>all the negative reviews of Stadia are by people who already have console/gaming PCs.
I think most people who like playing video games would have some sort of video game playing hardware, yes.
So, they might have opinions on what a good gaming experience is. And, if they have issues with Stadia (incl latency, graphical fidelity, and library), then it might be coming from an informed place.
Yup, also if you look at people who write or have written netcode for a living, it's been pretty obvious where Stadia was heading for a while.
You need to be able to reliably hide ~100-200ms of latency for any number of reasons. If you're talking about something that runs over wifi you're fighting against the laws of physics and no back-end infrastructure is going to save you.
The way you do that is with client side dead-reckoning and/or lock-step simulation with a pre-shared seed. Streaming video eliminates any opportunity to do that and you don't get to buffer frames ahead like Netflix can.
There's a reason OnLive folded and I wouldn't be surprised if you see Stadia EoL'd in the next year or two.
I agree with your overall point but not so sure on this specific piece:
>If you're talking about something that runs over WiFi you're fighting against the laws of physics and no back-end infrastructure is going to save you.
Signal propagation speed (velocity factor) over air is only a bit slower than cat7 copper, and on-par with fiber optic. From memory:
In a vacuum: VF = 1 (equal to speed of light)
Over copper: VF = .75 (one quarter speed of light)
Over air or fiber: VF = .66 (one third speed of light)
Wi-Fi is measurably slower than wired copper cable for a few reasons but I don't understand physics to be one of them. With the distances relevant to last-mile connectivity like we're discussing here, that difference in VF appears to be immaterial.
Then again, you could argue that the air being a contended medium is by it's very nature a physics problem, and that does seem to be the root of most of the drawbacks of running low-latency workloads over wireless. Not necessarily an insurmountable physics limit though, with spectrum allocation, code division, and beam forming already doing a lot to mitigate the issue, there's the potential for further developments that might bring things to parity.
Still, you're not going to get past the VF limit I described above but again, if we're only talking about in-home WiFi rather than across the entire service-to-consumer run, it doesn't play a large part.
100-200ms latency seems a little extreme to me. Its an 8ms ping from my home to the nearest Google edge node (supposedly where Stadia nodes will live), 9-11ms on my WiFi. Compression and decompression of the video stream with hardware acceleration and tuned for speed over quality should only add maybe another ms or two. Where is the extra 80+ms coming from?
All it takes is someone turning on a microwave down the block from you to drop 4-5 packets in a row and you get close to the 100ms number.
Channel congestion in dense urban areas approaches the same limit, heck Ars was just talking about a technique today[1] that because it doesn't respect channel slots drops throughput of any router within 60m by 20%.
100-200ms is an upper bound, however they were reliably doing client-side latency prediction back in '96[2] over 300-400ms dial up connections with little fidelity loss including servers supporting 200+ concurrent players.
I'm sure there's a niche of people for which streaming is usable, however we would design games with latency in mind(and how we'd deal with it) as a part of fundamental mechanics of the game. Here they're just slapping them on a video stream and hoping it works.
To be fair, network connections have significantly improved since the 90s.
Your point still stands though, on a wireless network, it only takes some interference from other devices to cause latency spikes that make latency sensitive games frustrating to play, even if the actual upstream connection is of high quality. I've had quite a bit of issues with playing games on wireless networks in urban areas that go away magically by using an ethernet cable; the stuttering that happens when ping spikes from 30ms to 400ms is very jarring.
If Stadia is used for games that are not latency sensitive, such as deckbuilding games or turn based strategy games, it's fine. Using it for competitive first person shooters is likely to be very frustrating on anything but a wired network with good upstream connectivity or a wireless network with no major noise sources.
Yeah and wifi is already pulling out all the tricks like FEC, CSMA and the like just to scale with usage.
At the end of the day the internet is a packet switched and not circuit switched topology despite telecom pushing hard for circuit switched in the early days. Preference is given for throughput and not latency(although network-next is working to address some of that).
Serious question, what about the general increase of bandwidth usage?
In my mind things like Stadia will increase the amount of data traffic exponentially, even at a low market penetration rate. I think I saw one estimate that it's going to be 120gb on average per hour. Even at 100gb, the extra traffic at scale is going to be insane.
Regardless of if current infrastructure can handle x% market penetration of that kind of usage in a, what, 1 year time? Should the infrastructure even be required for that kind of insanity? We're not talking about life-saving data transportation. We're talking about god damn video games. Or am I way off my whiskey fueled rocker?
> Compression and decompression of the video stream with hardware acceleration and tuned for speed over quality should only add maybe another ms or two.
This estimate is waaaaay off. Nobody has been shipping video encoders or decoders setup to pull 1000fps nor tuned for latency.
Realistically it's 8-14ms to decode if you're lucky (low latency hardware+firmware), and comparable to encode. And these streams are made entirely of key-frames, which is a hyper unusual scenario for video systems.
That all adds up to maybe something like +30ms over baseline. Which is a lot. If you have a console try playing it on a TV with and without game mode enabled. That's about a 20-30ms latency difference in general.
