Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

OnLive Latency Tested

Soulskill posted more than 4 years ago | from the in-the-case-of-ping-v.-pong dept.

Networking 204

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."

cancel ×

204 comments

Sorry! There are no comments related to the filter you selected.

Color me surprised (0)

Anonymous Coward | more than 4 years ago | (#32837362)

The end of another vaporware company. That is all.

re: (-1, Redundant)

f3rret (1776822) | more than 4 years ago | (#32837368)

I'm surprised, really I am.

Also: FP

Usage caps (5, Interesting)

KiloByte (825081) | more than 4 years ago | (#32837400)

And with the bandwidth this service uses, you'll hit your ISPs "unlimited" cap in what, 6 hours? A day?

Re:Usage caps (1)

RuBLed (995686) | more than 4 years ago | (#32837448)

I for one wants OnLive to succeed. It would be almost like how MMO games "provided" us DSL back then. (at least in our country) This would provide faster broadband services although ISP's could just jack the prices up for these "premium" services instead. The latter is what I fear would probably happen.

Re:Usage caps (1)

Ferzerp (83619) | more than 4 years ago | (#32838824)

And I want a flying car and a teleporter, but that doesn't meant I would go out and buy something that didn't work because I like the concept of what they claim they want to do.

Wishful thinking becomes dangerous to the point where you are tempted to sink money in to something that so obviously will not work.

We would all *love* the free energy pseudoscience crazies to actually make good on their claims, that doesn't mean it is a worthwile endeavour that we should waste money on.

First post! (0, Troll)

Anonymous Coward | more than 4 years ago | (#32837408)

Only took 137 ms.

"masses of bandwidth"? (1)

Eunuchswear (210685) | more than 4 years ago | (#32837416)

Latency was also reduced still further simply due to the masses of bandwidth FiOS offers compared to bog standard ADSL: in our case, 25mbps."

Maybe you want to look for a better ADSL provider. 25mbps is not much faster than a good ADSL2+ line.

Re:"masses of bandwidth"? (5, Informative)

_merlin (160982) | more than 4 years ago | (#32837436)

The statement is silly because latency isn't directly related to bandwidth. Switches, bridges, repeaters, modems, routers and other such devices all add latency. If FiOS reduces the number of these in the chain, the latency will be reduced. I'm not saying it necessarily does - just that it could provide better latency without having more bandwidth because of other factors.

Re:"masses of bandwidth"? (1, Insightful)

Anonymous Coward | more than 4 years ago | (#32837894)

I suppose there may be a connection if the bandwidth isn't enough to cover the constant streaming, but that'd mean the latency would slowly increase until it was unplayable; which obviously wasn't the case.

As you stated, it's probably a case of higher bandwidth needing better infrastructure; the better infrastructure providing an improved latency.

Re:"masses of bandwidth"? (4, Insightful)

KiloByte (825081) | more than 4 years ago | (#32837994)

Let's assume you have a hop where distance/c = 10ms and packet's length/bandwidth = 10ms. This means, the head of a packet arrives in 10ms, the tail in 20. No routers or bridges save for the most unaware repeaters will handle the packet until it arrives completely. Only then they will examine it and start sending it forward.

Thus, the final latency will be:
a) distance/c, plus
b) time spent in queues, plus
c) time needed for the bodies of packets to arrive.

To reduce a), you need to be closer to your destination. To reduce b), you need an ISP who oversells less. To reduce c), you need bandwidth on all hops.

Re:"masses of bandwidth"? (1, Interesting)

Anonymous Coward | more than 4 years ago | (#32838350)

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)

Re:"masses of bandwidth"? (3, Informative)

ledow (319597) | more than 4 years ago | (#32837576)

In the UK there aren't many options at all. Eurogamer.net is UK-based, hence the mention of BT.

My company don't want the expense of using leased line and other specialist stuff, just an ordinary thing that can work like a home package over an ordinary phone line. The FASTEST damn thing we can have is a single or multiple ADSL2 lines. We have basically unlimited funds for such things and often specify overkill-measures (i.e. 3 or 4 ADSL2 lines from seperate suppliers rather than 1 leased line). We get 20Mbps sync and a little less real-world speed. We are approximately 10 metres from the exchange. We are in an affluent and well-populated area of London.

