Hacker News Reader

Comment by solardev | Hacker News Reader

Google's product strategy..

Google: We're going to launch a fancy new product in an already-crowded field, not market it at all, and everyone will jump on board and love it!

Everyone else: No way, it's a PITA to port to, and knowing Google, they're going to sunset it soon.

Google: No we won't, we're fully dedicated to this thing

Everyone else: No way, they haven't added meaningful features in years and they're going to sunset it soon.

Google: We're shifting our focus but we're still fully committed to this thing.

Everyone else: No way, they never cared about the home gaming segment and they're going to sunset it soon.

Google: We promise, we're in it for the long haul.

Everyone: No way, they're going to sunset it soon.

Google: Sorry, there's no way we could've seen this coming. We devoted an entire month of resources to this project and thought that was enough! Sadly it hasn't met our expectations. We're not sure why more publishers and players didn't get involved, but we're going to have it to shut it down.

Everyone: No shit, lol.

At least they're giving refunds.

View on HN

I mostly agree with you, but I downvoted your comment because

> We devoted an entire month of resources to this project and thought that was enough!

...is not just wildly inaccurate, but also disrespectful to the people who spent years working on it.

Sorry, but as a player, I honestly can't tell where that effort went... feels like they abandoned it as soon as it launched.

Even at the time of its death, crossplay, cross save etc. didn't work except for a tiny handful of games. The Stadia Plus Chrome extension made a bunch of improvements on their own. There was never a desktop app. No ultra wide support. No RTX. Never got vsync working right.

In the years it was alive, what did they add? How come GeForce Now saw such activity and Stadia got... nothing? I wouldn't blame the people working on it, but some manager in Google really screwed that up.

GeForce now also had the major advantage of being able to play games your already bought on Steam

Yeah, and it was a single switch that publishers could check on/off, instead of having to port their entire game to Linux and Vulkan (what a sibling post said)

Oh, so that's what Vulkan is for. I've been wondering why it exists as I learn graphics programming. I've read that it's faster than OpenGL but also much harder to program for (which is saying something because OpenGL takes some not-insignificant mental gymnastics, at least for a beginner) or vice versa.

Vulkan is needed because the performance trade offs made by OpenGL / direct3D aren’t appropriate for some graphics engines. Vulkan allows knowledgeable & well funded developer teams (like Unreal) to make their own abstraction layer on top of the graphics card’s capabilities. And in doing so, get more performance out of their engines.

Having only OpenGL/D3D was like if the only way you could do network programming was via python. Python is easy enough to use but it’s not well optimized in every use case. Vulkan is like the C of graphics programming. It’s useful when you want to get in there at a lower level, because the higher level abstractions leave too much performance on the table.

Thanks that's a helpful explanation.

D3D < 11

Vulkan is not limited to Linux at all. In fact, it is an open standard made by the same people as OpenGL.

OpenGL was a high level interface, so to speak. And due to this, it made some things easier and other things harder. Most notably, you didn't have the low level access some people/projects wished for. Some time ago, there was a push for a more low level interface and Vulkan was supposed to be the platform-independent, open solution. Naturally, when your interface is lower level, you also need to provide more boilerplate. So Vulkan is slower to get up and running but that's not a particular fault of Vulkan I'd say. The story around it is rather interesting. It is based on AMD Mantle which was discontinued on behalf of Vulkan. Unfortunately, not every vendor was pleased with an open interface. Microsoft released DirectX 12 and Apple released Metal as direct competitors with very similar (if not identical) goals, fragmenting the whole landscape in the process.

Today, Vulkan is supported on Linux and Windows. Apple provides GPU drivers for macOS themselves (not the GPU vendor) and never implemented Vulkan support to not cannibalize its own Metal (I suppose). However, there exists MoltenVK, an open project that creates a translation layer from Vulkan to Metal.

