-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
-
We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.
Google Stadia will be “faster and more responsive” than local gaming hardware in the near future predicts Stadia dev
- Thread starter Deleted member 44371
- Start date
I have no doubt Stadia will be faster and more responsive than the worst local gaming hardwire / setup. After all, plenty of players are ok with laggy games, so there is definitely a market for it.
If we're talking about beating the best possible local condition, that's just impossible unless dark magic is involved.
If we're talking about beating the best possible local condition, that's just impossible unless dark magic is involved.
I've posted my thoughts and I don't want to run around in circles debating minutiae with you. The article that was posted in the OP received an update to clarify -
Right there confirms processing dedicated to multiple renders/variations due to prediction. This is the X times (10 was the example I used, obviously this will fluctuate dynamically) required to have things ready to go or override user inputs as Stadia or developers choose. So whatever shortlist of renders Stadia prepares will require more processing power to deliver. Whether some in game/stream events are 1 to 1 or 1 to 4 or 1 to 10 will vary on implementation, game actions, inputs, predictions etc. Anyway you cut it the concept of negative latency will require more processing than a local hardware console, therefore it raises the question of how the investment vs performance per player per game is delivered by Stadia.
I reserve serious judgement that Stadia will outstrip next gen consoles within 12 months as the VP claims. Perhaps for games with lower population and high server instance dedication but hugely popular games struggle to even deliver patch downloads let alone predictive renders, streaming and remote inputs.
Last edited:
Again I reference the comments and sales pitch by the VP, this thread was posted on it. Even for the US I have serious doubts Stadia will push past next gen consoles visually or internetworking wise. Lower cost entry? Sure Stadia and xCould etc stand to reason. Performance visually and networking wise I do not agree with the statements by Stadia's VP.
I use my Australia edge case as a method to highlight the facts. Just as relevant for this in the US. Also the global market share of gaming is not just USA-centric anymore.
OK, so you actually don't have a source to what you said. Why is it so hard to admit that? Will you simply keep saying that Google said that Stadia will provide 10 times the processing power? I also asked you 10 times what? Xbox One, PS4, etc?
What are you understanding from the highlighted other than the game will only move to the direction from the player's input? We are all clear that some type of prediction will happen, the thing is that we don't know exactly how it will work yet. Where are you getting that Stadia will override the player's input? Is that what you are understanding from " Whether or not a player takes the anticipated path or another entirely remains dependent on local player inputs."? It will definitely require more processing power, but again we don't know how much. If the motion vector example from DF is correct, a lot of the information on the simulated frame will be reused from the previous one. Is that why you understand that Stadia is saying that they will offer 10 times the power? You need to understand that a lot can be happening here that we don't know yet. Time Warp is another cool technique that is already being used right now on VR headsets and Stadia might be using something similar, to simulate possible frames reusing the previous frame information.
As you can see there is a lot to learn here and all of it is super interesting, if you care to take the time and see it. You are right, we don't know if this predictive technique will be used on every single game, some of them, half of them, only one, etc. We don't even know exactly how it will work to make an educated guess on that.
Last edited:
Nobody is talking about negative latency, there is always latency locally too so they only need to be lower than that.
Digital Foundry already saw the same latency (166ms iirc) on ACOdyssey back on the last Stadia demo as on Xbox One X locally. And things can obviously improve.
Unless I'm misremembering that video, the latency you're talking about was actually what was introduced in addition to the game's input lag. But even if it wasn't - that's not even the most inherent problem with streaming games. It's the variance of that latency that'll always be the issue.
That makes it a complete no-go for any reaction heavy, or timing heavy, games. Rhythm games you can just through out. Fighting games... I don't know why SNK are bothering to put Samurai Shodown on there, as I'm sure even casual fighting game fans will notice that variance. Some games are better for streaming, yes, but "better" doesn't mean "good". Trying to calculate what someone's inputs will be before they do them doesn't solve this problem.
I've played VF5 on PS Now and it feels fine. Expecting Stadia to be as good if not better than that. The tech works
No,
"The Stadia tests were carried out on a Google connection vs our Stream tests 'in the wild' on a 200mbps connection. However, the Stadia tests include display latency on a Pixelbook (which we cannot measure) and were also carried out via WiFi networking. Both of these factors will add additional latency, while our Stream demo tests were carried out via a LAN hook-up to the router."
"Based on the results we're seeing here running the PC version at 60fps, latency should drop by 33ms. The Xbox One X version is already bafflingly high in latency terms, to the point where Stadia can match it. Based on the PC results, running at full frame-rate would actually see the streaming version surpass the Xbox game."
If the ping will be stable or not will depends on several factors, but it's not like everyone will have a highly variable ping. Also, have you ever played an online fighting game? Are those imposible too due to how ping can vary? If you care about how Stadia will handle variable latency you can see the video below.
