The next big update to the game is coming along nicely and we're progressing through the last few big things to do list before Steam Early Access (more on that later). The focus for this update has been sorting some big old bugs and generally making the game feel less like a pre-alpha and a more like an alpha. We also found some time to include some extra little features at the same time.
At time of writing, I estimate that 0.0.8 is about 60% complete. That puts us on target for this version to come out mid-October, somewhat quicker than usual. As always I’ll post the patch notes for it as soon as I have a draft of them.
Path-Finding System Greatly Improved
This was by far the biggest task for this version. The way AI navigated in Atomic Society needed a big overhaul. Workers tended to take strange routes, get confused, bug out and walk through walls, etc. Small things but they happened constantly. It made everything in the game feel just that little bit annoying.
However, Nick has been busy in his underground lair and managed to drastically improve the way the AI picks its routes. This makes the whole game feel better. Watching people live out their lives looks more natural and realistic.
For example, before this version every building was shaped like a box to the AI regardless of the actual detailed shape of the model. Now the AI can accurately detect the shape of the building can even walk in and around its grounds and side-step obstacles. They’re much more aware of what’s going on around them, including their fellow citizens. They also no longer ghost through each other but step around one another, etc.
The same improvements apply to controlling the Town Leader, which is now more logical and consistent. And as an added bonus, all these improvements have given the game a small frame-rate boost too due to efficient coding.
The 0.0.8 pathfinding system doesn’t drastically change the game, but it does make everything in it better to play, so it was definitely worth doing.
Another overdue problem with the game involved placing buildings. Often players would be unable to put a building exactly where they wanted for no obvious reason. There was a reason of course, a bug, and how the game handled the radius of objects.
Thanks to Nick's tinkering with pathfinding, we were also able to improve how the game scans the landscape. It should now have a much better idea of where obstacles are and the shape of the ground. We’ve also improved how the terrain deforms when you place a building so when this task is finished, you should be able to build towns more or less however you want to lay them out.
We’ve added a new feature this month that a lot of people have been asking for. In 0.0.8 you will be able to tell your scavengers to focus on one particular resource. No more getting angry at them as they fill their bags with scrap metal if you needed lumber.
To do this you just open your storehouse menu and tick the resource you want most of all. The scavengers then obey your orders and use common sense to work out where the nearest ruin containing that item is.
Aside from curing some player frustration, this also adds a little gameplay as you are now in control of your resources to a degree. You can send them off to get one thing while you gather another as the leader. Though I might need to do some re-balancing when this is finished, the game might be a bit too easy now you're not fighting the stupidity of your scavengers anymore!
Early Access Release
We’ve been thinking about bringing forward the Early Access launch. We were aiming for early 2018 but Atomic Society is already quite fun in its own limited and slightly buggy way, so why not make it easier for people to get the game?
I first thought about bringing the release forward while I was browsing Steam. I like diving into Early Access and seeing what weird and interesting games are out there that don't get the attention of the front page. Seeing all these games, it struck me how Atomic Society is already a lot more polished than a the majority of games out there and they seem to be doing okay, by my standard of "okay".
It doesn't matter we've still got a lot to add. For those who enjoy Early Access games (like me) it's fun supporting a game and seeing how it grows up, each patch can be fun to check out. And bad reviews don’t phase me anymore. I’ve seen games with negative reviews turn it around with a few good updates because Steam focuses on recent reviews. I'm not going to be scared away by a red thumb symbol.
The other reason we were holding off was to get the marketing right and have lots of content in the game before the mainstream press found us. But the mainstream press already found us 2 versions ago against our plans, and we survived that. Plus, I don't don’t care about lots of marketing these days. I’d be perfectly happy if things just carried on as they are now but with the added ease of Steam.
Anyway, long story short: Expect an announcement about our Early Access release in the not too distant future. We'll need to get 0.0.8 out first, and then do a mini-version to prepare the game for Steam, but after that we’ll be ready to start announcing dates. It will be this year (so in the next 3.5 months) unless something unexpected happens.
0.0.8 brings a new map for you to make towns on. This map is called Valley and is a forest biome. It’s a verdant woodland set on the steep sides of a valley around a huge lake. You can see right across the map from one side to the other, at least you can if you settle your town high up.
The aim of this map was to make the most vertical landscape in the game so far, and to make a nice relaxing location! After months making deserts and frozen tundras it’s nice to be back in peaceful surroundings.
I think the unique, crater-like layout of this map will lead to some interesting town shapes.
New Gameplay Building: Luxury Tower
A ultimate type of shelter has been added to 0.0.8. This is a huge building, the biggest you can make so far. It’s essentially a lot of luxury homes welded together into a battered tower. “Luxury” is a state of mind in the post-apocalyptic world after all.
It’s designed to an end-game building, for those times when you have a lot of resources and you want to quickly provide housing for all the remaining homeless citizens without having to build a ton of houses, so its very expensive to build. It also looks cool.
