I led the user experience design of Steam Deck from conception through development.

Valve

Valve has a history of trying to bring the experience and low barrier of entry of console gaming to the PC faithful. It’s not always been a smooth ride and has resulted in some spectacular failures.

Steam Deck is a handheld gaming computer that lets PC gamers take their Steam experience anywhere they want to go. My role was to take our hardware plan and build the experience that would make it a reality.

What is PC gaming?

One of the biggest questions I found myself asking at Valve is what exactly is PC gaming really? And how is it different from console gaming? There was general consensus that we wanted an ‘appliance-like’ interface, consistent hardware, and improvements to make playing on the go more seamless. But I was warned that we were not making a console, it should still feel like PC gaming.

Over time, I learned, when people were talking about PC gaming, what they really meant was the experience of using Steam.

Users wanted to be able to ‘own’ their games, have backwards compatibility and access to their existing library. They wanted features like Steam chat, deep customization for their inputs and settings, and the ability to use mods. They also wanted to have access to an open hardware experience. Peripherals, external storage, mice, keyboards, monitors should just work. Some folks might even want to leave our UX entirely and use the desktop or install windows.

Finding the flow

The principles I used throughout the design process were focused on drastically reducing the barrier to entry. This is a PC that you don’t put together, manage drivers, or install an operating system on.

Our north star was that you should be able to pick-up Steam Deck, find a game–owned or in the store–and play it almost immediately.

The first mechanic I prototyped and tested was navigating games. I took lessons from TV streaming apps, and other game launchers. Using a directional pad or joystick as the primary input is a very different challenge than most mobile or desktop apps. The Steam deck has just about every kind of input–touch, joystick, trackpads, you can connect a keyboard and mouse, and it even has a gyro. We built a hardware mule so we could quickly prototype the system navigation and test it across all those modalities. By delivering on the core experience first, we could then build the rest of the Steam ecosystem around it.

Building an OS

We worked hard to make it easy to chat, see notifications, setup inputs, and manage connected devices without impeding the core path to games. One of the biggest challenges we worked through was building all the features needed for settings and services that are normally handled by Windows or Mac OS. Through testing I identified what worked well for multi-modal interaction and began to establish a design system.

We built a system of components that could quickly snap together to build features.

I also worked hand in hand with engineering to bring these components to life, often writing the css and functionality myself. We were able to build out settings pages in a piece-meal fashion as the development happened. As the team went further in the process we also began to establish layouts and workflows to handle some of the most common challenges.

One specific issue was converting Steam’s litany of right-click context menus into patterns that were usable without a mouse. We did this programmatically by designing a drop in pattern that converted the context menus into layered pop-ups the user could easily navigate using touch, joystick or any other input.

Accessibility

Beyond building an interface that would work across sever different input modalities, I also made accessibility a critical feature. We set out to vastly improve color contrast, tab navigation, and also make the device work across most of the 27 languages Steam supports.

One major accomplishment was integrating localization and language support into the onscreen keyboard. Most applications and software can rely on the operating system to provide an IME (allows for input kanji and other characters that don’t easily fit on a qwerty layout) and onscreen keyboards for each language. Our experience lives entirely inside a browser so we had to do this work from scratch. I built an adaptable keyboard that could handle multiple layouts, use all the Steam deck –including touch–and support Japanese and Chinese languages.

Over a million Steam decks were pre-ordered and delivered to customers with very positive reviews of the new interface.

Next project: Blue Origin