Measure your ping to several Google servers and see how much it varies on the server with the lowest latency. Try it 10 or more times and see what range do you get.
No it was the total latency, compared to PC and Xbox One X. PC was faster though, so that's the real competition, consoles will be seen in the rearview mirror pretty soon I think. In general people don't realize how much latency they're already dealing with in console gaming. Even fighting games have a couple of frames latency even without the console-HDMI-AV receiver-HDMI-TV latency chain included.
Anyway, I'll test it all out pretty soon. Can't wait! The controller being connected directly to the Stadia servers instead of a box and then the servers is brilliant. It'll be interesting to compare the latency if you use a regular XB1 controller. And of course compared to consoles.
I wonder if this alludes to some sort of predictive modelling of the player inputs. That would seem like a Google thing to do.
Depends on where you measure it. Google measures it at their data center and says 'we sent you the output BEFORE your several possible inputs, hence negative latency'. For end user though, none of it matters because as you said, you can't have less than zero latency at your end.
'as a person with no professional experience in a completely unrelated field, here's my opinion'
LOL
I'm not just being an ass, I think half of all my posts on this forum have been about machine learning related things.
Speculative input will probably fool most people, but interpolating input will probably have it's own drawbacks just like frame interpolation or up-scaling.
I disable ASW on Oculus because of the downsides.
I disable ASW on Oculus because of the downsides.
Here is a direct excerpt from the Edge magazine. Here it seems that Madj Bakar is referring to 30fps games using a wireless controller. It might be good to update the OP with this as it is more accurate representation of what he said.
"But latency is the thing that gets the most attention. And while it's already proven to be more than playable, [Madj Bakar, VP of Engineering] expects further improvements. "Ultimately, we think, in a year or two, we'll have games that are running faster and feel more responsive in the cloud that they do locally, regardless of how powerful the local machine is," he claims. These improvements will come via a term which sounds rather slippery. "Negative latency" is a concept by which Stadia can set up a game with a buffer of predicted latency between the server and player, and then use various methods to undercut it. It can run the game at a super-fast framerate so it can act on player inputs earlier, or it can predict a player's button presses. These tricks can help the game feel more responsive, potentially more so than a console game running locally at 30fps with a wireless controller."
Great, I still don't want it.
Predicting my inputs is... not going to be the cure to latency, haha.
Predicting my inputs is... not going to be the cure to latency, haha.
That's a tall order.
That video (and the pilot episode of The Americans) has led me to discover a wonderful band who made a lot of wonderful records. :D
Its not rocket science and you are making false assumptions about me. I have plenty of experience with several internships. Oh and networking and transmissions lines is definitely a very electrical engineering centric topic. Maybe you should go read up on what electrical engineers actually do before opening your mouth.
Also, maybe you should go read up on propagation delay before you started ignoring the content of my post.
Imagine a world where technology attempted to solve these issues in creative ways for a better user experience...
I can't speak to what the quality will be when it releases. However, you're the best case scenario and even without lag mitigation I don't think you'll experience any noticeable input lag with any controller based game. <20ms is pretty close to imperceptible.
Here we go, some more data. 34 million gamers play 22 hours per week on average. That is equal to 88 hours a month and 1,386GB of data using the 4K service or 792GB on 1080P. What will end up happening of course is that the majority of hardcore gamers that will subscribe to Stadia wont be moving their full gameplay time to that single service and will probably divide their time on other platforms. This will gradually increase as the latency, image quality, game library and latency improve, as well as data caps. An interesting data point is that data caps are not as wide spread as I had in mind, with only 62% of the gamers reporting having one.
"First, we estimate that approximately six million out of the 34 million daily gamers would eat through their data caps if Google Stadia becomes their primary gaming destination."
"According to The NPD Group, 34 million gamers play 22 hours per week on average. If these individuals switched to using Stadia as their primary gaming platform, they would eat through even the highest data caps (usually around 1 TB, or 1,000 GB), coming in at roughly 1,386 GB monthly. It is important to note that this assumes that users only use their connections to game with Stadia; it doesn't account for other activities like TV and movie streaming, video chatting, and general browsing. It also assumes a one-user household."
"17 percent of the 943 respondents to our survey indicated that they had a data cap. A further 21 percent said that they didn't know whether their ISP had one or not."
I can't stop laughing. Family, roommates, parents, siblings, Netflix, music streaming, uploads, work activities and more are all apparently not on Stadia's marketing radar.
Did you read the rest of the article to see the actual % for every group? It has always been known that Stadia was not going to be for everybody day one. What we now have is a clearer picture of how it looks.
