Hey guys!
Before shapez 2 releases on August 15 (that’s less than a week!) I wanted to do a personal post about the history of shapez 2 and what to expect during early access. We’ve been super transparent in the development process so far, and thus I thought it would be a great opportunity!
This post is a bit longer, but I tried to include a bunch of images and interesting insights! If you'd like to skip to the Early Access expectations part, click here.
TL;DR
- Shapez 2 releases August 15, 13:00 CEST
- It should be polished & enjoyable and basically bug-free
- No roadmap; Instead we will survey after the release what you want us to actually work on
- First update might take a few months
Why shapez 2?
Back in early 2021, I was struggling to come up with new ideas for content and updates in shapez 1. The game loop felt very closed, all new ideas evolved around adding more complex shape mechanics at the end of the game, which would only benefit a fraction of the playerbase. Popular mods like shapez industries did exactly that, but they also heavily modified the early game already, which I didn’t want to change (as I felt it was already in a good place - except for blueprints being a bit late).
Performance was a frequent complaint, but there was nothing left to optimize. Shapez 1 is written in JavaScript, so it’s literally just a website. There is no support for efficient rendering, memory management or proper multithreading and the language itself is at least a magnitude slower than C++ or C#. If you look at it from this perspective, it’s actually quite insane how well the game runs up to a certain scale.
These were just a few of the factors that led me to the conclusion to focus on something new, instead of adding more content and features to the first game. Now that shapez had shown huge promise (which, by the way, was totally unexpected!) my goal was to now do it “properly” with a real engine and build a base that could be expanded for many years.
Due to personal reasons it took me a bit longer to actually start, but back in July 2022 I wrote the first line of code for shapez 2, without any clue how it’d eventually turn out.
Researching
As during that time a lot of sequels couldn’t deliver on the expectations, I was very worried and careful to make sure we build a great sequel that is both liked by existing players and potential new players. While this sometimes meant going for less experimental changes, I think we found a great compromise by now.
To determine what defines a “shapez” game, I did a survey in the shapez 1 Discord server. I wanted to avoid changing the essential things that made shapez stand out in the first place:
The results showed me what to keep and expand on, as well which mechanics we could probably exchange or get rid of entirely. For example, only 18% of players voted for the complex shape mechanics that are introduced in level 20. I never really liked the mechanic in the first place, and so we decided to remove it and add more intuitive mechanics instead.
Additionally, I wanted to find out what new features could be exciting for a sequel:
These surveys immensely helped us to determine what to focus on. If you look closely, you’ll see that we’ve implemented almost everything from the list already 🙂
The Discord
Starting from day 1, I shared a lot of development progress in the shapez 1 discord. I wanted to try making a perfect game (spoiler: quite impossible) and I knew I could only do it with the help of the community.
The early community input helped a lot, and during the whole development process, many ideas were discussed and then later discarded. I never held something back, instead as soon as there was an idea or prototype (be it just a paint sketch) I tried to get feedback as soon as possible.
We even let the community decide the art style for shapez 2! I initially didn’t want to make another space-themed factory game, so we proposed a bunch of concepts:
The initial winner was D, which later evolved into the space theme
Unfortunately developing a sequel isn’t as easy as just doing a survey and implementing everything. We had to make a lot of tough choices, and sometimes even decide against the initial community response. As the discord is somewhat of an echo chamber of players that like shapez enough to be interested in the early development of a sequel, we always had to be careful to not just listen to this vocal minority and actually get different opinions and perspectives.
Some of these topics evolved in very heated discussions with more than 1000 comments. To name a few:
- Straight Stacker vs Bent Stacker (spoiler: both made it in)
- Art Style (CPU Theme vs Space Theme)
- Floating shapes (aka Level 20 Logo Shape from shapez 1)
- Cutters always cutting from north to south, regardless of orientation
- RGB vs RYB color scheme (now a setting)
- The name of the new ‘Diamond’ shape
If you are interested, all of these are still available in the #discussions channel on the shapez 2 discord!