In terms of what the average home user can have, only Virgin media fibre really beats the other offerings but that's highly variable and although you are told "up to 50MBps", the infrastructure isn't there to give you that as usable bandwidth.

To be honest, I'm impressed they managed to get what they did considering the state of UK broadband. Of course, you can pay stupid money and get serious pipes put in but that's hardly a "real world" scenario for the average home user. It's not unimaginable, though, that a true gamer might have the best a home user can be offered - which in the UK is a 25/50Mbps fibre service.

Re:"masses of bandwidth"? (1)

Eunuchswear (210685) | more than 4 years ago | (#32837658)

We get 20Mbps sync and a little less real-world speed.

So, like I said, your ADSL line is a little bit slower than his "masses of bandwidth" FiOS line at 25mbps.

Re:"masses of bandwidth"? (1)

Eunuchswear (210685) | more than 4 years ago | (#32837684)

In the UK there aren't many options at all. Eurogamer.net is UK-based, hence the mention of BT.

Get a lot of Verizon FiOS installations in the UK?

Re:"masses of bandwidth"? (2, Informative)

ledow (319597) | more than 4 years ago | (#32838030)

Nope. But some, obviously:

http://www.verizonbusiness.com/uk/products/internet/fios/ [verizonbusiness.com]

Were you trying to suggest that Verizon only do business in the US?

Re:"masses of bandwidth"? (1)

Eunuchswear (210685) | more than 4 years ago | (#32838134)

Well, yes, I was.

Didn't know they were selling that in the UK.

Re:"masses of bandwidth"? (1)

TheThiefMaster (992038) | more than 4 years ago | (#32838620)

The tests Eurogamer did were in the US, over Verizon FiOS (to give "OnLive the best possible ISP service we could find"). OnLive's not yet available in the UK.

It still sucked.

Re:"masses of bandwidth"? (2, Informative)

neokushan (932374) | more than 4 years ago | (#32837700)

I'd just like to add that I'm a Virgin Media customer with the 50meg connection he speaks of and I quite regularly max the hell out of it, at peak times. 6MB/s downloads are no problem.
However, he's right in that some areas have massive congestion problems and will suffer from issues, but unlike DSL, it wont have anything to do with how far away you are, if you're in a Virgin supplied area, you can get the 50meg.

Re:"masses of bandwidth"? (1)

Dexy (1751176) | more than 4 years ago | (#32838184)

Virgin have a horrible throttling policy [virginmedia.com] though, which, while it doesn't affect 50Mb customers, would make gameplay hell once the threshold is reached for anyone else.

Re:"masses of bandwidth"? (1)

mattsday (909414) | more than 4 years ago | (#32838240)

Agree here - I'm on the 50mbps service and usually see 40-45mbps without many issues.

Re:"masses of bandwidth"? (1)

GigaplexNZ (1233886) | more than 4 years ago | (#32838648)

My company don't want the expense of using leased line and other specialist stuff <snip> We have basically unlimited funds for such things and often specify overkill-measures

I don't follow.

Re:"masses of bandwidth"? - poster is inaccurate (1)

kangsterizer (1698322) | more than 4 years ago | (#32838642)

ADSL maximum is 8mbit over copper and 12mbit over ISDN.
ADSL2+ is 24mbit maximum

So, 25mbps is not standard ADSL, *it does not even exist.* Moreover, reaching ADSL2+ maximum theoretical speed (24mbit) is extremely unlikely. Most of us have a speed between 1mbit and 20mbit in most cases (depending on line quality, modulation type, line distance to central, etc).

Disclaimer: that's sync speed (IP bandwidth is therefore LOWER)

And that means...? (1)

iamhassi (659463) | more than 4 years ago | (#32837422)

Is this bad or livable? From what I recall of first person shooters a 150-200ms lag isn't bad, but your review just gives the raw numbers and never says if the games were still playable or not.

Re:And that means...? (1)

mykos (1627575) | more than 4 years ago | (#32837456)

That is just your input lag. Normally, you experience very little input lag (less than 5 ms) on your own computer. Then you have to add your network response time on top of that. You're looking at 1/3rd to 1/2 second before you actually respond ingame. Say goodbye to all multiplayer that isn't turn-based.

