Today, the Google Stadia streaming service officially launched for those who picked up the Founder or Premier Edition.
Well, sort of anyway. Some people have it, a lot of people don't, we certainly don't and it appears the team at Stadia give different answers to different people on when you will actually be able to access it. I've also seen plenty of people whose orders have been cancelled without warning or explanation. Even worse still, some people have been sent their hardware without an access code. Google have, so far, done a terrible job at communicating on Stadia and so the initial launch doesn't seem to have gone down well at all.
Oh, they also have the most ridiculous launch trailer I've ever seen:
Direct Link
The more I think about it, the more amazed I am at just how badly this has been managed. Take a look at the actual proper Stadia website for example, there's not a contact or support link in sight.
So they originally had a tiny lineup, then did a Reddit AMA where they said a bunch of the hooks they used to get people in weren't actually ready at all and hastily announced a few more games just before release. Nothing about it has so far looked like they've been in any way prepared to launch a gaming service.
One thing that I've seen confirmed now, as many suspected, is that input lag does seem to be a real problem. Google talked big about their powerful hardware and everything they were doing to bring it down, but it seems they haven't solved anything at all so far. Looking at the Eurogamer article, the input lag table included was quite impressive. This video from The Post also makes it look pretty awful.
From what Jason Schreier of Kotaku said on Twitter from "one person involved" that "preorders were below expectations". I really can't get my head around that. Somehow, they didn't get as many preorders as they had hoped and simultaneously failed to get them into the hands of people who did buy into it early. Words are honestly failing me right now. Incredible.
Eventually, at some point we will get access to it to report on how it works when played on a Linux desktop. When that is, I can't tell you, Google can't either. We're playing delivery bingo right now.
Article taken from GamingOnLinux.com.
Quoting: KetilI am looking forward to some input lag benchmarks. How many milliseconds does it take from you push a button until you get a frame that is affected by it? How much does the input lag vary? Under which conditions does the input lag get too high? I would also like to see some statistics of where the latency is occurring, but this probably requires cooperates from the stadia team.Honestly, I've no idea on how to actually accurately measure input lag. Been looking around, doesn't sound like it's even remotely easy to do. The best guides I've seen mention using a really good camera to show both the screen and key presses, which sadly the best I have is a webcam and it's not all that great either. We just don't have the funding to splash on stuff like that.
Last edited by Liam Dawe on 20 November 2019 at 12:13 pm UTC
Quoting: Liam DaweHonestly, I've no idea on how to actually accurately measure input lag. Been looking around, doesn't sound like it's even remotely easy to do. The best guides I've seen mention using a really good camera to show both the screen and key presses, which sadly the best I have is a webcam and it's not all that great either. We just don't have the funding to splash on stuff like that.
German magazine c't used a 1000 fps cam. I would guess you have some of those lying around? ;)
Quoting: EikeOh yeah, four collecting dust in.....a local store somewhere :PQuoting: Liam DaweHonestly, I've no idea on how to actually accurately measure input lag. Been looking around, doesn't sound like it's even remotely easy to do. The best guides I've seen mention using a really good camera to show both the screen and key presses, which sadly the best I have is a webcam and it's not all that great either. We just don't have the funding to splash on stuff like that.
German magazine c't used a 1000 fps cam. I would guess you have some of those lying around? ;)
Quoting: Liam DaweI didn't expect you to do these measurements, but I hope someone will do it and that I will learn about it. From the scientific side of this, I have found some articles about cloud gaming, but I still have questions they haven't answered yet. I have not done a comprehensive search though.Quoting: KetilI am looking forward to some input lag benchmarks. How many milliseconds does it take from you push a button until you get a frame that is affected by it? How much does the input lag vary? Under which conditions does the input lag get too high? I would also like to see some statistics of where the latency is occurring, but this probably requires cooperates from the stadia team.Honestly, I've no idea on how to actually accurately measure input lag. Been looking around, doesn't sound like it's even remotely easy to do. The best guides I've seen mention using a really good camera to show both the screen and key presses, which sadly the best I have is a webcam and it's not all that great either. We just don't have the funding to splash on stuff like that.
As for designing latency measuring experiments, I would look into a keylogger, measuring the time accurately, and record the desktop at the same time, with an accurate timestamp. It is probably a good idea to test both with and without keylogger and desktop recording to see if it feels different. I believe the extra latency added by e.g. obs would be predictable enough that we can compensate when analysing the results. At least within in error of about 2 ms. If I record obs with obs including a clock with microseconds, I see that the difference between the first and the second clock is about 32-34 milliseconds with a frame rendering time at around 7.5 ms at 30 FPS shown in obs stats. This shows that the previous clock is exactly one frame behind the current frame, and the timestamps match up when I step through the video.
Last edited by Ketil on 20 November 2019 at 1:56 pm UTC
Quoting: KetilI believe the extra latency added by e.g. obs would be predictable enough that we can compensate when analysing the results. At least within in error of about 2 ms. If I record obs with obs including a clock with microseconds, I see that the difference between the first and the second clock is about 32-34 milliseconds with a frame rendering time at around 7.5 ms at 30 FPS shown in obs stats. This shows that the previous clock is exactly one frame behind the current frame, and the timestamps match up when I step through the video.
From the few measurements that I read about, latency is estimated between 100-200 ms. So that amount of accuracy is much better than what is needed.
BrazilianGamer Nov 20, 2019
Quoting: Liam DaweWell that gives me a little hope. What connection speed? How far away was your Chromecast from your router?
My internet is gigabit, on Xfinity. I don't get the full gigabit speed, but it is still plenty fast (maybe 700Mb). I did have a similar experience during the beta on 250Mb internet.
My router is maybe 15 feet away or so from where I setup my Chromecast.
I also live in Chicago; I wonder if I'm close to an edge server, which might account for my low latency.
Quoting: GuestQuoting: BrazilianGamerThe trailer is cringe indeed
Oh I'm sure the trailer can't be that ba....what the heck was that thing?! This is why adverts shouldn't be made while on drugs.
When I saw it I thought "Why are such drugs only given to advertising clowns?!? :D
For instance, each meters adds a 3ms sound latency; and, when earing your own voice, a 15ms latency is noticeable.
That gives you a little more respect for the work of musicians when they try to work together onstage.
Patreon
PayPal
See more from me