In total, we did around a hundred surveys on the discord so far. I’m very happy with how it turned out, and I really like this way of developing a game together with the community instead of presenting a finished product. This way we can be sure that starting from day 1, we are moving in the right direction!
This also means despite launching into early access, the game has already been hugely impacted and improved by community feedback. The main reason we are launching into early access is to show our commitment to keep updating and improving the game even further!
The move to 3D
I knew from the beginning that I wanted shapez 2 to be 3D, simply because I saw a lot of people not trying the game because of the simple graphics. So besides improving the gameplay, I also wanted to package the game nicer to make it appeal to a wider audience.
shapez 2 very early on in development
It also added the opportunity to make all buildings open and animated, something that sets shapez 2 apart from all other factory games: because all operations are ‘logical’, we can visualize how a shape is cut, stacked, rotated and so on. Other games only have black box machines that receive inputs and spit out the result and we believe this is something truly new to the genre (which we are very proud of!).
However, while 3D was the most prominent addition in the beginning, over the course of time so many new gameplay features and mechanics have been added that instead of calling it “shapez 1 but in 3D” it feels more like shapez 1 was a prototype for shapez 2, and shapez 2 is now the real game.
While switching, we also decided to switch to a new engine (Unity) which allowed us to use a more performant language and rendering pipeline.
Growing the Team
Whereas shapez 1 was developed by me alone back then, I realized that shapez 2 was too ambitious to develop alone. I wouldn’t be able to focus on all areas with enough attention if I would have to work on the development, game design, marketing and art.
Over the course of the two years, the team grew to 8 full time members, which was only possible because we got funding from the German government, which we are very grateful for! Without this team, shapez 2 wouldn’t have been possible and I’m very proud of what the team has achieved together.
Iterating and iterating
Since you always just see the end product, it’s easy to forget how many iterations sometimes are required to get something “right”. There are so many mechanics we tried that didn’t work out or required a major rework: Fluid packing & simulation, Trains, Platform building, Research, Shape crates, and many more.
For example, we implemented around 10 (!) different research systems until we found the current one, which we now quite like (until #11 arrives). You can read more about this journey here:
https://steamcommunity.com/games/2162800/announcements/detail/4103413768091304193
Game Design Challenges
One of the biggest challenges was (and still is) balancing the puzzle part with the logistics part. Our surveys show that the player base is split, with some players preferring more complicated shape puzzles whereas others are focused on building and scaling huge factories.
(Left = Puzzle , Right = Logistics)
Finding the balance has been very hard, and is one of the reasons we have added so many customization, difficulty and game modes - to make sure everybody can enjoy the game how they want to.
While talking to experienced players all day in discord, it’s easy to forget how difficult the game can be for first time players. We tried to get better insights by running a lot of playtests (we captured around 450 hours of playtest footage) with different audiences and optimizing the early game experience, but finding the perfect ‘default’ difficulty was very, well, difficult. Adding more complex game modes like the insane or hexagonal mode definitely helped though, because now you can simply adjust the difficulty to your liking.
What to expect from Early Access
We’ve spent a lot of time polishing the game already and are really happy how it turned out. We believe it’s in a great state and is meant to be played. With 4 game modes, there’s plenty of content to be explored. There should be at least 40 hours of content but it’s easy to get over a hundred hours or more, as our playtesters have proven already.
As we have over 1000 patrons playing the game on a regular basis already, we were able to fix a lot of bugs and issues already. There are no major issues known to us at the time of writing, except for the usual driver-related crashes (which we can’t really do anything about - except recommending a different backend like DirectX 12 / Vulkan). We also spent a lot of time on the tutorial and UI/UX, making the belt placement really easy to use and everything feel smooth (which we’ve already gotten a lot of praise for!).
That being said, there are still a bunch of things that can be further improved – we don’t stop when something is just ‘good’ but always try to go the extra mile. For example, space belt placement is missing a few quality of life tweaks like automatic tunnel placement, unfinished space pipe visuals and so on. However, all of these should be minor in the grand scheme of things – it’s more our perfectionism speaking here.