Re:And that means...? (-1, Troll)

QuantumG (50515) | more than 4 years ago | (#32837540)

Normally, you experience very little input lag (less than 5 ms) on your own computer.

And you measured this how?

Re:And that means...? (3, Insightful)

Eraesr (1629799) | more than 4 years ago | (#32837708)

Less than 5ms is nonsense (a simple framerate calculation , but Digital Foundry did quite a few input lag [eurogamer.net] tests [eurogamer.net] .

Anywhere from 60ms to over 100ms is common. Apparently gamers start to notice input lag at 166ms. Also, input lag and network lag shouldn't be confused with each other. The ping values you see in your game aren't 1-on-1 comparable to the input lag rates reported here.

To be honest, the 150ms input lag surprises me in a positive way. It's much lower than I had expected. For a game like UT3, 150ms is probably way too much and apparently that's one of the faster games, so OnLive's input latency is probably still too high for most games.

Re:And that means...? (0)

Anonymous Coward | more than 4 years ago | (#32837770)

Wrong. 5ms is faster than achieved, but >40 ms is very poor for gaming system (I see the article talks mostly about consoles).

Re:And that means...? (3, Insightful)

Eraesr (1629799) | more than 4 years ago | (#32837842)

It's not about consoles only. Your PC has the same problem. An average input lag of 5ms is impossible because when you take a game that renders at 60fps, every frame is roughly 16ms on the screen. If you push a button, then your action won't be visible until the next frame. So I'd say that at 60fps you'd have an average input lag of 8ms and that's not taking into account factors like game code processing the input and updating the gamestate or the lag caused by your LCD display.

(Older) PC games can be run at higher framerates because the hardware can handle it, so that might potentially decrease that 8ms average, but 5ms is only achieved with a 100fps framerate (when you assume 5ms as average, if you assume it's the slowest it'll ever be then you need 200fps).

Again, your LCD display has an inherent delay of 40 - 80 ms as well. The idea that 40ms input lag turns a game unplayable is a grave error. I mean, the article already points out that 60ms is basically as low as input lag on a non-LCD screen goes.

Re:And that means...? (2, Interesting)

Purity Of Essence (1007601) | more than 4 years ago | (#32838084)

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 [cowboyprogramming.com]
Measuring Responsiveness in Video Games [cowboyprogramming.com]

Some examples from one of the above links:

Games that run at 60 fps:

PS3 System menus: 3/60ths
Guitar Hero 3 (XBox 360): 3/60th
Ridge Racer 7: 4/60ths
Virtua Tennis: 4/60ths
Ninja Gaiden Sigma: 4/60ths
PixelJunk Racers 4/60ths

Games that run at 30 fps:

Genji: days of the Blade: 6/60ths
Tony Hawk's Proving Ground: 8/60ths
Blacksite: Area51: 8/60ths
Halo 3 (XBox 360) : 8-10/60ths
EA's "Skate": 10/60ths
GTA-IV: 10/60ths
Harry Potter: 10-14/60ths
Heavenly Sword: 7-18/60ths

Re:And that means...? (0, Troll)

B1oodAnge1 (1485419) | more than 4 years ago | (#32838692)

(Older) PC games can be run at higher framerates because the hardware can handle it, so that might potentially decrease that 8ms average, but 5ms is only achieved with a 100fps framerate (when you assume 5ms as average, if you assume it's the slowest it'll ever be then you need 200fps).

You don't seem to realize that if an average PC gamer is maxing out at 60 fps then he is complaining constantly about the lag and making plans to build a new system.

60fps is bare minimum for a decent experience in a non multiplayer game, and unplayable against other humans.

Re:And that means...? (0)

Anonymous Coward | more than 4 years ago | (#32838804)

With a stopwatch.

Re:And that means...? (5, Insightful)

f3rret (1776822) | more than 4 years ago | (#32837478)

150-200 ms latency in a modern FPS is nearly unforgivable. I played TF2 a lot for a while and if I ever had more than 40-70 ms latency the hit detection would start to suffer and you'd get shot through walls or just not hit.

I expect a system like OnLive might work better with strategy games and other types of games that are not nearly as fast paced as most modern shooters are.

