Game development isn’t a science. It’s a creative process. It’s important that during the process we highlight what we’re doing wrong and what we’re doing right.

Making mistakes is healthy. Not learning from them isn’t.

1. Devblogs

We started doing devblogs about 40 weeks ago. Previous to this we would only really post on this site when we were releasing an update. The idea comes from a talk that Robin Walker made at SDC where he highlighted that doing all this development work on a game then preventing your work as a list of changes is short-changing yourself. It makes sense to properly highlight the work we’re doing. This not only presents it to the community, but it also presents it to the development team too. Kind of like an open review of our time spent.

Our devblogs got away from us a bit. It got so that we were working harder to impress the readers of the devblogs than the players of the game. This isn’t sustainable because some weeks we are going to just be fixing bugs or enhancing performance, which isn’t fun to read – so we’d feel like we’re wasting our time doing those things. So now we’ve reined in the blogs slightly. We’re trying to be more even handed, only posting stuff that is actually in-game and only posting concept art if we have a good reason (instead of filling space).

2. Cheats

Cheats were a huge problem in the old version. It’s fair to say that it killed Rust for a lot of people. We wrote some simple checks and banned a shitload of people, VAC banned a shitload of people, but we ended up spending more time writing anti-cheat and trying to trap cheats than actually developing the game. This is obviously not a feasible situation.

So we went about seeking a company to out-source this stuff to. In the newer version we opted to use Easy Anti Cheat. We chose EAC because it’s the less obtrusive solution we’ve found, the guys really love what they’re doing, and the company is small enough to give a shit about us.

EAC has worked out good for us so far. They have done a lot of work for us, like making OSX and Linux versions of their client. We’ve had some short-term problems with their infrastructure dropping clients from servers but they’ve always been on hand to fix them.

It’s naive to think that we can completely remove cheating from a PC game. The idea is to make it difficult and unattractive.

3. Early Access

Our launch on Early Access overwhelmed us. It’s awesome that we sold a shitload of copies and got a shitload of players, but it’s not sustainable. We would much rather have slow burned and been a popular game in a year, than had the huge initial spike. The money we made is great but we didn’t particularly need the money for development, and once we sold a million copies a lot of people expected that we’d hire 50 people to make the game in 2 months and the fire them all. This is probably a miss-understanding of what Early Access is (it’s not kickstarter) and why we’re all Game Developers (the fun of developing a game, not the fun of writing pay-cheques).

If I could have done it again we would have sold it exclusively on our website for a year before launching on the Steam Store. The only downside to this is that we’d have to handle all the billing, and all the chargebacks and all the credit card fraud. So maybe we’d convince Valve to handle the billing while only being discoverable via our website or something.

4. The Reboot

We decided to scrap the old codebase. I made an extended post on Reddit on my feelings about how this has gone, whether or not rebooting was a good idea. I’ll highlight some of the points here.

Our general feeling is that although the reboot was inevitable the focus should have been on getting the reboot to be a cleaner 1:1 copy of the legacy version. This is where we were failing for a long time, getting sidetracked ‘developing’ the game before the foundations were laid. We should have reached the same stability, gameplay and performance levels of legacy before branching out into other things.

This is a lesson we have learned, and hopefully you guys have seen the difference in the last couple of months.

Grade: C-

We made progress, but it’s superficial. The game looks better than it ever has, but what we gained in graphics we have lost in gameplay – and that’s unforgivable.

Next Year

In the coming months we’re going to be clawing that back. We’re going to be re-introducing a lot of the legacy features such as Radtowns, Blueprints, Workbenches, Roads, Researching. We’re going to look at bringing back the legacy map and making it possible for user created maps. There’s even been talk of bringing zombies back – but we won’t make any promises on that.

Once everyone’s happy with the gameplay, stability and performance we’ll start exploring the mindmap.


Comments are closed.