Google Stadia will be “faster and more responsive” than local gaming hardware in the near future predicts Stadia dev

[Alucardx23]

So at best you're going to be shaving off 16ms of local input latency. Relative to the network travel of your input to the servers, the processing and rendering, and getting that back to your screen, it's like a drop in the ocean. It's irrelevant. There won't be that much of a difference in input latency shit factor going from 166ms to 150ms, for example.

Click to shrink...

You are confusing framerate with input lag. If you increase the framerate from 30fps to 60fps, you will see a new image every 16.6ms instead of 33.3ms, but the game simulation and polling might be running at a higher rate than the game framerate. Below you can see how God of War goes from 138.7ms of input lag running at 30fps to 59.5ms when you unlock the framerate and it can go all the way up to 60fps.

https://www.youtube.com/watch?v=7grQXtsGSxQ

Also see this example from Counter Strike. Increasing the framerate from 60fps to 144fps, will lower the amount of time it takes to show a new frame by 9.7ms but the input lag will go down by 39.2ms.

[Sponsored] Beyond 60FPS: How Running Games at 144FPS/240FPS Can Improve The Gameplay Experience

Turning v-sync off and running games unlocked has always lowered input lag and made for tighter control in gaming - but now we have a range of 144Hz and 240H...

www.youtube.com

On the Killzone example below, you will see the framerate doubled, so increasing the framerate to 60fps will show a new frame 16.6ms faster than running at 30fps, but the input lag went down by 20.2ms.

 

[AtomicShroom]

You are confusing framerate with input lag. If you increase the framerate from 30fps to 60fps, you will see a new image every 16.6ms instead of 33.3ms, but the game simulation and polling might be running at a higher rate than the game framerate. Below you can see how God of War goes from 138.7ms of input lag running at 30fps to 59.5ms when you unlock the framerate and it can go all the way up to 60fps.

https://www.youtube.com/watch?v=7grQXtsGSxQ

Also see this example from Counter Strike. Increasing the framerate from 60fps to 144fps, will lower the amount of time it takes to show a new frame by 9.7ms but the input lag will go down by 39.2ms.

[Sponsored] Beyond 60FPS: How Running Games at 144FPS/240FPS Can Improve The Gameplay Experience

Turning v-sync off and running games unlocked has always lowered input lag and made for tighter control in gaming - but now we have a range of 144Hz and 240H...

www.youtube.com

On the Killzone example below, you will see the framerate doubled, so increasing the framerate to 60fps will show a new frame 16.6ms faster than running at 30fps, but the input lag went down by 20.2ms.

Click to shrink...

I know this. What I don't know is how people claim this will make the network latency not be utter garbage.

 

[BlueManifest]

I hope someone makes a video after stadia releases where they get random gamers and have 2 TVs set up

one is running stadia one is running a console and they don't know which is which and they ask them which has worse latency, I want to see the results

they would have to find a game that looks like each other though so the stadia one doesn't stand out in looking better

 

[Wamb0wneD]

Has someone already posted this?

They deleted the tweet... R* doesn't want that other talks about the game or maybe RDR2 won't run at 4K/60FPS on Stadia?

Click to shrink...

They don't want their PC release of RDR2 undermined lol.

Rockstar: RDR2 is coming to PC!
Google: But you don't need one tho.
Rockstar: Oh really now?
Google: Shit sorry!

 

[Alucardx23]

I know this. What I don't know is how people claim this will make the network latency not be utter garbage.

Click to shrink...

The God of War example shows that it went down by 79.2ms just by increasing the framerate from 30 to 60. So imagine this scenario, someone is playing God of War locally on a vanilla PS4 with 138.7ms of latency. Someone else is playing the PS4 Pro version of God of War on PS Now with 59.5ms of input lag, so 79.2ms less than the local version. If you add 50ms of network latency to that, it would increase the input latency to 129.2ms. Compare that to the 138.7ms of input latency of the local version being played on a vanilla PS4.

 

[Kieli]

The God of War example shows that it went down by 79.2ms just by increasing the framerate from 30 to 60. So imagine this scenario, someone is playing God of War locally on a vanilla PS4 with 138.7ms of latency. Someone else is playing the PS4 Pro version of God of War on PS Now with 59.5ms of input lag, so 79.2ms less than the local version. If you add 50ms of network latency to that, it would increase the input latency to 129.2ms. Compare that to the 138.7ms of input latency of the local version being played on a vanilla PS4.

Click to shrink...

Input lag can be more predictable. Ping/network lag can be all over the place. ISPs have built-in redundancy and mechanisms to get network packets to you despite network partitions (e.g. some hardware goes down), but you can't control that.

