Slashdot Mirror
OnLive Latency Tested
The Digital Foundry blog has done an analysis of recently launched cloud gaming service OnLive, measuring latency across several different games. Quoting:
"In a best-case scenario, we counted 10 frames delay between button and response on-screen, giving a 150ms latency once the display's contribution to the measurement was removed. Unreal Tournament III worked pretty well in sustaining that response during gameplay. However, other tests were not so consistent, with DiRT 2 weighing in at 167ms-200ms while Assassin's Creed II operated at a wide range of between 150ms-216ms. ... OnLive says that the system works within 1000 miles of its datacenters on any broadband connection and recommends 5mbps or better. We gave OnLive the best possible ISP service we could find: Verizon FiOS, offering a direct fiber optic connection to the home. Latency was also reduced still further simply due to the masses of bandwidth FiOS offers compared to bog standard ADSL: in our case, 25mbps."
And with the bandwidth this service uses, you'll hit your ISPs "unlimited" cap in what, 6 hours? A day?
The creatures outside looked from Alt-Right to Antifa; but already it was impossible to say which was which.
As an early adopter (read: 1-year free trial ;) the service works fine (6Mb cable conneciton). For twich games you will notice a little sluggishness, but overall, its not difficult to adjust. Essentially, all the games play like a good latency online game. The only thing i'm not sure I like at the moment is the some of the minor artifacting you'll see due to the video compression. Again, this only really comes into play if you stop and look for it, during action you'll not notice it too much as you'll be busy paying attention to other things ;). Though right now, I cant say for sure how this service will perform in the future, as you apply for entrance into the service currently. Once anyone can join whenever they want, its hard to say how quickly OnLive will adjust to increased congestion.
I love to slaughter the english language.
In theory they are separate, but for example I upgraded my connection to 4x the previous bandwidth, and to my surprise the latency dropped down by half.
Almost all games have at least 4 frames of controller response lag, some games much more. At 60 fps, that's at least 67 ms of essentially unavoidable latency before the image even gets to the display.
It breaks down like this: one frame to read the controller, one frame to process the control, one frame to draw the response, one frame to display the buffered image.
TFA largely glosses over that fact, but they do link to a previous article that address this phenomenon. Here are some other ones.
Programming Responsiveness
Measuring Responsiveness in Video Games
Some examples from one of the above links:
+0 Meh
Your simplified view does not really account for reality:
- "Most unaware repeaters" are quite frequent on long-distance optical links. Do not ignore their influence!
- You will find that technologies like DSL may induce quite a bit of latency due to features like interleaving, this latency may overshadow anything mentioned in a) to c)
- The influence on first-hop bandwidth is usually marginal. With current bandwidth (assume 16Mbit/s) and usual packet length for games (assuming 200byte), this will give you 100 MICROseconds.
Also I just have to add my favorite quote regarding bandwidth and latency: Never underestimate the bandwidth of a truck full of tapes hurling down the highway (Andrew S. Tanenbaum)
OnLive seems to be a DRM pusher's wet dream:
1) You can't play without constant internet connection.
2) Can't trasnfer saved data to an offline version of the game.
3) You are renting the game and thus you own no physical copy of the game which you can resell or lend to others to use.