1793

No major triple-A studio is going to sign an exclusive deal with Stadia, for a price that would be realistic for either party. They cost too much to make, and Stadia just doesn’t have the player base to make you your money back, let alone make a big profit.

Any bonafide exclusive triple-A games early on, would have to come from Google itself (or a second party arrangement with Google paying for development)… but Google made no real investment in development infrastructure. They have one confirmed studio, and it’s a tiny indie outfit that has never shipped a game, was acquired less than a month ago, and whose first game can’t be exclusive due to pre-existing contracts. Christ, Amazon doesn’t even have an announced service yet, but we already know they have at least two large games in internal development.

I assume these 10 “timed exclusives” Google announced, will be about 80% indie games, with maybe 1-2 “double-A” games thrown in there, since they know a year of only indie exclusives isn’t going to fly with their users.

If they had an exclusive AAA game on the horizon for this year, I assume it would have been at least announced before the November launch. And I’m pretty sure Jade Raymond herself has already admitted that it’s going to be “a few years” before you see real Stadia exclusive games (which is completely insane, by the way).

Like I said before - based on their investment so far, Google should have just made Stadia a service like Shadow: a means to play PC games you already own. There’d be no complaints about the lack of exclusives, you’d immediately have access to the wide world of PC games to buy, and all that would really matter is the core tech - which is the only thing that is mostly in place and working right now anyway.

But of course, that isn’t as financially lucrative as being an actual platform holder, so they went the cloud console route - but without making the big investments in content that every other platform holder makes, so far…

1794

The only way to get a 60fps video output stream is on chromecast ultra right? If you were playing on Windows on a PC monitor inside Chrome you’d have to disable G-Sync and set it to 60hz for proper smoothness since 120/144/165 Hz wouldn’t be exactly 60 or a multiple of 60?

1795

Incorrect.

1796

image
image1466×1436 396 KB

From the GDC state of the industry report

1797

How much money are they losing right now? Wouldn’t trying to support “PC games you already own” cost them even more?

I consider this an experiment. If Google is willing to eat their losses, let them do whatever they want. My epeen is not threatened.

1798

You might consider it an experiment, but they are selling it as a service and platform, haha.

I have no idea what their economics look like right now - probably not too bad yet, because their actual investment in Stadia has been so pathetic. They acqui-hired one tiny indie outfit, and are slowly hiring for an internal studio, two months out from launch. Beyond that, they tossed a few bags of cash out for timed exclusive indies like Gylt.

My point is that if Stadia were just selling you access to a high-end PC to play your PC library on for $10 a month, the expectations of them change. Suddenly people don’t really expect exclusives, because you’re not a “platform” - you’re a facilitator. They don’t expect robust services and a steady stream of ports for your platform either.

Yeah, you lose some of the gimmicks like YouTube integration, but you don’t even necessarily have to give up on your storefront - Stadia could still have it’s own storefront built into the service, competing alongside Steam and others.

This is what Stadia is acting like it is right now, because it’s certainly not acting like a serious competing platform. Someone actually did the legwork on Reddit, and figured out that Stadia right now, actually has fewer games than the ORIGINAL XBOX had at this same point in it’s lifespan (and the PS1 too). That’s despite coming out at a time when the industry itself was smaller. And in addition to having the bigger library, it also launched with a bonafide exclusive system-seller, in Halo CE.

If Google don’t want to make the risky investment in their own platform’s success by buying studios and really going “all in”, why should anyone else? Go the more modest, less risky (and thus less lucrative) route, and just be a service that lets you play PC games without a PC, and suddenly the value proposition for Stadia changes.

1799

The idea doesn’t start making any more sense, no matter how often it’s repeated.

  • The service would be totally at the mercy of Microsoft, who are still just as rutheless about exploiting Windows as they always were. Hell, it’s less than six months from when they changed the licensing model of Windows Server in a way that made it totally uneconomical to run on AWS and GCP, but exempted Azure.

    They’d do the same with client Windows in a heartbeat if a service like this being run by one of their main competitors gained any traction. (Assuming that using those editions of Windows in a cloud setting is not already categorically forbidden).

  • The abuse potential is massive. It would take like 5 minutes from the service opening to somebody starting to use GPUs for mining crypto-currency 24/7 rather than playing games. (Or running ML models, or whatever could profitably use a lot of GPU/CPU). A flat rate of $10 would have to be making assumptions about typical utilization being lower than 100%.

    Things like Stadia and xCloud are protected from this since users can’t run arbitrary software. Your model no longer has that advantage, so the machines need to be billed based on actual usage just like cloud servers are, which I think would be a very hard sell to consumers.

  • It’d be a total commodity market, with no way for Google to differentiate themselves from anyone else who tries this, no real competitive advantage, and customers who are presumably very price-sensitive. The margins would be awful

    The idea of making up for those margins by running a PC game store assumes that anyone would rather buy a PC game from Google than from Steam. We know very well that wouldn’t happen. The EGS launch showed just how hostile PC gamers are to using non-Steam stores.

