Dabe Alan

Valve’s challenge isn’t to keep PC hardware open, but to close it down

Valve’s challenge isn’t to keep PC hardware open, but to close it down

The video game industry has been buzzing with the news of the “Steam Box,” a sort of PC-based console for playing video games on your television, using a controller. One company touted Valve’s investment in its platform, and showed off a prototype of a small, modular computer. Valve is reported to be looking at other prototypes, and may allow many different models of the “Steam Box” to be released. Even the name is more of a media creation than an official designation.

“We’ll come out with our own and we’ll sell it to consumers by ourselves. That’ll be a Linux box, [and] if you want to install Windows you can,” Valve’s Gabe Newell told Polygon. “We’re not going to make it hard. This is not some locked box by any stretch of the imagination. We also think that a controller that has higher precision and lower latency is another interesting thing to have.”

That’s Valve’s box, however. Other models will be released, with different specifications, and this is where things begin to get interesting. What the hell makes something a Steam Box, and just how important is closing, or at least tightening, the design of a gaming PC?

You likely already own a Steam Box

This isn’t a new console, remember. If you take your gaming PC, hook it up to your television, and play games with a controller, you have your Steam Box. The question is whether hardcore PC gamers are interested in leaving their monitors for their television, and giving up a mouse and keyboard for a controller. The Steam Box is challenged by the idea that many of us who love PC games are more than comfortable with a smaller monitor on a desk, with a mouse and keyboard used to play.

But Valve isn’t interested in converting you. If you’re already buying Steam games and you’re happy with your setup, they’re happy. The idea is to grow the market for Steam games, and to head into the living room. That’s why so many comparisons are being made to consoles; gamers who want to put something under their television in the living room want something that’s easy to use, easy to set up, and “just works.” They want to buy a system like they buy a console: Go to the store, point to the one they want, plug it in, and play some games.

Fewer choices, more gamers

This is why the idea of a modular system is so compelling. Get rid of model numbers of CPUs and video cards, and just explain performance in a way that’s easy to understand. Give components a number that roughly equates to their performance. Let’s say stores stock Steam Box shells, and you can choose between options that are listed on a scale from 1 to 10.

You buy the system for $400, get an “8” in the GPU for $300, pick a “7” in the CPU for $250, and you have your $950 console. As new parts come out prices go down, and the numbers go up. Make the case easy to open, and the parts easy to replace. This is a simplified description of a way to fix a complex problem, but you get the idea. Fewer choices, better communication, and simple metrics for performance and value are all needed to improve the consumer experience when it comes to mainstream PC hardware. RAM is already easy to understand: More is better. Speed is an issue, but you can fix that by setting a minimum speed for inclusion in the box.


This will never happen, as it would require Nvidia and AMD to get rid of their fancy combinations of letters and numbers and mazes of different amounts of RAM and clock speeds in their unending parade of video cards, and the same thing would need to happen for CPUs. Buying performance parts for gaming PCs is incredibly consumer hostile at the moment; it takes an invested individual to really understand what you’re getting with your hardware, and why it’s important. The system described above may also force hardware manufacturers to change how components connect to the motherboard in order to make your average consumer comfortable with lifting a flap, removing their old GPU, and replacing it.

You’re going to argue that you know how to do these things already, but you’re not the target market. Neither am I. We’re already a part of this ecosystem. It’s a matter of grabbing people who aren’t comfortable building their own computers, while still keeping an upgrade path available.

The problem is that Valve is going with Linux for the “official” version of the hardware, which means limited game support, at least at launch. While the Humble Bundle has been doing much to get developers to support both Linux and OSX, the vast majority of games sold on Steam are Windows only.

Communicating the fact that this hardware is a PC that doesn’t play most PC games is a rough task, as is asking consumers to purchase a copy of Windows and install it before playing their favorite games. Other companies may release similar PCs with Windows pre-installed, but Valve is going to be the leader here; what they do sets the tone for everyone else.

Valve has a tricky balancing act on its hands. It needs to keep all the open aspects of the PC software ecosystem while closing the hardware aspects of the device in order to lure a new audience. Buying, upgrading, and understanding the systems has to be dead simple. Complex stats and options need to removed in favor of easy to understand metrics for power and game performance. The trick isn’t to give consumers more choices, it’s all about giving gamers fewer options, but to have each one be worthwhile.

I can already hear the comments screaming, but this is the reason Apple does so well selling hardware. You go into the store, and you’re given a few options between different models, each one aimed at a different consumer. It’s easy to weigh the pros and cons of each one and make a buying decision.

Going into a big box retailer or to a company’s web page and selecting a PC, on the other hand, is a mess. Dozens of different models, options, incomprehensible jargon, and uninspired designs rule the day. Making a purchasing decision, and then feeling comfortable about it, is not an easy thing.

So what should Valve do?

There is plenty we don’t know about the hardware that will eventually be released for this initiative, or the ultimate role Linux will play in Valve’s own system. But the more choices and options given to consumers, the more the Steam Box will simply look like a standard PC without a monitor. PC sales are down 21 percent, and mirroring the problems of the PC market for what amounts to a console launch is a mistake.

The trick, and this may be an impossible task for a console based on PC hardware, is to close things down. Offer fewer choices, with each level bringing specific value. Once again, you need to allow customers to come into a store, point to the system they want, and understand their options quickly and simply.

Valve needs to create a rigid set of rules and standards for third-parties who want to release a system under the “Steam Box” banner. As long as the system does, or can, run Windows, the software and ecosystem will be open. The trick is to turn the hardware options into a walled garden, with Valve having ultimate say over what does and doesn’t count as a Steam Box. Anything less will be a missed opportunity to expand the world of PC gaming.