Debugger

Why Game Streaming Will Finally Work

Services like OnLive have tried and failed. Stadia and xCloud should succeed.

The future of gaming won’t require a powerful console sitting under your TV. In fact, it’s boxless, wireless, and streaming from a magic server, miles and miles away.

At least that’s the vision spun up by some of the world’s most important companies. Google is preparing to launch its Stadia streaming gaming platform later this year, Microsoft has xCloud, Ubisoft’s got UPlay+, and so on. The race is on to become the dominant cloud console — but it’s actually a race that started quite some time ago.

A service called OnLive offered streaming games back in 2012 that connected players to a virtual desktop running on a remote server. The company lasted a few years and came to offer a dedicated device for connecting to its service, but it eventually folded because of lack of customer interest. OnLive was only ever available in a few markets, and it had data centers installed directly in the heart of those cities.

There is also Sony’s PlayStation Now service, which launched in 2014 and allows gamers to stream an array of titles to their console or PC. Nvidia’s GeForce Now is another, though it’s still in beta.

These earlier efforts never quite took off, in part because they didn’t work well for a wide range of people. PlayStation Now, for example, still has issues on slower internet connections. Infrastructure has been another major problem: Streaming services rely on servers around the world, and proximity to a user’s location can be a major issue.

So what makes the modern wave of game streaming any more promising? Two things: better technology and the sheer scale of the companies entering the field. Here’s how it all works, and why the next generation may finally fulfill the promise of these earlier services.

Faster, better, stronger

One of the problems with early game streaming services was that they were, in essence, streaming video of a game directly from a powerful computer and nothing more. They relied on streaming technologies designed for one-way video streams of content to your machine — think Netflix, for example. But streaming video games requires a fast, two-way connection — the game needs to make its way from the server to your screen, and any inputs you make must then be communicated back to the server. Then it streams back the reaction to what you just did, and the cycle continues.

The internet has come a long way since these first services were imagined. Stadia, for example, uses a standard called WebRTC, which is largely known for making video calls faster and reducing latency by connecting people’s computers directly to one another, rather than routing them through a centralized server. You’ll still need a good internet connection for this to work properly (more on that in a second), but the technology is better equipped to mitigate latency issues than previous efforts like OnLive were.

By using WebRTC along with other standards like QUIC and BBR, which further reduce latency and connection congestion, Stadia is able to make an end run around a lot of the problems encountered by early services, because the video stream and controller commands aren’t intricately linked together.

When your connection gets flaky — as even the best broadband definitely will sometimes — the stream will still feel responsive. Instead of glitching around the screen like in traditional multiplayer games — itself a major problem that can make games unplayable, as you miss your targets and those with better connections destroy you — the stream will simply drop video quality to compensate, becoming pixelated or distorted as when YouTube has buffering issues. Because the congestion is being managed by these new protocols and your controller commands are processed separately from the video stream, your character still responds, and as the connection recovers, the video quality ramps back up again.

The real magic, however — and what makes me so optimistic about Stadia — is that no additional software will be required. Google promises that you’ll be able to play on any device that has a Chrome browser, with a single click. If it works — and early reports indicate it very well could — Stadia will completely transform how games are experienced on every level.

And the gaming industry could use some disruption. When you buy a game today, you’re inevitably faced with a 40-gig update and a lengthy install process before you can even start playing — and that’s assuming your drivers are all up to date and compatible with the software.

Game streaming will remove all of these pain points. By lowering the barrier, these services can expand the appeal of video games to the millions of people out there who are gaming curious, but who never made the move to buy a console, or who find themselves unsure of what they’d like to play. It’s easy to imagine that a future streaming service might offer the first 30 minutes of a game for free as a one-click taste, like a traditional demo on steroids.

Of course, this is all assuming your internet connection is up to the task.

The cloud is already in your city

(Google Cloud’s reach as of 2019. Credit: Google)

The problem that killed every streaming service before now was latency — the time between issuing a command like “move forward” and the remote game actually responding.

With OnLive, you might have been looking at the game on your screen, but it was rendered far away in a data center. Actually doing something involving the controller’s buttons meant sending a signal up to that remote server, which then needed to be processed and sent back your way. The process only took milliseconds, but the delay was perceivable as lag to the player.

The thing is, OnLive actually worked pretty well if you happened to have a data center physically located in the city you lived in. The further away your data center was, the more OnLive became a laggy mess. The company’s challenge was building out data centers full of powerful servers in hundreds of cities across the United States.

(Microsoft’s existing Azure data centers. Credit: Microsoft)

Google and Microsoft don’t have this problem, because they’ve spent a decade building hundreds of data centers in every corner of the world for their cloud platforms. This infrastructure is what allows services like Google Drive to work.

Because these companies want services like Windows Update and Google.com to load as fast as possible, they have dedicated content delivery networks. These data centers are as close as possible to the majority of their users, hidden in cities and suburbs to shave crucial milliseconds off loading times. Previous services like OnLive simply lacked the massive cloud infrastructure Google and Microsoft have spent billions to develop over the years.

All that said, you may have noticed significant gaps in those connectivity maps above, like New Zealand floating out there without any connections. Game streaming services clearly aren’t going to reach everyone immediately, particularly those in countries where connection speeds are slower, or where there aren’t enough people to justify a dedicated data center.

That’s precisely why Microsoft is still building Xbox hardware, with a new console coming in 2020, and why most of us are looking at cloud gaming the wrong way. It’s the next step in the evolution of gaming, which will make games accessible to millions of people for the first time without buying consoles — but it won’t replace hardware for a long time. They’ll co-exist for years.

Still, it’s clear now that streaming technology has completely transformed. All that remains to be seen is whether anyone actually wants to pay for it.

Fascinated by how code and design is shaping the world. I write about the why behind tech news. UX Manager @ Shopify. https://twitter.com/ow

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store