Swider wasn’t the only Stadia developer blindsided by Google’s late September announcement that the streaming gaming service would be shutting down next January. Game makers who talked to Ars (and some who shared their surprise on social media) all said they had no indication of Google’s shutdown plans before the public announcement.
“During correspondence [with Google], we are exchanging emails—nothing showed us it could be the end of Stadia,” Swider said.
What a shitshow.
I do not get how Alphabet can stay successful the way they keep canceling everything. Microsoft had to go through many iterations of their Xbox consoles before gaining success but it was that commitment that made them able to get developers to stick around not to mention people buying their products. Google meanwhile has such a horrendous record that I can’t see why anyone would want to risk dealing with them, developers and customers alike.
The original Xbox was a success, considering it sold more units than established rivals such as Nintendo, unless the benchmark for success is “selling as much as the PS2” (which benefitted from huge brand name recognition, out-of-the-box DVD playback and the huge PS1 back catalog), which was always an unrealistic standard.
Why was the original Xbox a success? Because Microsoft went in with an “in it to win it” attitude. They sold it at a loss, the product offered features that other consoles didn’t (such as an HDD and integrated ethernet), it had a huge game catalog at launch (I was honestly surprised when a magazine was presenting the launch of the original Xbox and the launch catalog spanned pages) and they had SEGA and several talented first-party studios make AAA games for it.
Compare and contrast with something like Stadia, Windows Phone, or Windows 8, where arrogant companies launched their ecosystem-less platform on the world expecting the world (that is, developers and customers) to axiomatically give them a place in a mature market. Build it and they will come! No, it doesn’t work that way when talking about ecosystems! Even Android risked a similar fate in its first years if it wasn’t for the carriers pushing it as if their bloatware depended on it (which it did).
Don’t forget the 360/PS3 era proved custom CPU designs were too complicated to program for. Now they’re all using practically stock x86/Tegra chips..
dark2,
That’s true about custom CPUs. Developers prefer to write once and run everywhere. For better or worse this discourages unique programming models. A custom CPU might offer a ton of compute power, but if developers only bother to write generic CPUs algorithms without optimizing for the special processors, then their potential gets wasted, which was the case for the PS3 for a long time. Using mostly standardized components makes portability so much easier and it’s become a market advantage even if it makes consoles less interesting.
To be fair, I can’t blame Sony for Cell, IBM duped them into thinking the 7 SPEs could function as a CPU and as a GPU equally well. Then Sony saw the final chip and realised that, while true in theory, it’s wasn’t true in practice, since there was no way the 7 SPEs could match the performance of the Xbox 360 GPU no matter how well you programmed them, So, a traditional GPU was added (sourced from Nvidia). So, Sony was stuck with this weird architecture, where the system had one good-enough GPU, one very weak PowerPC core, and 7 SPEs that weren’t good at functioning as a CPU or as a GPU. But it wasn’t the same situation as the PS2 where the architecture was intentionally bad so they could claim “128 bits of powerrr!”.
But you are right, their choice to go with custom designs backfired badly for them, and that’s why nobody has tried it ever since.
BTW everyone expected this day would come when Stadia launched with a mostly barren ecosystem of games, but I am also surprised by how fast they wound it down and how little notification they gave developers. That’s some late-2000s-Microsoft developer backstabbing right there (see: Silverlight). Remember how nobody took Microsoft’s new platforms seriously after the fiasco that was Silverlight and how everyone stuck to win32/win64 afterwards? This is Google’s problem too now: Everyone is confident with developing for the Android API but doesn’t trust Google on anything else.