Re:And that means...? (1)

The MAZZTer (911996) | more than 4 years ago | (#32837552)

Actually TF2 isn't bad compared to other games. Source has lag compensation that actually tries to adjust the game temporally based on a player's lag... in short, it figures out where everyone was from that player's POV at the moment they shot and uses that to figure out if the player hit with his shot, etc. I've played online games where even 70-100ms ping is unplayable (I'm looking at you, D.I.P.R.I.P.) but TF2 can remain smooth at 150, and playable at 200-300ms (though the lag becomes noticeable then as people can more easily seem to hit you around corners... in actuality they're hitting you where you were 300ms ago).

On the other hand, unless you compile code into the game itself to compensate for input lag in a similar fashion, I expect any game would be similarly frustrating over 50ms lag or so.

Re:And that means...? (1)

alexhard (778254) | more than 4 years ago | (#32837630)

I think you're overstating the difference the lag compensation of tf2 makes. It kinda makes the 70-100ms interval playable, but anything above that is still very noticeable and extremely frustrating unless you're playing a camping engie.

Re:And that means...? (1)

Ruede (824831) | more than 4 years ago | (#32838846)

yeah and high ping bastards have an advantage....
great... that is why i hate the source netcode.

anywhere around the world ppl started to get better internet connections.... (56k died out.) but valve decided to make the game playable for those suckers that had to keep a 56k modem.

Re:And that means...? (1)

SpazmodeusG (1334705) | more than 4 years ago | (#32837588)

The latency from where the game is running to the multiplayer game server will be very low. So you won't get bugs that are normally attributed to that sort of thing. As far as the games concerned you'll have a terrific ping to nearby game servers.

The input and display lag isn't even knowable by the game for the remote desktop trickery they are doing. Instead you get a game running perfectly smoothly on the datacenters computer with your inputs being completely out of sync with the display.

Re:And that means...? (1)

KovaaK (1347019) | more than 4 years ago | (#32837610)

Keep in mind that there's a difference between input latency (which this story is talking about) and multiplayer network latency that you are talking about. In the case of multiplayer games, there is generally enough code on the client-side to predict what your screen will look like so that it feels snappy to you. Regarding getting hit through walls, that happens because the server doesn't think that you are behind the corner yet, so you are still a viable target to your opponent. The original quake didn't have any of the movement prediction, so there was movement input lag when you played online - thus when you thought you were behind cover, you really were. It still let you move your mouse without waiting for the server to agree that you were facing the new direction (every modern FPS does this as well as let the client predict movement before waiting for the server to agree). OnLive, on the other hand, will affect both movement input and mouse control input with this increased latency.

Basically, a good LCD HDTV can add 40-80ms of input lag, and hardcore gamers such as myself complain about that (justifiably so). OnLive adds this same kind of latency. If it really is 150-200ms before multiplayer network latency, it is indeed useless for playing multiplayer games online.

Re:And that means...? (0)

Anonymous Coward | more than 4 years ago | (#32837676)

"hardcore gamers as myself". Get over yourself, you are a dying market.

Re:And that means...? (0)

Anonymous Coward | more than 4 years ago | (#32837848)

Why do you assume that by using the label "hardcore gamer" that he was elevating himself?

Seems to me like you need to chill out.

Re:And that means...? (1)

Razalhague (1497249) | more than 4 years ago | (#32837802)

Basically, a good LCD HDTV can add 40-80ms of input lag, and hardcore gamers such as myself complain about that (justifiably so). OnLive adds this same kind of latency. If it really is 150-200ms before multiplayer network latency, it is indeed useless for playing multiplayer games online.

I wouldn't be at all surprised if OnLive started putting servers for multiplayer games right in their datacenters, giving you very little network lag to compensate for the input lag.

Re:And that means...? (0)

Anonymous Coward | more than 4 years ago | (#32837500)

in the client and server open source game Sauerbraten [sauerbraten.org] , you can set up a server and play across europe in 50-90 ms. If the game doesn't trust much the client to avoid cheats more network load and lag than games like sauerbraten is inevitable but I'd rather risk cheaters and play smooth. YMMV

Re:And that means...? (1)

Xest (935314) | more than 4 years ago | (#32837594)

Put it this way, 150 - 200ms is worse than I used to get playing Quake 1 on a 33.6kbps dialup modem.