This is much more in line with what I thought regarding gamers using more data in general and how the average gamer's playtime would stack up against their data caps with Stadia use. No way are people playing 22 hours a week going to be able to transition to Stadia completely...but I'm willing to bet most people in that group aren't even interested in doing so in the first place for a wide variety of reasons. If you use Stadia along with other platforms it probably wont be a problem for you depending upon data use by others in your household, but unlikely that anyone will be able to transition to Stadia completely any time soon.Here we go, some more data. 34 million gamers play 22 hours per week on average. That is equal to 88 hours a month and 1,386GB of data using the 4K service or 792GB on 1080P. What will end up happening of course is that the majority of hardcore gamers that will subscribe to Stadia wont be moving their full gameplay time to that single service and will probably divide their time on other platforms. This will gradually increase as the latency, image quality, game library and latency improve, as well as data caps. An interesting data point is that data caps are not as wide spread as I had in mind, with only 62% of the gamers reporting having one.
"First, we estimate that approximately six million out of the 34 million daily gamers would eat through their data caps if Google Stadia becomes their primary gaming destination."
"According to The NPD Group, 34 million gamers play 22 hours per week on average. If these individuals switched to using Stadia as their primary gaming platform, they would eat through even the highest data caps (usually around 1 TB, or 1,000 GB), coming in at roughly 1,386 GB monthly. It is important to note that this assumes that users only use their connections to game with Stadia; it doesn't account for other activities like TV and movie streaming, video chatting, and general browsing. It also assumes a one-user household."
"17 percent of the 943 respondents to our survey indicated that they had a data cap. A further 21 percent said that they didn't know whether their ISP had one or not."
Thanks for the info.
I think that we can agree on this being a gradual change and very few people of that group that plays 22 hours every week will move all of their gameplay time over to Stadia if they do subscribe. A big factor for this will be the Stadia game library, as well as other factors that should be more important on average for this group, like image quality and latency. As all of these improves so will the % of that group that finds Stadia good enough.
I just don't see why you would ever buy a game on Stadia over, say, buying on PS5 or a gaming PC. You literally get less for the same amount of money.
I have been playing games since the original Nintendo and have owned one or two consoles for each generation after that, as well as gaming PC's. What has me more exited than everything for Stadia are the incredible things developers will be able to make once they put all of that hardware to work together. I also expect for the hardware refresh rate for services like Stadia to improve even faster than the half life console refresh we got this gen. This is all gradual of course and I also plan to buy a PS5 for the exclusives, though we still don't know how much will PS Now improve after they fully join forces with Microsoft's Azure. For all we know all PS5 games could be day one on PS Now as well.
You don't have to wait for your game's to download or worry about your hard drive space filling up
you can play your games over more devices
you will get to play at max graphical settings and 4K
you don't have to have any hardware with a loud fan that takes up shelf space and uses a lot of electricity
less is more
better question is why would I buy a multiplatform game on ps5? other than just so I can buy games physically and no I'm not worried about the internet being down where I can't play, I play too much anyway already so I won't mind if the servers are down for a few hrs if that ever happens
Last edited:
What other device are you gonna play Stadia on exactly? Your laptop? The display isn't going to be nearly as good as your television. Your phone? Why do that at home when you could use your TV, and if you aren't home then...your 4g isn't gonna give you good performance with stadia.
If you have a PC you will be playing at Max settings at 4k. With PS5 you will be as well.
You think the device you are playing on isn't gonna get hot and run a fan? If you are using a PC or Laptop...yeah it will. If you are using a phone...RIP your battery.
You may not have to download the game, but you don't own it. Say good bye to mods, say good bye to playing if internet goes out, say good bye to playing locally on a device in places where internet isn't good enough to support stadia (airplane, hotel, etc.).
The value just isn't there
Yea i can play at max on a PC for 700$ - 1,000$ no thanks
I don't care if the internet goes out for a few hrs, my internet hasn't gone out once this year
chromecast doesn't have a fan and uses basically no electricity
I'll use it on other devices like at friends houses, mobile devices where I have good WiFi or around the house etc, maybe even computer at work lol
I mean yeah if you are only gonna use it on mobile at your house, sure that will probably work...but that kind of defeats a lot of the benefits of playing on mobile and will also shred your battery. If none of those things I listed matter to you...great, you are one of the few people who Stadia sounds great for, if all the games you want come to it that is. For most people, it doesn't make sense, for all the reasons I listed and more.
That also reminded me, if my electricity ever goes out I won't be able to play PS5 but I'll still be able to play stadia on my phone through my battery backup router or possibly through cellular
I think my electricity has gone out more than my internet now that I think about it lol
You're not gonna be able to play Stadia with your cellular network.
Yea I have a battery back up router for security reasons since it's tied to the alarm system and everything I never thought about being able use stadia through that during a power outage
Isn't this about decoupling CPU processing and GPU render?
That way input latency will not be dependent on the FPS from the GPU render.
That way input latency will not be dependent on the FPS from the GPU render.