  • The addressable market for the “rent a gaming PC from the cloud” is far too small. Shadow already has exactly this service. They appear to have 70k users. For a project like this to get green-lit at Google, they’d need to show a plan for how this becomes a billion dollar business. Probably more, given how low the margins would likely be.

  • There’s no technical upside. All the important parts of the stack would be controlled by third parties and there would be decades of legacy to support, so shipping interesting new features would be hard or impossible. What you end up with is something that’s strictly worse than a PC on some axes (latency, graphics quality) but with no compensating advantages.

    Why would any gamers be excited about that? And yes, maybe some of those potential features are just gimmicks, and others will be snake-oil like “negative latency”. But there must be others that are actually compelling. Maybe Stadia never gets to those compelling features given their trajectory. But launching a service where they can’t be implemented in the first place is just playing to lose.

There are two companies that could plausibly run the service you suggest, since they have some unique way of mitigating some of the flaws.

  • Valve, since they can restrict it to just playing Steam games, and consumers will be happy. I suspect they’d end up having to do it on Linux+Proton rather than Windows though.
  • Microsoft, since they control the platform. But it’s hard to see them doing both this and xCloud.
1800

Some interesting and valid points in there, though to be clear my point in that post was semi-facetious, in the sense that Google aren’t operating Stadia like a platform anyway. I still think their best strategy would have been to market Stadia as a sort-of low cost-of-entry buffet streaming service for super-casuals, versus the crowd enticed by things like 4K60 and 10.7 TFLOPS.

I think you’re vastly overstating the risk of this. I also wasn’t suggesting giving people completely unfettered access to a PC in the cloud. More the idea of building “Stadia” as a client that gives you access to a variety of existing PC game launchers like Origin, Steam, etc.

No competitive advantage? They have datacenters all over the planet that they have complete control over, and an influence and means of advertising that outfits like Shadow could only dream of. Cheaper to deliver the service, and lower latency.

Making this change would definitely eliminate their (proposed) current differentiation factors like the YouTube integration, but you could absolutely come up with new ones - such as their own version of a Game Pass, tied into the Stadia storefront I mentioned before.

I agree the margins would be way smaller, compared to operating an actual platform - but that was kinda my sarcastic point. Platforms are lucrative, but they also require large investments in robust services, and GAMES. Google have been operating on the cheap, when it comes to development infrastructure, and we all know the state it released in, features-wise. Make small investments, get small returns.

The EGS showed people are hostile to launchers buying exclusive timed rights to third-party games. It also definitively showed that people will ultimately grin and bear it, if your store offers something lucrative that you can’t get elsewhere.

Shadow isn’t really a meaningful metric for the potential of the market. For one, I don’t think I’ve ever seen an advertisement for Shadow… ever. They are a company so small they don’t even appear to have a Wikipedia page. I didn’t even know Shadow existed, until after Stadia was announced, and people started talking about other cloud services.

They also aren’t the only competitor, what with nVidia getting into the same space with Geforce Now. So clearly at least someone is seeing potential in this market.

For many of the same reasons Google tried to hype people up about Stadia. Steadily improving hardware on the backend at no additional cost. Real console/PC games on a variety of devices, and on the road. No massive initial investment in local hardware.

Literally the only meaningful difference would be no exclusive games, since it’d cease to be a discrete “platform” any more. Though even then, if Google retained that Stadia store I’ve mentioned, there’s nothing to say they couldn’t still offer exclusives only available through their store if they really wanted to.

I’m not going to sit here and pretend I’ve run any numbers on whether offering such a service makes sense for Google economically; we are just jawing on a forum after all. My real point was just that if Google are going to operate Stadia like they have, then Stadia itself should change to suit those expectations.

Google seem to want the lucrative money that comes with operating a platform, without spending money on the things that give a specific platform value in the first place. Companies like Sony and Microsoft understand this, which is why they’ve actually expanded their first-party in the past couple of years with studios like Ninja Theory and Insomniac. Google acquired an indie studio that haven’t released their first game yet - and that happened over a month AFTER the November launch, haha.

1801

Google already rents virtual machines where you can install whatever programs you want. But they’re expensive. (Can’t recall the exact rates.) Amazon and Microsoft do this too.

1802

I didn’t even know this. Took a look at the pricing chart for Google’s offering, and quickly blacked out from the relentless wall of numbers, haha.

https://cloud.google.com/compute/all-pricing

1803

At Google you get $200 worth of free service just for signing up if you want to try it out. It’s what I did. Did some raytracing that my PC is too puny for.

1804

Welcome to the Cloud. There’s reasons for it, but I doubt anyone can find the optimal price… which is kind of fine.

1805

Stadia: where turning on auto-updates is akin to a journey into Mordor, but completely re-engineering your home wifi network (don’t need a video game console but buying an expensive new router) and settings is something everyone has the time and know-how for

1806

I wonder if in terms of hardware performance Stadia is a better value than Google Cloud. (I’m not saying it’s a good value to a gamer.)

