This has the best chance of succeeding. Using the newer QUIC protocol to cut down on latency, this may actually have a very good chance. People complain about latency, but a lot of console gamers don't even enable game mode on their tv. Of course this isn't going to be an option for the 'leet pro gamers' as those shooting games need every bit of latency removed. For the other 99% of games and gamers, this could be perfect.
I disagree this could be perfect for 99% of gamers out there. If you want to game with lower quality compressed graphics and compressed sound with input lag, Stadia is for you. A few websites have posted graphics comparison of Assassin's Creed Odyssey between Stadia, PC, PS4, and Xbox. Stadia graphics look pretty compressed and look the worse in the comparisons. If you already have a console or PC, which is the majority of gamers, I don't see what the point of Stadia is.
The stat Google is touting as Stadia being more powerful than a PS4 Pro and One X is a bit silly. It's a streaming platform. I checked out the announcement and it was pretty funny the latency issues during the demo. I'm highly skeptical Google can keep latency as low as they are claiming. No current cloud services have 0 latency. Stadia may work fine for single player games, but no way there won't be latency issues playing a game like Battlefield V.
According to Google latency won't be an issue for multiplayer. They had a segment of the presentation where that was discussed and the reasons latency won't be the problem some are expecting based on current gaming platforms. It will still be a few months before you can see for yourself if it's a problem or not. Heck, Microsoft thinks it's the way forward too and creating their own version of it.
FWIW I didn't see the "funny latency issue" you were mentioning but the livestream is still available to look at it again.
Early testing shows there are latency issues with Stadia. I don't think it will be any different when it comes out. There are too many variables such as internet speed, TV/monitor, etc. I read quite a few people complaining of input lag playing Doom at the demo area. I really don't see who the target audience is for this. Gaming on a PC/Console will still be a way better experience.
Who's the target audience for gaming on an iPhone or iPad or Apple TV? That one is spending $B's on it, latency be damned. Stadia will have better games and more devices those games can be played on. Not sure what your objection to that would be.
"It'll be good for casual gamers, probably eat away even further at the low end of the console market, but I don't see it having much impact with high-end gamers.
My round-trip latency to the nearest Google datacenter, as tested by pinging their DNS, is 8-10ms at work, ~20ms at home. My primary gaming PC hits the 200fps cap on Doom 2016 - that's 5ms. So even if they throw enough rendering power at the games to render in zero time, they still would be worse latency for me - and latency is, at this point, the killer issue.
10ms isn't bad. Add some time for rendering, they're still under the 16.7ms needed for 60fps. So they'd be comparable gameplay quality to current-gen consoles. My home internet would have latency similar to 30fps - still tolerable, particularly in slower games, but not ideal.
I can see this being a perfectly fine way to play RPGs, action/adventure games, stuff like that. An extra 20ms is meaningless in a turn-based game. It would also likely do well in online games - there's a lot of latency anyways, this just puts that in the controller-to-renderer phase instead of the renderer-to-gameserver phase. You do lose the benefits of prediction but that hasn't been super-critical for years now.
It will be a suboptimal way to play some genres - shooters, fighters, racers, some sports games. The lag will be noticeable even on good connections, but we're at a point where this is at least viable. I can see budget gamers going in on it - people who might have bought a Switch or a used PS3/Xb360, instead going with streaming. Especially if it has a wide range of games, and regularly cycles in new ones. Spend maybe $20/mo on a subscription instead of dropping $60/mo on a single game. I don't think even that will bring in the hardcore gamers... they'd be behind the release schedule and possibly lose access to old games.
But the hardcore gamers like myself are not the bulk of the market. The gaming PC might be the best way to play, the big consoles might be what most people see as mainstream, but historically, most gaming sales were on cheap handhelds - the Game Boy and DS. Now that handhelds are basically dead (save for the half-portable but high-cost Switch), the bulk of the market has switched to smartphone games. But that's a very painful way to play a lot of games - Fortnite might make money on it, but even with a controller, it's an awful way to play it. I can see streaming to a TV or non-specialized computer being very enticing to such people.
I think Google has a chance to succeed. They need to focus on the value - keep it cheap ($10-20/mo - and maybe have a free ad-supported tier with a limited game selection), and put it on every device possible, to the point that most people will already own a compatible device. Get a very wide library of games - not just old ones, but also recent titles, no more than two months behind console release. Don't bother with undemanding titles - if someone wants to play Undertale, they'll just buy it because it can run on any computer made this decade. Get all the main publishers on board - out of EA, Activision, Ubisoft, Bethesda and Square Enix, they can afford to miss no more than one. Don't even try to secure exclusives - that's a PR nightmare right there, as the Epic Store is finding out.
They've got a shot. Not with me, but they've got a shot with the general market. It will be interesting to see how it pans out."
Comments
"It'll be good for casual gamers, probably eat away even further at the low end of the console market, but I don't see it having much impact with high-end gamers.
My round-trip latency to the nearest Google datacenter, as tested by pinging their DNS, is 8-10ms at work, ~20ms at home. My primary gaming PC hits the 200fps cap on Doom 2016 - that's 5ms. So even if they throw enough rendering power at the games to render in zero time, they still would be worse latency for me - and latency is, at this point, the killer issue.
10ms isn't bad. Add some time for rendering, they're still under the 16.7ms needed for 60fps. So they'd be comparable gameplay quality to current-gen consoles. My home internet would have latency similar to 30fps - still tolerable, particularly in slower games, but not ideal.
I can see this being a perfectly fine way to play RPGs, action/adventure games, stuff like that. An extra 20ms is meaningless in a turn-based game. It would also likely do well in online games - there's a lot of latency anyways, this just puts that in the controller-to-renderer phase instead of the renderer-to-gameserver phase. You do lose the benefits of prediction but that hasn't been super-critical for years now.
It will be a suboptimal way to play some genres - shooters, fighters, racers, some sports games. The lag will be noticeable even on good connections, but we're at a point where this is at least viable. I can see budget gamers going in on it - people who might have bought a Switch or a used PS3/Xb360, instead going with streaming. Especially if it has a wide range of games, and regularly cycles in new ones. Spend maybe $20/mo on a subscription instead of dropping $60/mo on a single game. I don't think even that will bring in the hardcore gamers... they'd be behind the release schedule and possibly lose access to old games.
But the hardcore gamers like myself are not the bulk of the market. The gaming PC might be the best way to play, the big consoles might be what most people see as mainstream, but historically, most gaming sales were on cheap handhelds - the Game Boy and DS. Now that handhelds are basically dead (save for the half-portable but high-cost Switch), the bulk of the market has switched to smartphone games. But that's a very painful way to play a lot of games - Fortnite might make money on it, but even with a controller, it's an awful way to play it. I can see streaming to a TV or non-specialized computer being very enticing to such people.
I think Google has a chance to succeed. They need to focus on the value - keep it cheap ($10-20/mo - and maybe have a free ad-supported tier with a limited game selection), and put it on every device possible, to the point that most people will already own a compatible device. Get a very wide library of games - not just old ones, but also recent titles, no more than two months behind console release. Don't bother with undemanding titles - if someone wants to play Undertale, they'll just buy it because it can run on any computer made this decade. Get all the main publishers on board - out of EA, Activision, Ubisoft, Bethesda and Square Enix, they can afford to miss no more than one. Don't even try to secure exclusives - that's a PR nightmare right there, as the Epic Store is finding out.
They've got a shot. Not with me, but they've got a shot with the general market. It will be interesting to see how it pans out."