something about it. I was able to track it down to Intel GPU driver issues and are preparing a minimal reproducible project to send over to both Intel and Unity for analysis. In the meantime, however, I was able to get it working and *might* be able to work around the bug. No promises, however.
At the beginning of this week I got Helk up my ass regarding terrible PVT performance on high-end hardware. We’re talking drops from 100 Fps to 25 Fps. This was most definitely not supposed to happen, but it did. Sorry guys.
I was able to replicate it and figure out why it was happening. It seems to happen mostly on NVIDIA GPUs, I believe because of their particular cache architecture. In case you’re wondering why I’ve been pressing so much for this feature, here’s an example of the performance improvement when it does work properly:
Rendering the terrain only, at 2326x909, maxxed anisotropy and parallax on a NVIDIA GT 750M and considering only the terrain timings: top is running without PVT at ~34 Fps; middle is running the current implementation (with a few improvements already) at ~104 Fps; and bottom is running with the new compressed cache at ~238 Fps. Of course, these values will change depending on a lot of factors, like how much of the terrain you see, how far away you are from it, your aniso and parallax and, finally, everything else that’s running on the GPU.
Based on these numbers, I believe it was important to halt everything and get this fixed as soon as possible. Not only because it was affecting your gaming, but also because it’s a really important optimization that affects most mid-range machines and high-end machines running at high resolutions or maxed anisotropy and parallax mapping. I’m also taking the opportunity to add real-time PVT cache compression to reduce VRAM footprint by around 65% while simultaneously improving performance. If these latest changes don’t fix it for everybody, I will fold. Also, I’m not done with them so they’re not on any branch yet.
Regardless I took the opportunity to add back this option:
Just keep it disabled for now, unless it actually improves your framerate. It’s also now disabled by default.
Yesterday I started revamping the standard Rust shaders to add support for masked tinting and microfiber-based fuzz, aimed at cloth and fabric materials:
Also fixed a water2 bug on NVIDIA cards running in DX9, showing some weird random lines instead of an opaque layer of water.
I hope all this technical mumbo-jumbo doesn’t turn you off. I’ll go straight to the point next time. Speaking of that, next week I'll finish the outstanding tasks and get back on track.
This week was pretty busy for me as I wanted to get a new weapon done in time for the patch. I managed to finish the view-model animations for the two-handed sword, as well as the third-person anims.
I also fixed some minor bugs with some of the third-person animations, such as when the player deployed/reloaded/attacked with a weapon, his eyes would close shut.
Next week I’m gonna continue work on the melee weapons that Dan made. I still have three more to go (a hacksaw, a two-handed cleaver, and a two-handed baseball/mace thing).
I finally finished up these armor sets! They took a bit more planning and tweaking than I was expecting, so I haven’t finished the wood helmet, but that extra work means we should be able to use their pieces in fun ways in the future. Since the bones, boards, ropes, and other pieces are modular, we can use them to make lots of unique variants using the same pieces, and they could even be combined into armor using multiple different materials.
Next Week: A lot of the work I have to do over the next week (and possibly more, depending upon what types of snags I do or don’t run into) has to do with retrofitting clothing to fit the new female model.
I made some good progression on the crossbow! It’s been super fun to work on this asset and hopefully it will be super fun to play with! In any case, ya’ll want pictures so here is what it looks like.
I’ve yet to do the LODs for this and check it out in Unity.
Next week I’ve got some bug fixes to check out with the female character, as there's some seam issues going on with the skinning. I’ll also be finalising some heads and different skins, so hopefully I’ll be able to get those checked out asap.
I'm working on the finishing touches of the sphere monument update. It’s sitting in its own branch at the moment, as it will be waiting for a wipe to be released to you.
Next week the tier two furnace should be ready to be implemented.
I finished up work on the large fractioning tower! They should look pretty intimidating with the flames shooting out the top.
I also started back work on cars. Like I said a few weeks ago, I think cars will work best if they are constructed modularly, so I started with some tests. Basically you start with a default chassis, and I'm only showing one size here, but I imagine we would have a few different sizes both smaller and larger. The chassis then has a two column grid on it that you use to place the modular parts. The parts I have here all take up 2x2 grid spaces but I think with the larger chassis we could have flat beds that take up a 2x4 or larger area. Even with three or four different pieces you can already see the variety of vehicles that you can get. When there are more parts to pick from and different size chassis to build on I think everyone’s vehicle will feel like their own.
Next week I’ll continue with more car stuff. If this modular system looks like it could work with the rest of the team, I want to design a library of modular parts.
This week I finally finished up the first real pass at weather-influenced ambience. You’ll hear rain hitting the tree leaves when it rains in the forest, and some days will be windier than others. Things like wind volume levels and the thresholds that we switch between different weather sounds will probably need a bit of tuning still, but I’m pretty happy with where we’re at so far.
Most of the ambient sounds have also been cleaned up and improved now.
I made all of the ambient sounds stream from your hard drive instead of loading into RAM, which should lower memory consumption a bit (and is probably necessary with the number of these sounds we have now).
Next week the way we’re managing that all of the ambient sounds and determining which ones to play right now was great when we only had a few variations, but it kind of blows now that we’ve added so many more sounds. I think there’s a lot of room for improvement now, so I’m going to spend a little time next week refactoring the ambience manager.
Once that refactor is done I’ll be spending the rest of the week running through my list of sounds that need adjustments and polish. There’s a few things in there that I’ve been wanting to change for a while now so I’m excited to finally have some time to do this!
This week has been a bit interesting. This is due in no small part to the fact that I received a reminder of one of the perils of being a concept artist, that being that you can work on something for a week and polish it to a shine only to have it scrapped completely. I had about 11 different stackable hats/headgear pieces (based on the ones from last week) but it looks like none of them are going to see the light of day, and I'm not sure I can show them off in the devblog if they're just flat out not gonna make the cut.
I can however show you some more beaten-up satellites.
Aaaaaaand buttloads of icons. They're all pretty much variants of existing icons so I've put them all in one smaller image for your viewing pleasure.
Next week I'll be looking into making new beds. These aren't to replace the sleeping bags and they're not quite a retooling of the Legacy beds, but something that could prove very useful in game.
If you want to follow this project you can sign up to the mailing list.
We'll only update you about this project, we won't spam you about other stuff or sell your email address.