1807

I spend some occasional time in the Stadia Reddit (it’s amazing, by the way), and the “True Believers” there have really latched on to the whole “no updates!” thing as their rallying cry for the true value of Stadia. It makes me a little sad, really. While being prompted for a game update or something can indeed be a minor irritation… that’s all it is. And if your console is in standby, it should ideally handle that stuff during downtime the majority of the time.

But all of that is beside the real point: If all you can earnestly rally around is not having to install updates, things are pretty dire at the moment.

1808

Consoles have had rest mode for years now though I don’t dispute huge day 1 patches being a pain. That’s something Stadia games won’t have to worry about because they will never get a day 1 port or the data centre will be days, weeks or months late in updating. But, sure they hitch their wagon to this.

There is no value anywhere on either the consumer or publisher/developer side who has to pay for the ongoing hosting costs on top of posting and support UNLESS Google throws a lot of money at them. The only upside is for Google. Well, except for the long term anti-consumer control a game publisher could exert over making all games an expiring-server GAAS.

1809

I just feel kinda bad for Stadia boosters right now. They are getting mocked and laughed at by a huge portion of the gaming community right now, which of course causes them to band together even more. They just want to play games like the rest of us.

Some of them are obnoxious, of course. But such is the case for any fanboy. It’s actually funny in the case of Stadia fanboys, because there’s so little there to latch on to.

Google on the other hand, I have no trouble admitting I’m taking a certain measure of enjoyment at seeing them step on endless rakes. Not because it’s Google, but just because of the hubris with which they handled this entire rollout.

They presumed to saunter into a fiercely competitive industry with a completely unfinished service missing a bunch of core and promised features, they did so with some ports of old games, and an underwhelming indie exclusive, and they were shocked that pre-orders of their Founders kits were “below expectations” (according to Jason Schreier at Kotaku).

What’s that Microsoft and Sony? You have over 30 internal studios between you? Well we have one, that hasn’t even staffed up yet. Come, consumers - sign up for our nascent platform on an unproven technology, featuring no killer apps. Google commands you!

1810

Amazon and Disney have shut down bigger studios with further-along projects for less money than this. I don’t think setting up 1 Jade Raymond-led studio and buying another small one is much of an obstacle to Google pulling the whole plug on the half-assed experiment in 1 or under 5 years. Like others have said, that GPU-driven cloud can be easily repurposed.

Speaking of cloud VM GPU performance capabilities, the Stadia booster crowd are truly “non-gamer elite” clueless! They cling to the fact of data centre customized Vega 56 GPUs without concrete public specs have magical performance advantages on account of them being “customized” when PC hardware enthusiasts know the rough ballpark of what AMD’s GPUs were capable of for certain generations. Their magical thinking extends to another one promised but as yet undelivered feature: just slapping on multi-GPUs to the cloud! “The cloud” will be infinitely scalable and run circles around consoles! Surely, Google will just offer it to game makers and end user suckers subscribers for free right? SLI and Crossfire multicard support has all but died but supporting it in the cloud will be easier I’m sure. The final magical thinking arising out of cluelessness is that somehow “Linux and Vulkan” allow for faster performance because it’s “‘more bare metal.” The Stadia boosters have no fucking clue their ports are just ports of the PC versions with things like graphics detail settings hidden (or not hidden and badly left in like the patching dialogs of NBA2K20) — devs will have spent 90% of their time optimizing for the hardware base that will make 99% of their money not wasting it on Stadia.

One last edit re: the magical custom GPU: Stadia boosters keep bragging about the compute power as a proof of pixel pushing dominance (Hahaha) when it’s easy to tell how it performs exactly as its AMD consumer counterpart in the same games and shares the same framerate ceilings

It’s easy to laugh at them but like someone else said it’s also east to feel sorry for them that they request “features” inherently part of an open PC system but impossible in a walled off Stadia rent-a-cloud ecosystem. Other recent complaints of Stadia boosters from their subredddit:

Wanting to use a friend’s save game to skip tedium of redoing content (something sort of like this is promised for the future)
Adjusting graphics quality more granular level to compensate for how slow it can get (vs a handful of quality/speed presets)
Being able to get Skyrim on Stadia …with mods
Being able to run cheats in a sandbox environment and certainly won’t cheat in multiplayer against others
Being able to play offline (yes this last one seems beyond stupid or maybe they were trolling or utterly uninformed)

1811

I think this is the reason they never spent a bunch of money competing with Microsoft for studio acquisitions over the past couple of years - because they don’t have complete confidence that the service will succeed, and if it doesn’t, they don’t want to have spent a bunch of money on studios infrastructure.

The problem is that you can’t ask for commitment from customers to a brand new platform, when you yourself aren’t showing that kind of necessary commitment to it.

I get that Rome wasn’t built in a day, and neither were the PlayStation and Xbox divisions, but to not have a single major system-seller in the vein of something like Halo Combat Evolved? Seriously? To not at least buy and establish 3-4 studios?

I just don’t get the purpose of this half-in strategy they’ve pursued thus far.

1812

They don’t have to go all-in. They don’t have to ship any hardware.

It’s like saying GOG didn’t go all-in because they only had 100 or so games at the beginning instead of 10000. (I don’t actually know how many games were on GOG on day 1.)