 

[Alucardx23]

Input lag can be more predictable. Ping/network lag can be all over the place. ISPs have built-in redundancy and mechanisms to get network packets to you despite network partitions (e.g. some hardware goes down), but you can't control that.

Click to shrink...

You can measure your ping to Google servers on the link below. You will see that the range it varies is not so much. Even on local gaming the input lag will vary based on how the engine works and how variable the framerate is. That's why you have to do multiple test to get an average.

GCPing.com

Measure latency to Google Cloud regions

www.gcping.com

https://www.youtube.com/watch?v=Ewj8GBboI7Q

The most important thing here is that it's not like we don't have real examples of the experience people can get with cloud gaming services. Look at the review below for the Xcloud beta.

"playing though Xcloud on a good Internet connection feels practically indistinguishable from local play. Diving in to Halo 5: Guardians on my phone (with a cloud save automatically downloaded from Xbox Live) felt like slipping into a familiar coat, with the same smooth frame rate and snappy controls you'd expect on an HDTV. While there may have been a frame or two of added delay between my inputs and the on-screen action compared to the local version, I'd be hard pressed to tell the difference in a blind test."

Project Xcloud preview serves as a passable, portable Xbox One

Small headaches are worth it to get a tiny Xbox on your phone.

arstechnica.com

 

[Tarot Deck]

"precog trickery"

Basically the servers will try to predict the player's inputs, and send multiple different results at once for the most likely possibilities. Then the input actually happens, the client chooses the output frame that matches best, and possibly does some light image manipulation if something is slightly different, like the direction or magnitude of an analog stick.

Microsoft experimented with something like that a while ago and apparently got great results. Technical paper on it:

https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/outatime_techreport2014.pdf

Graphs from the paper:

On the left, general opinion score from people playing. In the middle, player performance based on health remaining at the end of a segment. On the right, time taken for players to complete the segment.

Click to shrink...

That's... Cool and scary at the same time

 

[BlueManifest]

You can measure your ping to Google servers on the link below. You will see that the range it varies is not so much. Even on local gaming the input lag will vary based on how the engine works and how variable the framerate is. That's why you have to do multiple test to get an average.

GCPing.com

Measure latency to Google Cloud regions

www.gcping.com

https://www.youtube.com/watch?v=Ewj8GBboI7Q

The most important thing here is that it's not like we don't have real examples of the experience people can get with cloud gaming services. Look at the review below for the Xcloud beta.

"playing though Xcloud on a good Internet connection feels practically indistinguishable from local play. Diving in to Halo 5: Guardians on my phone (with a cloud save automatically downloaded from Xbox Live) felt like slipping into a familiar coat, with the same smooth frame rate and snappy controls you'd expect on an HDTV. While there may have been a frame or two of added delay between my inputs and the on-screen action compared to the local version, I'd be hard pressed to tell the difference in a blind test."

Project Xcloud preview serves as a passable, portable Xbox One

Small headaches are worth it to get a tiny Xbox on your phone.

arstechnica.com

Click to shrink...

Mine is 28ms what's yours, that website says it's not secure too for some reason

 

[Alucardx23]

Mine is 28ms

Click to shrink...

I envy you so much :D. The lowest I get is 60ms. If you had the PS4 Pro version of God of War running on that server with 59.5ms of input lag and add the 28ms of network latency + 10ms to capture and encode each frame, you would be getting 97.5ms of input lag. Compare that to the 138.7ms of input lag that someone playing the game locally on a PS4 vanilla would get.

 

[Kieli]

You can measure your ping to Google servers on the link below. You will see that the range it varies is not so much. Even on local gaming the input lag will vary based on how the engine works and how variable the framerate is. That's why you have to do multiple test to get an average.

GCPing.com

Measure latency to Google Cloud regions

www.gcping.com

https://www.youtube.com/watch?v=Ewj8GBboI7Q

The most important thing here is that it's not like we don't have real examples of the experience people can get with cloud gaming services. Look at the review below for the Xcloud beta.

"playing though Xcloud on a good Internet connection feels practically indistinguishable from local play. Diving in to Halo 5: Guardians on my phone (with a cloud save automatically downloaded from Xbox Live) felt like slipping into a familiar coat, with the same smooth frame rate and snappy controls you'd expect on an HDTV. While there may have been a frame or two of added delay between my inputs and the on-screen action compared to the local version, I'd be hard pressed to tell the difference in a blind test."

Project Xcloud preview serves as a passable, portable Xbox One

Small headaches are worth it to get a tiny Xbox on your phone.

arstechnica.com

Click to shrink...

I'm not disputing that in ideal circumstances, cloud gaming can offer a good experience. I guess where I have concerns is when claims are made that cloud gaming will be more responsive than local gaming hardware. Given the state of internet in NA, we are most certainly not there yet.

