5 ms·

I wonder what went wrong. But at the same time this feels inevitable. The best AAA studios on earth still strongly depend on purpose built hardware to deliver

I wonder what went wrong. But at the same time this feels inevitable. The best AAA studios on earth still strongly depend on purpose built hardware to deliver their experiences. The broad spectrum of smart media devices is a hellscape compared to Xbox, PS, Nintendo and PC. There is no way you are going to ship something like GTA6 to a Vizio tv without doing rendering on cloud GPUs. Netflix would also have to compete with platforms like Steam, which are already expanding into broader ecosystems.

disqard 2y ago
Yeah, I think controlling the h/w platform is necessary (but not sufficient) for success here. Probably also needed is patience, which (for example) was missing with Stadia (Alphabet) and with Luna (Amazon).
- bob1029 2y ago
  > patience This is the key with game development. Getting a concept to work is a non-linear path at best. Most business leaders do not understand this process and will ruin it the instant they attempt to observe and quantify it. I've seen it almost ruin boring-ass B2B SaaS. Fun isn't even an objective here and that's the hard part. Most can't cope with the chaos of this artform when it's done without restraint. I think you have to be moderately insane to run a AAA studio effectively.
- root_axis 2y ago
  I don't think patience would have made a difference, it's simply that the economics of cloud gaming are untenable. The type of people who reliably spend money on AAA games don't want the degraded experience that cloud networking introduces. These days they also understand that they can potentially lose access to their cloud catalog if they are banned, the internet goes down, or if the publisher decides to withdraw titles. By comparison, buying a PS5 or a Nintendo every 5 years amortizes nicely when the gameplay experience is objectively superior and you have a sense of security that you own the game. Even worse for the business model is the ongoing costs of running burstable GPU clusters at scale - clusters that use a lot more power and are driven by components that are a lot more expensive while also being more prone to failure. The situation is even worse today with transformers putting extreme strains on commercial GPU supply for the foreseeable future. All this before content licensing and production costs which is its own formidable challenge as netflix already knows. Cloud gaming has no hope of success.

disqard 2y ago

Yeah, I think controlling the h/w platform is necessary (but not sufficient) for success here. Probably also needed is patience, which (for example) was missing with Stadia (Alphabet) and with Luna (Amazon).

bob1029 2y ago
> patience This is the key with game development. Getting a concept to work is a non-linear path at best. Most business leaders do not understand this process and will ruin it the instant they attempt to observe and quantify it. I've seen it almost ruin boring-ass B2B SaaS. Fun isn't even an objective here and that's the hard part. Most can't cope with the chaos of this artform when it's done without restraint. I think you have to be moderately insane to run a AAA studio effectively.
root_axis 2y ago
I don't think patience would have made a difference, it's simply that the economics of cloud gaming are untenable. The type of people who reliably spend money on AAA games don't want the degraded experience that cloud networking introduces. These days they also understand that they can potentially lose access to their cloud catalog if they are banned, the internet goes down, or if the publisher decides to withdraw titles. By comparison, buying a PS5 or a Nintendo every 5 years amortizes nicely when the gameplay experience is objectively superior and you have a sense of security that you own the game. Even worse for the business model is the ongoing costs of running burstable GPU clusters at scale - clusters that use a lot more power and are driven by components that are a lot more expensive while also being more prone to failure. The situation is even worse today with transformers putting extreme strains on commercial GPU supply for the foreseeable future. All this before content licensing and production costs which is its own formidable challenge as netflix already knows. Cloud gaming has no hope of success.