Some games, unfortunately. Lots of devs are choosing not to join it, which is very silly to me to be honest. It has absolutely no impact on their sales or development. I bought the game on steam, they have my money. Why do they care what machine I use to run it? Better yet, why do they require I use my hardware? Blizzard famously (famously in the GeForce Now/cloud gaming community that is) pulled WoW without notice when people started playing it in larger numbers.

It’s very irritating.

In a way that's also a disadvantage - people who have a Steam library probably already have something to play it on and have at best a limited need for GFN. And if they have a PC + Steam then with Remote Play they already have their own streaming solution if they really want one.

How is that a disadvantage relative to Stadia? If they don't want to stream a game at all, why would they bother with either service?

And if they did want to stream a game from time to time (for me it was being able to play on a laptop anywhere, not just on my own LAN), Steam games were usually cheaper from sales and resellers, there was actual multiplayer, and the Steam ecosystem has a bunch of other advantages. It was just a better service.

And GeForce Now had better hardware, ray tracing, customizable graphics settings, vsync, etc.

If you don't need to stream a game, well, you don't need any of these services. But if you did, GeForce Now and Steam were just plain better than Stadia.

It's a disadvantage compared to the Xbox and PS streaming services - both of which cost less and come ready-made with a large library of games. It's definitely better than Stadia which seemed to consistently be the worst out of all the streaming services (which I think today's news vindicates).

GFN is good especially if you live in an area serviced by it's RTX 3080 streaming "rigs" which make it reasonably good value for money vs the cost of building an equivalent PC. But it's still the most expensive service on the market PLUS you also have to BYO library which rules out entire groups of potential users - Nvidia knows this as well based on how hard they're pushing the free-to-play titles. On top of that a lot of titles aren't even playable on it; I have a steam library of ~700 titles and apparently only ~140ish of them can be played on GFN.

Personally I like the GFN idea of being able to "rent" hardware and that's a business model other companies have proven works to a certain degree since ~2015 (Shadow.tech being a good example). However I don't think it has the same mass market appeal of the other surviving services which give you the hardware + a huge library of content just like Netflix/Disney+/Stan etc.

Stadia needed you to double dip if you wanted to play games you already owned on Steam, but GFN did not. That’s easily hundreds of dollars of a game library.

but also disrespectful to the people who spent years working on it.

Something being tragic doesn't make it not a waste of time. The writing was on the wall since day one. Nobody was going to pay full price for games that evaporate the moment the servers shut down being offered by the most ADHD company in existence. If you chose to spend your time and energy working on that and expected to accomplish anything other than being paid you were a fool.

The excerpt I quoted isn't about whether it was a waste of time or a savvy career move, just about the total amount of work that went into it.

> Nobody was going to pay full price for games that evaporate the moment the servers shut down

Eh that's true for Steam, Epic, GoG, et al. If anything, regular people trust Google way more than those other companies since they use the browser/search engine/mail/youtube every single day

I do think it was doomed from the start, but only because the subset of people that have high speed internet, disposable income, like gaming and DON'T already own a PC or console is a very very small market

Ha, I literally described Google as having corporate ADHD just yesterday

I can see where you're coming from.

I think it's appropriate enough to interpret the commenter in jest and see that their point is that google spent 1/nth of the required time/resources necessary for it to succeed.

In that light, it reads to me more like having some understanding for the people working on it that they were set up to fail, and the failure is not a reflection of their effort.

It's exaggeration for rhetorical effect, an extremely common stylistic device.

when everyone is "exaggerating" about how Google shuts every single thing under the sun down, it ceases to seem like a joke.

When was it a joke to say Google sunset every side project of theirs?

Imagine you go to work at Baby Mulching LLC. This company has existed for two decades, and is dedicated to dropping infants into industrial shredders. Everyone hates Baby Mulching LLC, and they're an enormous black stain on your resume, because there is one single thing they do, and it's grind up children.

