I’ve released a game on PC game pass (and Steam), and I can tell you that it’s painful for the devs too, before the players ever run into these issues.
One thing that was especially frustrating is that there is no way to automate the process of uploading a build. You have to drag and drop giant files (which you first had to get hold of from the build server in a usual setup). And click buttons and stuff. And wait a lot between steps.
When we mentioned the desire to automate this, so we could automatically deploy eg nightly builds, MS sounded like that was an interesting idea they hadn’t heard of before. WTF.And stuff like that missing will automatically mean that the quality of the build on that platform is worse. No nightly build, but only build on demand requiring human work time and frustration means no frequent testing by QA on the platform, until they absolutely have to.
When we mentioned the desire to automate this, so we could automatically deploy eg nightly builds, MS sounded like that was an interesting idea they hadn’t heard of before. WTF.
When you receive a request for a feature that you know would be good, but for whatever reason you can’t implement it (perhaps there are other things you need to pay more attention to, or perhaps there’s an idiot manager establishing dumb priorities), giving the response you received is one of the least anger-inducing ones. It’s likely they’ve been repeatedly asked to make the process less painful, but whoever is in charge of managing devs doesn’t care.
It’s doubly absurd considering Microsoft owns one of the biggest build and deployment automation pipelines as part of their Azure offerings. Most of it is aimed at Azure, but so much of the Xbox backend is just Azure under the hood anyway. Azure Pipelines should have had integrations for this on day one.
That’s embarrassing. How long have CI/CD pipelines been a thing?
And yet people somehow keep telling themselves that Microsoft has the best developer tools of all platforms (coincidentally mostly people who have never worked with other platforms).
It’s baffling that a software company isn’t able to give best software experience in their own operating system and software stack.
That would assume their primary goal is giving a great experience. At best that’s going to be secondary. More likely tertiary or less.
Primary goal is to move everyone to windows store, then lock it down like apple store.
Literally why Valve started working on Proton.
Microsoft fails so miserably at this stuff it would be kind of funny if it weren’t so gratingly infuriating.
Don’t forget collecting user data for ads and selling to other companies
This is the best summary I could come up with:
That’s really one of the main issues with the Xbox app — the consistency of the content delivery experience is sub-par compared to, well, Steam, which right now remains totally and absolutely the benchmark for these types of services on PC.
The Xbox app also frequently sports separated multiplayer pools that won’t connect to Steam, owing to the extra hoops developers often have to do to get that working correctly.
And sure, Microsoft can’t wave a magic wand and force developers to prioritize their platform over Steam — assuming they would even have the resources to do so — but for the end user, all they see is an inferior product offering with PC Game Pass branding on it.
I’m also endlessly annoyed that there’s no cloud save upload visibility in the app too, which has created version mismatch scenarios for me moving between my ASUS ROG Ally, my main gaming PC, and my Xbox.
I’m not suggesting Microsoft should, or even could, simply dump its PC Game Pass operation into Battle.net, but it might be worth exploring how a different content delivery system, bypassing the aging mechanisms of the Windows 8 era, might solve a lot of these problems.
Microsoft needs to explore ways of achieving this absolutely desirable pro-consumer goal of “any game you own, on any platform you want, with all of your progress intact” — all without increasing the burden on its developer partners in the process.
The original article contains 1,754 words, the summary contains 241 words. Saved 86%. I’m a bot and I’m open source!