I used to achieve around 120 - 140 on dialup to UK servers. ISDN was around 60 - 80ms, ADSL around 20 - 40ms.

So in other words it sounds terrible in this day and age. Almost certainly bad enough to make games like Guitar Hero or other games that need rapid responses completely unplayable on this kind of service if they ever tried to offer it.

Re:And that means...? (2, Funny)

Krneki (1192201) | more than 4 years ago | (#32837652)

It is bad, because we are talking about input lag, not response lag. And while it might be OK for Joe the console player, but it is unacceptable for competitive PC players, who tweak every single input device in order to lover lag.

Re:And that means...? (0)

Anonymous Coward | more than 4 years ago | (#32838062)

Joe the console player is the target audience.

No, Joe the very light console player is the target audience.

Anyone else would get a better deal buying the system and games out right.

A $100 graphics card will beat this... (2, Informative)

Joce640k (829181) | more than 4 years ago | (#32838440)

No fast action game will work with that latency - the graphics might be smooth but the input response is like playing at five frames per second.

There'll be some games which work in this format but they won't be first person shooters or driving games - think flash games but multiplayer and in 3D.

Is it worth subscribing and being nickle-and-dimed for every minute you're on there instead of playing all the free flash games on the web? That's what they're betting the company on.

Re:And that means...? (1)

Bakkster (1529253) | more than 4 years ago | (#32838152)

And while it might be OK for Joe the console player, but it is unacceptable for competitive PC players, who tweak every single input device in order to lover lag.

Isn't this service intended for 'Joe the console player', and not for competitive PC gamers? In other words, it's an entirely reasonable trade-off.

I think the issue here is all the 'hardcore' gamers who worry about this kind of stuff were never the target market, because they're 'hardcore' enough to put tons of money into their own rig. The recreational player just wants the cheapest method that's acceptable, and that's what the service is aimed for. Of course, it's mostly the hardcore gamer who posts on forums, not the recreational ones, so it's nothing but hate in the internet echo chamber.

Re:And that means...? (1)

twistedsymphony (956982) | more than 4 years ago | (#32837988)

Personally I think the lag is unacceptable unless approved by Korean Starcraft Players.

Re:And that means...? (1)

Yvanhoe (564877) | more than 4 years ago | (#32838090)

Input lag of 150-200 ms is simply not playable. Remember that a laggy server hide the latency through interpolation and let's you have a short latency for your input.

200 ms is the kind of reaction time you can have at 5 fps. Of course their image goes at 30 or 60 fps, but if you want to see what a lag of 200 ms represent, witness how difficult it is to play at 5 fps.

Re:And that means...? (0)

Anonymous Coward | more than 4 years ago | (#32838608)

To me, this is unplayable. It isn't just lag in the reaction of objects in the world around you, it's lag in the reaction of your input including menu interfaces. To get an idea, you'd have to play a game at 200 ping that does not have any client side prediction at all. The original Quake (not QuakeWorld) is a great example of this. It was considered almost completely unplayable at anything higher than 150 ping by most players. You literally could not successfully jump across some gaps to get powerups/weapons.

The reason people say FPS games work fine at 150-200 ms these days is because those FPS games have massive client side prediction. The client just pretends that your actions are instant until it gets a message from the server that contradicts your actions, and then performs some action to resolve it. Generally the server will also look back in time a short amount to determine if your actions 200ms in the past were valid based on the game state at that time instead of the current game state.

And there's pretty much no way they can resolve this lack of client side prediction because the entire point of the service is *not* to have all this data or the executable on the client. It'll probably be fine for RPGs and slow paced RTS games or strategy games, but when it comes to shooters or platformers this amount of lag is completely unacceptable and unplayable except by the most casual of players or the simplest of games.

Carmack gave a speech a while ago where he mentioned OnLive and said that the only reason he thought some games would be playable was because their engines already handled input poorly and had 50-100ms of internal lag that would help mask the effect.

Impressive (1)

mrsurb (1484303) | more than 4 years ago | (#32837432)

Getting that sort of latency within "1000 miles of its datacenters" is quite impressive.