New Attitude to Game Development
This is the personal section of this dev blog (skip ahead if that's not your thing). I’ve had a much needed attitude adjustment lately. In short: I've stopped worrying about how well the game sells and it’s amazing. I feel like a new man.
I used to worry every single day about how the game was being received, how it was selling, what I needed to do to make it more popular and liked. Long-time readers of these dev blogs might’ve noticed how stressed I sounded at times. My stress was getting worse the more copies we sold. I was always watching what "rival" games were up to, what indie devs were panicking about, worrying about what people were saying on our forum, worrying if nobody was speaking on the forum, worrying if nobody made Youtube videos of our game, being scared when people did make videos. And so on and so on.
At long last I got it into my skull that worrying about the future and trying to control it hasn’t made me even one tiny bit happier. So I’ve resided myself to accept whatever happens. If AS sells lots, all right. If 3 people buy it on Steam and then refund it, all right. Either is fine. And as soon as I truly accepted that my stress disappeared.
I guess I’m becoming… Optimistic?! Not about success, just optimistic that things will be fine however they pan out. I’ve stopped chasing the latest news, I’ve unplugged from Twitter, I let things happen as they happen. And because of this I’ve probably had my happiest few weeks of work on the game since the day we started.
New Convertible Ruin: Mansion
Nani has spent a lot of time adding a big new ruin to the game in 0.0.8. You will soon be able to convert (if you find it) an abandoned Victorian mansion into a post-apocalyptic home or storehouse. It should make a fine centrepiece of your society. It’s the biggest ruin in the game yet, and also the most detailed.
Making a convertible ruin is a lot of work. Like all ruins, Nani has to sketch out how the ruin should look, based on my concept. She then makes a fully intact version of the building in the art program we use called Blender. Players never get to see this version of the building, it looks as the building would in its prime, before the bombs fell. When we’re happy with how that looks, she smashes it up and adds decades of neglect!
This would be enough for ordinary ruins, but for ruins the player can convert, Nani then needs to work out how a bunch of post-apocalyptic engineers with limited supplies might try and restore it. And do all that with an extremely tight polygon count and range of textures. However it's all come together and our collection of ruins grows larger.
More Improvements in 0.0.8
That’s not all coming to 0.0.8. We’ve also been able to upgrade the game to the latest version of Unity which has a few minor benefits for players and will probably boost performance on older machines. The upgrade went smoothly for a change though Adam had a few problems on his 5 year old laptop with in-built graphics (only the finest hardware at Far Road Games).
Having watched people on Youtube get caught out by the new repair mechanic in the game, we’ve added in more useful information to explain why buildings are disappearing.
Adam has made it so players can now type in a number of loot to destroy, which is very handy if your storehouses are overflowing.
We've made several useful UI improvements again and fixed (*checks notes*) 7 important bugs at time of writing.
There’s more to come in 0.0.8 though. I’d like to try and squeeze in adding the ability for the Town Leader to help maintain buildings and possibly add little mounds of debris when a building collapses so you can salvage something back. We'll see how times goes.
A Look Ahead...
By the time next month's dev blog rolls around I predict that 0.0.8 will have just been released, or at least be very close to going out the door. We'll then talk about an Early Access release date and begin integrating the game with Steam. There are 1-2 features I want in the game before Steam, nothing too serious, just stuff like rebindable controls. It should be a fun (and stress-free) end to the year!
See you next month.
Version 0.0.7 Released!
It turned out the game needed six weeks of intensive tweaking and bug-fixing to finish the new update! But we did it and I'm pleased to say 0.0.7 is now available for those who enjoy seeing the game grow up step by step.
For those who haven’t seen what’s new yet, here’s a quick glance at some of the changes:
It’s a pretty big patch so you can read the full patch notes for it here if you like.
Version 0.0.7 certainly doesn't make the game feel like a finished product yet, so don't expect that. It is, however, the next big step on the journey.
Under (Release) Pressure
Now that version is out - and a quick hotfix patch we quickly released to stop all citizens spawning as female - I’m able to unwind a little.
I get so stressed in the run-up to release. Being a perfectionist and releasing a pre-alpha makes for a bad brain combo. I must have proof-read the tutorial 50 times for example and still managed to miss a serious typo!
It reminds me of when I used to do a bit of acting, that pre-show tension, when you're waiting for the curtain to go up before a paying audience. You're as prepared as you can be, but you have to get it right because people have expectations. It's even worse when you know that Youtubers could be streaming your update to thousands!
Adding to the tension, we still had a serious crash in the game with less than 24 hours to go until I'd definitively told people the version would be out. It was a crash that occurred when saving, destroying people's towns. Over many hours we pinpointed the problem to cops arresting themselves for being slavers and doctors euthanizing themselves! Fortunately, coder Nick, powered by pure caffeine, came to the rescue and was able to sort it out in the nick (get it?!) of time.
