"Google Stadia is not a product that exists because people want it"

As a consumer and a developer I don't really want Stadia to succeed because of the shift it represents in terms of ownership of media, as well as hack-ability of games. In general I want to have games locally so I can do whatever I want with them.

However, as an industry observer, it's hard to imagine that something like Stadia doesn't have the potential to be a huge success if they really could deliver on promise. With all the people watching people play games on Youtube and Twitch, it's hard to imagine you couldn't convert at least some small percentage of that massive audience into players if the barrier to entry was only a click of a button.

Stadia is a staple of what the open web should be for games. Just open a web browser and the game is available. Much in the same way that Photoshop should be available on the web.

Unfortunately it had to be Google that had to get their oar into this space and therefore it looks like just a toy demo for them.

I do wish engineers at Google should instead contribute to the technologies that advance the open web instead of vanishing Google products.

With such latency we would have 0 online shooters. But we have a lot of them.

I used to work on (the second generation) cloud gaming. It's very difficult to be profitable. Most people would think that the challenge is latency, streaming quality. But the real challenge is cost management. The cost can be easily as high as $2 per hour. At this rate, the monthly cost will buy you a decent computer or a console. It doesn't make sense.

Plus, to make cloud gaming work, you have to own everything. From the video encoding hardware, operating system, virtualization, game content, distribution channel, cloud ...

For example, we tried to hack windows to support existing windows games. It was very difficult. Windows isn't a multi-user system. There is no proper user isolation. We have to monitor the hard drive to see what files are touched by games, and try to resolve conflicts caused by different users using the same machine. It's very hacky. An easier way is just discarding the VM and refreshing the hard drive, but it will result in long loading time of several minutes, very high cost and poor experience.

And then there is the game store problem. Game publishers won't share revenue. You basically sell the games at the same price and also charge users for the cloud gaming platform for a compromised experience. doesn't make sense.

There is only one company that owns the whole stack, Microsoft. They own the OS and APIs (so that includes all the drivers), virtualization, games (XBox store), data centers (Azure).

A few years ago, YC invested in a cloud app streaming company (they kinda used the same technology we used.). I was surprised. After working on it for a few years, I would not invest if I were an investor.

Google's stadia seems to be smarter. As it tries to avoid some of the dead-ends we went to, especially trying to hack windows to make it cloud gaming friendly. Google chose to use Linux and develop games tailored for cloud gaming. I think that's a better choice, but will face content issues.

I read the article, but it feels like "don't build anything until you are sure users want X". The problem is that nobody knows what users actually want, especially when you build something new. And the market will keep throwing ideas in several directions until one day, an idea sticks and becomes successful.

Like, no gamer in pre-Wii times "wanted to play hours standing with a motion controller". Nintendo released the Wii and it was a huge success (for a while at least).

Maybe the author's point is that Stadia is not doing anything "new", and this may be a good point. But it's still early days, and Google has deep pockets. It's a little too early to discard Stadia.