I'm stuck on a dinky 50MBps connection because my ISP has no competition in my neighourhood. I can only dream about 1GBps needed to drive amount of data needed for 60+ FPS, 1080p+ video.

 

[BlueManifest]

I envy you so much :D. The lowest I get is 60ms. If you had the PS4 Pro version of God of War running on that server with 59.5ms of input lag and add the 28ms of network latency + 10ms to capture and encode each frame, you would be getting 97.5ms of input lag. Compare that to the 138.7ms of input lag that someone playing the game locally on a PS4 vanilla would get.

Click to shrink...

Something I'm confused about is there appears to be a server in my state but a server 2 states away from me gets better ms

the one in my state is 33ms but the one further away from me is 28

 

[Alucardx23]

I'm not disputing that in ideal circumstances, cloud gaming can offer a good experience. I guess where I have concerns is when claims are made that cloud gaming will be more responsive than local gaming hardware. Given the state of internet in NA, we are most certainly not there yet.

I'm stuck on a dinky 50MBps connection because my ISP has no competition in my neighourhood. I can only dream about 1GBps needed to drive amount of data needed for 60+ FPS, 1080p+ video.

Click to shrink...

What you define as "ideal circumstances" will also increase over time and that will vary per person, because each person will have a different tolerance for input lag for what would they consider an enjoyable experience. So far we can only assume how negative latency will work, but I can share some information with you that should put you on the right track to understand how it could work.






The important information you should want to have is what amount of people in NA are "there" by their own definition, when it comes to have an enjoyable experience on a cloud gaming service. We need to wait for the new cloud gaming services to come out, so we can have a much better idea of how that looks like.

 

[Alucardx23]

Something I'm confused about is there appears to be a server in my state but a server 2 states away from me gets better ms

the one in my state is 33ms but the one further away from me is 28

Click to shrink...

The difference should be related to how many hops it's making and the equipment (servers, cables, etc) used at those locations. Server 1, while physically closer to you, makes more jumps (hops) and the servers it's going through appear to have slower responses.

https://en.wikipedia.org/wiki/Hop_(networking)

https://whatis.techtarget.com/definition/hop

 

[Alucardx23]

Something I'm confused about is there appears to be a server in my state but a server 2 states away from me gets better ms

the one in my state is 33ms but the one further away from me is 28

Click to shrink...

I tried doing a tracert command on CMD, to the Google server with the lowest ping, but I was not getting accurate results.

 

[Spikey]

Click to shrink...

What is with all these Observers?

 

[AtomicShroom]

The God of War example shows that it went down by 79.2ms just by increasing the framerate from 30 to 60. So imagine this scenario, someone is playing God of War locally on a vanilla PS4 with 138.7ms of latency. Someone else is playing the PS4 Pro version of God of War on PS Now with 59.5ms of input lag, so 79.2ms less than the local version. If you add 50ms of network latency to that, it would increase the input latency to 129.2ms. Compare that to the 138.7ms of input latency of the local version being played on a vanilla PS4.

Click to shrink...

You're assuming a lot of things. First you're assuming that the roundtrip will only add 50ms of latency. It could very well be 100ms or more. Second, even so this would only ever be "somewhat okay" for games with an already outrageously bad input latency to begin with. For games that already run at 60fps with minimal input latency of around 63ms, Stadia would simply never cut it.

 

[Alucardx23]

You're assuming a lot of things. First you're assuming that the roundtrip will only add 50ms of latency. It could very well be 100ms or more. Second, even so this would only ever be "somewhat okay" for games with an already outrageously bad input latency to begin with. For games that already run at 60fps with minimal input latency of around 63ms, Stadia would simply never cut it.

Click to shrink...

I'm not assuming anything, I'm telling you what the ping is on my example. There are people here that have reported pings that are even below 10ms to a cloud gaming service, so what is your problem with me using a 50ms ping as an example? It could be 300ms or more, but that wouldn't answer my example at all, the key here is understanding that cloud gaming services can offer a lower input latency and as more and more data centers open, more and more people will have lower latencies to them. On most cases, when you increase the framerate you will see the input lag going below the ms savings for the framerate and the sames applies to games that already have a relative low input lag. God of War on a PS4 vanilla has 138.7ms of latency and millions of people enjoyed that game like that, please focus on this line. This allows me to say that the same game running on a higher framerate on a server, would also provide a good experience to millions. Counter Strike running at 60fps locally with 79.3ms of input lag and at 144fps on the cloud with 79.3ms, could compensate for 39.2ms of network latency. The same game running at 240fps would have 29.2ms of input lag, you could add 30ms of network latency to that and it would still be less than the 63ms you just mentioned. This is what I always say about this type of discussions, people like to make a lot of black and white arguments that cover everything and that simply does not apply on every case.