On your first day, you're asked to design a big toothed wheel. You ask your boss, will this be used to mulch babies? No no no, he says. It will be used to chop up entirely unrelated organic material. You spend three years on it, and once finished, it's immediately incorporated into the Baby Mulcher 3000, the next generation baby mulcher, which can puree one hundred toddlers per hour.

Everyone starts yelling at you for building a baby mulcher. But it wasn't your decision! You spent years working on it! This is so unfair!

Unfair, maybe. But was it unpredictable?

What relationship does this anecdote have to this scenario exactly, other than tenuously drawing a connection between Google and baby grinding?

As a third party with no skin in the game: no idea but it was hilarious.

I think what they're trying to say is that the stadia engineers should have seen the writing on the wall, coupled with the questionable history/actions of their employer when it comes to products regardless of their own good faith efforts as an employee.

I'm still not really sure what the moral of this parable is, but thanks for the most enjoyable read I've seen here all week, lol.

Disrespectful is interpreting words in bad faith. Obviously OP didn't literally mean 1 month.

>...is not just wildly inaccurate, but also disrespectful to the people who spent years working on it.

Fortunately you are here to be anonymously offended on their behalf.

I'm not anonymous; it's easy to figure out my identify from my HN profile.

I didn't work on Stadia, but I did work on Google Fiber for a couple years only to have almost all that work cancelled, and I'm still sad about it many years later. At the time, I really wouldn't have appreciated people on HN rubbing salt in the wound.

For what it's worth, I don't think anyone blames the engineers doing the work. That thought never even crossed my mind...

In fact I think Google has some of the best engineers in the world. Sadly, they're handled by some of the worst management and leadership in history. Your CEO has no vision or charisma and the whole company is just spinning in place, waiting for something to happen. Your middle management looks like little fiefdoms actively working against each other in some eternal civil strife. Google is its own worst enemy.

From the outside it looks like you guys are a bunch of really smart lab rats stuck in some perverse maze, being forced to run little Hunger Games style competitions to entertain some higher power.

I wouldn't want to work there for a million dollars. It jumped the shark a decade ago.

Fiber was just another tragedy in a long long list, another fight Google picked and then chickened out of. Google hasn't had a major success in recent memory, just a graveyard of half-assed attempts. But that's a bigger cultural problem at your company, not the fault of any engineer or single team. I sincerely feel for you, and I hope you got out and found something more constructive.

Edit: Sorry, not Fi, Fiber.

I left Google earlier this year. I think it was the right move for me at the right time. That said, I think your view of Google is a bit darker than the reality.

I hope so! They have a lot of talent, I just hope they eventually get better leadership to actually steer the ship somewhere, instead of just relying on the sheer inertia of ads.

You both have a common enemy - nobody is blaming Google being Google on the employees - they also get abandoned without support and their project cancelled.

Dude, they're making half a million dollars a year, they're gonna be fine

https://twitter.com/GoogleStadia/status/1552989433590214656

@GoogleStadia "Stadia is not shutting down. Rest assured we're always working on bringing more great games to the platform and Stadia Pro. Let us know if you have other questions." Jul 29, 2022

2 Months ago

how is this relevant to what they said?

When the people who make the product literally publicly lie (while being paid extremely well), comments like the OP's (that are obviously tongue in cheek) are certainly not "wildly inaccurate, but also disrespectful".

It's not a lie if they believed it. And I don't see how this justifies calling the developers lazy.

Why would you conflate some VP with the PMs and SWEs who built the thing?

> Everyone else: No way, it's a PITA to port to, and knowing Google, they're going to sunset it soon.

I’ve seen a few developers (Ryan Gordon, IIRC?) saying that the Stadia SDK was actually great. It was basically just a bog standard Vulkan/Linux environment with much fewer unpleasant surprises than consoles.

The whole stack was great. Google engineers worked their asses off on it. Very smart people throughout. The management was awful though.