Re:Impressive (0)

Anonymous Coward | more than 4 years ago | (#32837704)

I agree. But "impressive" does not necessarily imply "good enough".

Works Just Fine (5, Interesting)

Zediker (885207) | more than 4 years ago | (#32837468)

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.

Re:Works Just Fine (0)

Anonymous Coward | more than 4 years ago | (#32837584)

Then again, I'd be hard-pressed to bad-mouth anything that I've gotten a one-year free trial on.

Re:Works Just Fine (1)

Zediker (885207) | more than 4 years ago | (#32837636)

well, nothings stoping anyone else from signing up and getting a 1-year free trial as well. ;)

Re:Works Just Fine (0)

Anonymous Coward | more than 4 years ago | (#32837912)

Nothing except my integrity and burning desire to see OnLive die a death so horrible that its investors are bankrupted and nobody ever tries this bullshit again.

Re:Works Just Fine (1)

MattSausage (940218) | more than 4 years ago | (#32838650)

Ummm.. are you serious?

Why is Onlive such a dangerous thing for you to hate it so much? You prefer upgrading your PC every 6 months?

Re:Works Just Fine (1)

Krahar (1655029) | more than 4 years ago | (#32837622)

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.

What would be cool is if they could track your eyes using a webcam and increase the quality at the part of the screen you are looking at. Won't help if your eyes are twitching all over the place, but if there is anything you want to see in full quality, just look at it during the latency of those 230 or so milliseconds and the artifacts will disappear where you are looking.

Best part of OnLive - instademos (1)

D J Horn (1561451) | more than 4 years ago | (#32837626)

I too got in on the 1 year free trial to check it out. My connection is pretty average, 7meg verizon DSL, but being in the middle of nowhere I ping 150~200 in every game I've ever played, so I had pretty low expectations.

I was very pleasantly surprised when I gave it a shot. There is definitely noticeable latency, but I only really feel it when moving the mouse cursor around. Button based actions seem fine. I probably wouldn't play a twitchy FPS on it, but just about any other game doesn't feel strange at all. Playing Arkham Asylum with a gamepad feels great.

I have no intention of buying any games for it though - my only machine is a nice desktop so I have no need to. That being said, I am loving OnLive for the ability to launch it up and instantly play a demo of any game in there. No downloading, no installing, no waiting AT ALL. Not only that but the demos aren't specially packaged portions of the game - they are simply 30 minutes of access to the full retail game. This has already led me to make several purchases that I was on the fence about.

So in the end, having seen it for myself, I think OnLive is pretty cool and does have its uses. It's certainly not going to replace PC gaming as we know it, but I think this cloud based tomfoolery has a place in our future.

Re:Works Just Fine (1)

should_be_linear (779431) | more than 4 years ago | (#32837898)

And it will only get better: with increasing number of cloud-gamers, ISPs will start deploying OnLive servers on sites close to customers, decreasing latency dramatically. Another point that might improve in time is HW latency (modem / ISP routers, ...) which might be improved now that it is important for some BFUs and finally, OnLive might end up being property of Google, therefore spread much faster to remote locations near potential customers.

Re:Works Just Fine (2, Interesting)

gnieboer (1272482) | more than 4 years ago | (#32838192)

Also an early adopter, and I've found it varies widely by game. DiRT 2 was unplayable with that lag.

Some of the games worked fine, and IMHO the best thing it's got going for it is the ability to instantly play 30 minute demos of any game they've got, no need to install/uninstall more stuff on the home machine just to see if a game is worth it.

I also got kicked out several times due to "network issues" one night that was very frustrating (despite being on a reliably 16mbps connection->gigabit LAN). I think those factors, if not addressed, will prevent common user adoption (Win7 decides to background download some new service pack and hogs too much bandwidth and you're done with no understanding of why).

Bandwidth != Latency (1, Insightful)

Anonymous Coward | more than 4 years ago | (#32837470)

Bandwidth and latency are not the same thing. Increasing the bandwidth to 25Mbps will not help latency at all.

Re:Bandwidth != Latency (1, Interesting)

Anonymous Coward | more than 4 years ago | (#32837618)

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.