And then, feeling immensely excited and stressed, I released the new build and... It was all quite anticlimactic. You're so hyped and you've worked so hard, but most of the world doesn't even know the game exists yet. It can take days or weeks even for the news to spread. But that's a different marketing problem...
Smaller and Faster Updates From Now On
Because of this stress, and because most of the big core tasks in Atomic Society are more or less done (just about 700 medium-sized tasks left to do), we should be able to release more frequent updates from now on. This will be good for players, who get fixes and new things to tinker with more regularly, and good for us as they'll be less to test and fix each time.
So rather than 10 features per version over 3 months, I’m hoping for about 3-4 features every 6 weeks roughly. I wasn’t keen on this approach initially, as I thought smaller versions would be less exciting, and more hassle, but I might be changing my views.
We’ll see how it goes with the upcoming version 0.0.8 and whether it proves beneficial for the game and getting noticed.
Design Philosophy Ver. 240
Smaller versions will also fit neatly with the latest patch to my game design philosophy.
Basically, I’m now forcing myself not to plan ahead. I have dev scars now, I know how much work new features are. I can't make others work on something that isn't 100% what the game needs.
And you only know what the game needs when a new version is finished and you can actually play it without bugs or missing bits for the first time. Then you find out something I was certain the game needed a month ago might not be required anymore.
So from now on, rather than adding 10 new things to a version and hoping they’ll still be what the game needs in 3 months, I’m designing just 2-3 at a time, then we'll release and review. Only then do I decide what to add next.
This is a scary method for me. I like to plan and know where we’re headed. Now I don’t know where we're going beyond the imminent future. Now the game is dictating to me what it needs.
Where will it end up? How will I know when Atomic Society is “finished”? It reminds me of the old quote that “a poem is never finished, it is only abandoned”. Perhaps even the best games, that have years of work poured into them, are forced out into the world. One day you just have to move on.
For now, I think my satisfaction levels with AS, and the pressing need of not going broke, will be our judge, not any plans I invent. Right now I'm probably 28% satisfied with Atomic Society. I would like to get that figure to 40% before we even hit Steam.
If it’s taken us 2.5 years to get to 28% satisfaction, there's still a (far) road ahead...
Plotting Version 0.0.8
What I am sure about, and player comments seem to be confirming, is that the next version must be getting things we've already added to work more reliably.
Therefore, 0.0.8 is primarily going to be a bug-fixing version. Not the most exciting thing to announce, but it should make a huge difference as all those little glitches add up and I'm tired of seeing players frustrated by the same few things.
The core focus will therefore be path-finding, the routes and paths the AI takes. It needs a big overhaul that we've been putting off for ages. The second focus will be squashing some long-standing bugs, like not being able to build where you want to. And thirdly, we’ll do something about scavengers, as people want more control over what loot is coming in.
We've known about these problems forever, but only now does it feel we've got some breathing space to address them. As soon as that’s all done, we’ll release and then I'll play the game and work out where we go next...
The End of Summer Madness
The spark of interest the game received back in May, when a dozen random Youtubers examined it out of the blue, took about 3 months to die down and even now sales are a better than they used to be. The game has been publicly available for about 15 months, and 70% of all copies sold were in the last 3 months!
You can imagine what hope that gave us. Hope holds us together as a (mostly) unpaid team, as is the case no doubt for a lot of indie devs. Hope gets us through long days in our day jobs and long nights on AS. This summer helped a lot.
The problem with even a little sniff of success can be addictive. It's tempting to start to working for sales figures, not for yourself. It's never enough. Sales increased by 10%. That's becomes the new baseline to feel "good". Keep feeding the numbers. It's the same with social media. If I wanted to work for the bottom line, there are much better jobs to get than making games, so I'm trying my best to ignore it all.
Some new Youtube videos on version 0.0.7 by different people have appeared already, you can find the first parts of their series here, here, and here.
Patch notes for the 0.0.7b hotfix we've just released.
My post on how salvaging has changed in 0.0.7.
Future Prediction Time
I know I said I need to stop planning ahead, but it's okay to make mistakes in a dev blog!
At the end of the last blog I guessed that by the time I wrote this one, we’d all be exhausted after working overtime on version 0.0.7, and waiting to see if the version engaged the public. Aside from the fact we were delayed by 3 weeks getting the version out, that’s proven to be perfectly accurate.
So, by the time September’s blog is out, I won’t be insane enough to guess version 0.0.8 will be ready... But I’d like it to be about 70% finished, which would be nice. I'm looking forward to being able to polish stuff rather than invent new stuff for a change.
Anyway, thanks to all the readers of these blogs. They seem to get a pretty good reception.
Hope you enjoy version 0.0.7 and I’ll see you in the next edition…
The Road to Release
Every month we release a personal and honest look at the making of Atomic Society.