In the 10 years I was at Google I never really had to face a "crunch" situation but there I was right before Christmas working til midnight on the very tiny bit of Stadia I had somehow been dragged into despite telling my manager I didn't want to be anywhere near it because it was radioactive (My rule learned the hard way was always stay away from the "hot" project at a company like Google. It just becomes a feeding frenzy of empire building and egos, and steady incremental contribution will get you nowhere.)

I had it much better than most though. If I recall: The original app setup (integrated into the existing Home app) was tossed at the last minute. The entire out of box process for the controller redone in the process. In literally the last few weeks before ship date.

The controller folks I knew were heroes. It's sad to see their hard work thrown away.

There's no way Stadia in its entire existence made enough revenue to cover the sheer number of SWE-hours put into it, especially the spent SWE-hours caused by last minute product changes; which nobody further up ever had to pay a consequence for.

Do you think Google will unlock the Bluetooth for the controllers so we can still use them? Maybe a firmware update or open sourcing the drivers.

VS GeForce Now, where you check a box saying "I want my game to be streamable" and then... you're done.

https://partner.steamgames.com/doc/features/cloudgaming

Yup... until one day the game is taken off the story and that checkbox removed. Big business is rough.

Then you lost the 15 minutes you took filling out that form? And your players can still stream it elsewhere, like Shadow or Luna? Doesn't seem terrible.

VS the days/weeks/months it took you to port to Stadia, only to have the entire service shut down.

The point is that there are many PC streaming services that are basically "no porting required", whereas Stadia opted for a strange sort-of-console, sort-of-not model (despite PSNow and XCloud already being able to stream actual console games already) that required you to spend dev hours porting your game.

It backfired because most devs didn't want to (or couldn't afford to) port their x86 game targeting Windows to some tiny proprietary platform, ESPECIALLY when it's Google hosting it. Of the few that did (Orcs Must Die), Google arranged some exclusivity deal with them, lying that Stadia could do things no gaming PC or cloud streaming service could do due to Stadia's special scale or something. It was a lie. When the exclusivity expired, it showed up on Steam and GeForce Now and ran perfectly fine, and got a lot more players to boot.

Stadia had some really awesome tech -- namely the UX of being able to boot straight into a game without waiting for Steam Big Picture -- and some nifty (but relatively useless) side features like being able to capture a memory snapshot and resume that later, emulator-style. But I don't think their management really understood the PC gaming culture and what was truly important to its user base, and failed to take years of pleading and feedback into consideration. They just arrogantly did... something else (or nothing much? I can't really tell)... with Stadia and drove it into the ground. So sorry for all the engineers who worked on it and had to see it nosedive like that due to managerial incompetence.

None of this was a surprise to any gamer actually watching this space. Stadia came late, delivered less, and exited early. Google's product culture doomed it from the get-go.

depends on how you game in the future. "Show's not on netflix, but I can at least buy the Blu-Ray or subscribe to Hulu for it" may not cut it for some people and it's just "gone" for them. You already have gamers arguing over free launchers after all, I can't imagine asking those same kind of people to "just subscribe to a new launcher for X dollars per month again".

Some consumers can be fickle like that.

I'm not sure what you're getting at. Like DRM and subscription marketplaces where you basically rent the rights to a game for some indeterminate amount of time?

Most people seem fine with that, and Steam totally killed the physical games segment. There's GOG if you want a copy without DRM, but they're tiny and the prices are often worse than Steam or a Steam reseller.

Besides, if the game rental model is bad, Stadia does it even worse than Steam: tiny library, horrible pricing, incompatible multiplayer, proprietary ports on a different update schedule, exclusivity deals, all from a company with a proven history of abandoning their customers.

Steam on the other hand is run by one of the oldest, most respected PC gaming companies in the history of the hobby, owned and run by a bunch of gamers (not advertising execs), who singlehandedly transformed the way people buy and get games, and worked tremendously hard to make the service better abs better every year in the service of players. They've earned our loyalty.