Space pipes don't look like pipes yet
We would also like to explore a few more experimental game modes, if there is demand for it. For example having to explore the map, or introducing (as an alternate game mode!) energy or power requirements.
Performance
As mentioned above, shapez 1 started to bog down after a while when scaling to huge factories. We tried to improve this for shapez 2 and are happy with the results so far.
For shapez 1, your factories would start to lag starting at around 5,000 - 10,000 buildings, and would be almost unplayable at 40,000 buildings, depending on your setup.
In shapez 2, everything should currently run very smooth until 100,000 buildings, pretty smooth until 250,000 buildings and depending on your setup somewhat smooth with 500,000 buildings (30 fps on a higher end setup). Players have built factories with up to 1,25 million buildings and above, but they encountered significant lag. For reference, the average player has around 40,000 buildings when completing the game, so we already support 12x more than that.
Recorded on Alpha 23. Contains placeholder visuals (mostly the trains)
However, keep in mind that you might encounter lag if you are GPU bound (which is independent of the simulation). You might want to experiment with lower graphic settings then if you are experiencing low FPS for smaller factories. There are also some shaders and effects that are unnecessarily complex, causing the game to use slightly more resources than necessary. While it is not too noticeable on mid and high range setups, there’s still a few things we can enhance for the lower end setups.
While we are happy with the performance - this is just the beginning. We believe we can easily make it run twice as fast, as we still have some pending optimizations like multi-threading or certain optimizations for the space belts. However, since buildings are free to place and copy/paste over and over again, there will always be a limit to how much the game can handle. The limit may be 100 times greater than what it is now, but you'll still be able to hit that limit if you really want.
Future roadmap
You might be wondering - if we are satisfied with the game already, what will we be working on next?This is exactly the reason we are releasing the game now in Early Access - we need your input!
I don’t like launching with a roadmap immediately, especially not for early access. If the roadmap is already predetermined, what impact do players actually have?
Instead, we will be doing surveys after the release to find out what you want us to work on next! Additionally, we have a suggestions portal where you can leave feedback and suggestions, or vote on existing ones!
Early Access Duration
We haven’t set a duration for early access because it could be years in theory. This is not because the game is not ready but because we want to show our commitment to keep updating and improving the game even further.
How long the game stays in early access also depends on the success. Our dream goal would be to keep updating the game for many years while it finances the development. However, as the game industry is unpredictable, we don’t want to promise something we can’t keep.
That being said - please do not purchase the game if you don’t like it in its current state and just hope that it gets better in the future. Quoting Steam: “If you are not excited to play this game in its current state, then you should wait to see if the game progresses further in development.”
First / Future Update Timing
Finishing everything up for the launch has been a challenge, so we're taking it somewhat easy for a little bit. With so many big factory game releases and updates coming later this year, we wanted to make sure to hit this deadline. We will of course be monitoring the release and early reception – if a major issue pops up, we'll be sure to hotfix it.
Once the release has cooled down a bit, we plan to work on a few fundamental things that didn’t make it into the launch build due to time constraints. We want to refactor a few things that we had to hack in and also prepare the game to make it easier to expand and add new features (like multi threading). Of course, we don't know yet what you want us to work on, so we'll have to see!
However, this does mean it might take a while until the first major update. We're sure certain other factory games will keep you busy in the meantime though ;)
We're looking to use the following structure, depending on how things go:
- Major Updates: Every few months with big changes that justify picking up the game again and fit our quality standards (localized, well tested, etc.)
- Beta Branches: Irregular updates anyone can opt into, with previews of upcoming updates that might not be finalized yet
- Closed Betas: Irregular (but more frequent) builds for our patrons to try out bleeding edge features (untested and possibly broken)
Thank you for reading! We are super hyped for the release, and hope to see you there! I hope you will be able to enjoy the game as much as we do.
~Tobias