So, Stadia finally met its demise last week. After an unsuccessful launch with general apathy from the audience, there's an argument that the platform was doomed right from the very beginning - despite Google itself investing a gigantic amount of cash into the hardware and various publishing deals. Perhaps the writing was on the wall when Stadia shut down its own development studio, while the logical evolution of the platform - as a 'white label' cloud system that could be used by publishers - wasn't really compatible with the Linux/Vulkan foundation. Features failed to materialise (remember being able to access a game state by sharing a link?) while there were always doubts about the 10.7 teraflops of GPU power. If graphics performance was so high, how come Xbox One X at 6TF produced superior results? Was that GPU actually virtualised over two users? I'd love to know for sure.
The thing is, as many have stated, Stadia had much to commend it in terms of actually delivering a decent enough streaming experience - markedly better than xCloud in terms of quality. Regardless, the system did not attract much interest and it's time to face realities about the actual viability of cloud-based platforms in general.
Let's start with the technological arguments first. Put simply, if the platform holder is not responsible for the quality of the experience from beginning to end, there will always be problems. I've tried Stadia and xCloud on a 30Mbps fibre connection - and they simply did not work. I upgraded to Starlink and that didn't work either. Only now, with the arrival of fibre-to-the-premises (FTTP) and gigabit internet does it actually work for me. My experience won't be the case for everyone, but the point is, there's no way to 'debug' it and make it work. There are other technological limitations too in a bandwidth-constrained scenario - if I'm playing Stadia and someone else in the home decides to download a game or even just watch Netflix, there are going to be issues.
Alongside quality issues along those lines, Stadia made fundamental miscalculation - that selling games at full price on a patently unproven platform would work. There's a reason why xCloud and GeForce Now 'work' where Stadia does not: full-price purchases persist outside of the cloud on those platforms, via mature platforms where we can be reasonably confident that they won't disappear. Stadia always faced a confidence issue here, not least owing to Google's track record in abandoning its own services. Thankfully, refunds will happen in this case, but a fully cloud-based system is always going to have trouble selling full-price games that do not tangible 'exist'. And so, another one bites the dust and hopefully lessons have been learned.
- 00:00:00 Introduction
- 00:00:48 News 01: Google Stadia is shutting down
- 00:20:00 News 02: PS5 gets a 6nm dieshrink!
- 00:23:31 News 03: XeSS debuts, Arc A750/A770 priced
- 00:37:13 News 04: New Ryzen 7000 CPUs reviewed
- 00:57:49 News 05: Sony PC game announcements, leaks
- 01:02:18 DF Content Discussion: DLSS 3 reactions and follow-up
- 01:07:48 DF Supporter Q1: Is latency a big issue with DLSS 3? Could a future version have unnoticeable lag?
- 01:10:22 DF Supporter Q2: Could tech similar to DLSS 3 come to consoles?
- 01:11:40 DF Supporter Q3: Would a slower version of DLSS 3 be able to run on Ampere cards?
- 01:13:48 DF Supporter Q4: Is vendor-agnostic frame generation tech similar to DLSS 3 possible?
- 01:16:40 DF Supporter Q5: How does DLSS 3 cope with RT effects?
- 01:19:14 DF Supporter Q6: Would DLSS 2 Quality mode work better as a base for DLSS 3 than Performance mode?
- 01:22:15 DF Supporter Q7: What do you expect from AMD’s new graphics cards?
- 01:28:47 DF Supporter Q8: Thoughts on EVGA leaving the GPU market?
- 01:32:45 DF Supporter Q9: Is 8K excessive? Shouldn't we focus on visual features and not pixel count?
- 01:35:13 DF Supporter Q10: Could PCs move to unified memory instead of two discrete memory pools?
While the death of Stadia is the centre of discussion this week, there's still more to cover. It looks like the PS5's processor has indeed been shrunk from 7nm to 6nm, producing a markedly smaller chip but with only a conservative power reduction. It has allowed for a cheaper internal revision, but it seems we're still some way off a 'PS5 Slim'. We're looking to source a CFI-1200 unit and if any UK retail sources can help us buy a unit, please drop us a line at info [at] digitalfoundry [dot] org.
Meanwhile, Intel has announced pricing for its Arc graphics cards and while I can't comment - yet - on the quality of the GPUs themselves, what I can share is that the pricing here is so keen, Intel must surely be selling at a significant loss or at breakeven in a very optimistic scenario. Bearing in mind that Sony has put up the price of its PS5 with 16GB of GDDR5 paired with a 6nm 260mm2 processor, Intel is selling its own GPUs, based on a 392mm2 processor on the same fabrication node for $150 less... with a compelling software package added on for free, including Modern Warfare 2! It's a proper bargain then - in theory. Look out for the full Digital Foundry review soon.
There's masses of supporter questions this week, kicking off with a DLSS 3 special, where we tackle a range of queries in the wake of the exclusive first look we released last week. Then it's onto 'standard' supporter Q+A, covering off what we want/expect from AMD's RDNA 3 products and whether 8K is excessive (spoilers: of course it is!). This part of the show is defined by our backers at the Digital Foundry Supporter Program, where you can join our amazing community, talk to the team directly (we do actually interact with our Discord community) and get access to a bunch of early access videos and bonus materials. Join us!.