It's worse than that. Assume you're at 60fps like the average non-serious gamer. Assume you're lucky enough that video encoding / decoding adds only 1 frame of latency. Assume Google's proprietary technologies can minimize rendering latency so much that that's lost in the noise (I don't believe it). You've still got network latency. How bad is that?
Well, I'm in one of the 10 largest cities in the United States. Pinging 8.8.8.8 shows an average of 20 ms RTT, which is just over one frame. So is the total a 2 frame delay? No. Think about how Youtube works: they choose to introduce more latency than exists in the raw connection for reliability. My ping time shows a std deviation of over 10ms to 8.8.8.8. So if Youtube ran at a 20 ms latency (the true network latency), just about every other frame would have to briefly pause to wait for the next frame to arrive over the network. So you introduce latency. Youtube introduces about 3 seconds of latency on my network (a very rough estimate).
With a video game, the situation can't be fixed like that. You can't just run your video game behind your inputs by 3 seconds to iron out network latency spikes. So you've got to pick a percentage of packets that are going to arrive late (briefly pausing the stream), and just hope that you never get huge latency spikes of a second or so. (This actually happens all the time on busy networks due to buffer bloat.)
If you pick 95%, 1 out of every 20 video segments are going to have to pause for the network to catch up. Given we're expecting 60fps video, I'm assuming that's considered unplayable. So lets make it 99%. Assuming latency follows a bell curve, that means you have to introduce 45 ms of latency to allow smooth playback. That rounds to 3 frames of latency.
So all together, the "ideal" experience in my major city is to play a video game with 4-5 frames of latency, and realistically my developers are probably asking for a 100ms input-to-paint window. That's TERRIBLE to the point that it's probably only really usable for "clicker" time-waste games and turn based stuff. And the thing is, latency doesn't follow a bell curve. In my test lasting less than 30 seconds, I saw ping times well over 60 ms, and hardware associated with gaming will have its own latency issues too. So I expect even a stream running at 100 ms latency will still have brief pauses and jumps every few seconds, and probably noticeable ones every few minutes.
I think Stadia could work in big cities. If Google has servers in the city and you're using a wired connection then it should work fine for a large portion of games. It probably won't give a good experience for first person shooters though.
> You need to be able to reliably hide ~100-200ms of latency for any number of reasons. If you're talking about something that runs over wifi you're fighting against the laws of physics and no back-end infrastructure is going to save you.
There's margin no? I mean, right now I get 3ms ping to Google from my wired computer, and 10~15ms from my phone on wifi. Wifi doesn't add that much.
And I'm in a large European city, not in Silicon Valley.
I think anyone who has interest in video games should be concerned with what we give up by games a service model. It means all of your games can now be censored or deleted at any time without any notice to you. It means you can't mod your games as you please. It means you can't resell the games you've bought or share them with friends and family. It means you can't play without internet access.
Beyond the overpriced hardware and monthly fee there's a lot being sacrificed for the ability to game anywhere. Another thing you give up is your privacy as this will simply be one more way to track you, your location, what you're playing, when and how long you play, who you're talking with online and what you're saying. All of that will be tracked and logged and analyzed by google.
>The fact is that Stadia makes it possible to play games on devices where it previously was impossible is pretty compelling. (even if it's at 720p).
It makes it possible to play modern AAA games on devices where it previously was impossible.
That's a dramatic difference. There are plenty of incredible gaming experiences you can have on low-powered hardware, without needing a clunky streaming-as-a-service infrastructure to back them up. They just won't be the latest FIFA/Elder Scrolls/Call of Battlefield: Modern Duty.
Honestly, I tend to think that having environments that don't have full fat AAA games available is good for the gaming market as a whole. It opens up markets for indie developers and new formats. I'm thinking of the Wii and Switch for comparison-- if they had been built with the same spec hardware as their competitors, they would have been flooded with ports of the big multi-platform franchises, rather than the unique experiences they're known for.
Sonme of my favourite gaming experiences include playing Zangband on an already-obsolete Pentium II in 2003, and Cave Story on a pre-Atom netbook starring a 900MHz Celeron-M.
I think this is truly the crux of the product's potential. Triple A FPS games are never going to hit it big on a streaming service, especially in any sort of competitive circle.
The real opportunity is for games where latency isn't really important. Beautiful turn based games, cinematic choose-your-own adventures.
The second path is for games to be developed with appropriate tolerances & latency-forgiving mechanics. Take a modern platformer like Super Meat Boy or Celeste, where you are given a window of frames in which you can jump/react before failure. Even modern games like Mario Odyssey accommodate input latency of wireless controllers/HDMI displays. Similar accomodations can and will be made for games targeting streaming services.
> I think most people who like playing video games would have some sort of video game playing hardware, yes.
False. Almost everyone who owns a smartphone plays videogames now. The mobile game market is huge. A lot of people dont own hardware specifically for video games, but still like to game.
Its very possible stadia will work on android soon . Xbox has a similar service in brta that already works on phones.
The context here is obviously PC/console gamers. No one is clamouring for Candy Crush to be brought to Stadia. There may be overlap, but the mediums are completely different.
I think most people who like playing video games would have some sort of video game playing hardware, yes.
So, they might have opinions on what a good gaming experience is. And, if they have issues with Stadia (incl latency, graphical fidelity, and library), then it might be coming from an informed place.