A series of trends in controllers for competitive fighting games have led to the popularity of low-profile, all-button (aka "stickless" or "Hit Box-style") controllers using mechanical keyboard switches. The vanguard here includes the MPress and the Snack Box Micro. However, there are a few issues with the available offerings. The only commercial options with support for PS4 — the de facto standard tournament platform for the fighting game community (FGC) — are only sold in the US, making them especially expensive for players elsewhere in the world. DIY offerings typically require 3D-printed chassis, which typically require a harsh tradeoff between expense and aesthetics (not to mention durability), and struggle to provide PS4 compatibility due to Sony's authentication system for controllers.
Thus, having recently become frustrated with the weight and bulk of my very high-quality Blunderbuss controller, I decided to design a DIY solution for such low-profile controllers, that would support commercial PS4 encoders ("PCBs" in FGC lingo). I aim to design something that uses off-the-shelf parts where possible, and laser-cut panels wherever necessary, thus making the controller viable and affordable to build anywhere in the world.
I am making customisation a big focus. I want a builder to be able to adapt the design to their needs without requiring much technical skill with e.g. CAD software.
My "pet technology" for this project is using T-slot aluminium extrusions (aka 8020 profiles, structural framing, and a dozen other names) to form the frame of the controller. I am not the first person to think of this; in fact I first became aware of the existence of these extrusions because someone posted a controller build that uses them on Reddit. The extrusions are cheap, incredibly sturdy (sim racing enthusiasts build their rigs out of the stuff), and easy to come by anywhere in the world, and many sellers will cut them to size for you. They are also fantastic for modifications — a builder can easily decide to attach, say, a carry handle to the frame, or even something silly like a cup holder. (The FGC is not above employing antics to get inside an opponent's head.)
Essentially, what I am doing with this project is figuring out a lot of the "hairy bits", such as:
- How small the chassis can get without making it impossible to fit the encoder PCB;
- How to mount the panel that holds the keyboard switches at the correct height offset relative to the top of the chassis using a mechanism that is easy to customise; or
- How to use only universally-available fasteners to hold everything together.
With the hairy bits figured out, a builder can grab the files I produce, make simple changes (extend a line from 20 to 40 mm using drawing software), and have a full bill of materials plus all the laser-cutting drawings they need to order and build a custom controller to their specifications.