And furthermore, Nvidia is the provider of GeForce Now, not Steam directly. Their model is just way better, running off existing Steam and Epic games that don't require a new port or for customers to repurchase or lose their save games. Nvidia is also a gaming company. Not exactly as beloved as Valve is, but at least they're a long term player in the field.

Google just never really got it, at all. They arrogantly assumed they could just make a mediocre copycat and people would flock to it. Oh well.

Aside from the unpleasant surprise of there not being an audience.

>not market it at all

I noticed that too. Google never adversises its producta. Google is almost like the startup founder who thinks their side project is so good it doesn't need marketing or sales.

They absolutely inundated YouTube with ads for Stadia for a couple of months, to the point that it felt like half the ads I saw were Stadia ads. So that’s something, at least?

I wonder what the overlap is between Stadia's target market and people who watch YouTube without blocking ads.

It's not zero, because Chromecast doesn't block ads and that's the natural fit for Stadia. But it's definitely not 100% either.

I don't see ads at all, period. I either Ublock origin or pay for a subscription. I pay for youtube premium.

How would an ad reach me? Youtubers? Can Google use its "contractors" to promote its own product? That just feels weird.

I remember a talk given by chief of czech google at our highschool, she told us that we are one of the few countries where they had to use marketing to compete with for search market (I remember seeing ads for Google chrome in metro when it was new. ).. I think as a monopoly they just usually dont feel the need for marketing when there is no else running ads...

I've been seeing ads for Chrome around northern Seattle (Shoreline) area

At first I was baffled why they would possibly need to advertise Chrome in the US. But in retrospect it might just be feather ruffling in Edge (Microsofts) backyard?

There's also the big one right here: https://www.google.com/maps/place/47%C2%B036'07.6%22N+122%C2... when you get to downtown from the ID (where the 3d view shows a verizon ad, it's now a "no place like chrome" ad.

Are these new? Reddit had a particularly Firefox-meme laden week last week after all of the MV3 chrome discussions.

Chrome is core to Google's business so it's worth it advertising, they also advertise the Pixel phone a lot in high-end places in global cities because they need to gain market share for mobile browsing and phones against Apple.

Anything else, not that important, even GCP which is their major bet doesn't really get any ads because it's a business product so not much sense in doing so.

They seem to believe word of mouth/viral marketing works because it worked for gmail and chrome, so they didn't double down on Stadia before product market fit, and that caused Stadia to fail(along with Google's short attention span reputation).

>Google never adversises its producta.

Not really true (e.g. Chrome) but it's probably true they advertise relatively little given their size.

They just know first hand that online advertising is not worth the money!

I have seen plenty of (annoying) stadia adverts

Partial list of products where this applies: Meebo, Buzz, Orkut, Google+, Notebook, SideWiki, Schemer, Spaces, Checkout, Directory, Sync, Hangouts, iGoogle, Knol, Lively, Moderator, ....

There must have been hardly anyone stupid enough to buy many games. It’s probably not even much money to refund.

Not many games, but they didn't often have sales (so many were stuck at MSRP for years), and they also had some hardware sales (controllers, etc.)

For Google it's chump change, I'm sure, but I still appreciate it. I bought like two titles on there before realizing it was a doomed effort... meanwhile GeForce Now sees very active improvements and a much bigger library.

GeForce now is great and if I ever get a new gaming PC, I can play every last game on my own GPU. The tech works perfectly, the free tier is generous and the subscription is a value. Stadia was behind the moment they launched and flat out refused to give an inch to improve the product.

This is a company so obviously in love with its incredible success from ages gone by that it thinks we still owe it our brand loyalty. We do not.

Would be cool if I could use stadia hardware for GeForce now

Yeah, too bad those controllers don't work as generic Bluetooth ones :(

(if you didn't know, though, xbox controllers do work great with geforce now)

Reply on news.ycombinator.com