Re:Bandwidth != Latency (1)

icebraining (1313345) | more than 4 years ago | (#32838136)

Wrong. Latency is affected by the throughput, because it's the sum of the time it takes to be transmitted plus the time for that data to traverse the network equipment between the nodes.

Where precision is important, one-way latency for a link can be more strictly defined as the time from the start of packet transmission to the start of packet reception. The time from the start of packet transmission to the end of packet transmission at the near end is measured separately and called serialization delay. This definition of latency depends on the throughput of the link and the size of the packet, and is the time required by the system to signal the full packet to the wire.

Besides, if the link is congested, queuing delays will occur, as the packets will have to wait for the previous to be sent. This is important for OnLive, as it uses plenty of bandwidth to transmit the video feed.

Reducing network congestion by getting a "fatter pipe" always help reduce latency.

Head - Desk... (3, Insightful)

fuzzyfuzzyfungus (1223518) | more than 4 years ago | (#32837476)

Latency was also reduced still further simply due to the masses of bandwidth FiOS offers compared to bog standard ADSL: in our case, 25mbps.

Damn it, kids, Latency and bandwidth are not the same thing and anybody who makes that mistake should be forced to use a "1Gb/s" connection via fedex.

Yes, in the case of something like OnLive, which is basically streaming mouse/keyboard events one way and video the other, things will look substantially worse if frame N hasn't finished downloading by the time frame N+1 is ready for transfer(and then either has to be dropped, or delays frame N+1 even more than your connection's latency would); but having a fat pipe does not "reduce your latency". It is correct to say that 25mb/s FIOS is probably about the most generous test that is also remotely realistic for more than a tiny number of their potential customers; but the bandwidth thereof does not "reduce latency"...

Re:Head - Desk... (5, Informative)

JustinRLynn (831164) | more than 4 years ago | (#32837516)

Yep, you're absolutely right in that bandwidth and latency aren't the same. However, when used by TCP in latency sensitive environments, common asymmetric connections can quickly saturate their available upstream bandwidth. This means that they're not able to ACK incoming packets, effectively increasing their link latency and reducing its throughput. So, in reality, total throughput is a combination of link latency and the ability to quickly respond to the protocol stream to keep the bits flowing. This is why QoS for TCP is so important on heavily utilised asymmetric connections.

Re:Head - Desk... (1)

qbnstephen (1851328) | more than 4 years ago | (#32837598)

having a fat pipe does not "reduce your latency". It is correct to say that 25mb/s FIOS is probably about the most generous test that is also remotely realistic for more than a tiny number of their potential customers; but the bandwidth thereof does not "reduce latency"...

It's great to see somebody has their head screwed on right today.

Re:Head - Desk... (1)

Krahar (1655029) | more than 4 years ago | (#32837664)

Bandwidth can reduce latency all other things being equal. If you need to send a packet of 5kb then the latency of that packet will be lots of things that are independent of the bandwidth, but it also does include the time it takes to transfer the data in the packet itself over the wire, and higher bandwidth reduces that time. Not that it is going to be a lot of time for a 5 kb packet, but there can indeed be an impact of bandwidth to latency.

Re:Head - Desk... (1)

Krahar (1655029) | more than 4 years ago | (#32837716)

Actually I'm guessing that video frames are a lot larger than 5kb. If they are e.g. 1 mb then the time to transfer may even be more than the latency of the connection itself, depending on your bandwidth. So the latency as measured by the time from pressing a key and the man on your screen doing something is most certainly impacted by bandwidth in many cases. This effect is not limited to when the game can't finish sending a frame before the next one is ready, because even when that doesn't happen you still have to wait to receive the frame before it can be displayed.

Re:Head - Desk... (0)

Anonymous Coward | more than 4 years ago | (#32838390)

Just for fun: 0.2s * 25mb/s / (8 bit/Byte)= 625kB.

The frames can not be larger than that. Actually, they probably are half of that or so...

Re:Head - Desk... (1)

MichaelSmith (789609) | more than 4 years ago | (#32837694)

Latency was also reduced still further simply due to the masses of bandwidth FiOS offers compared to bog standard ADSL: in our case, 25mbps.

Damn it, kids, Latency and bandwidth are not the same thing and anybody who makes that mistake should be forced to use a "1Gb/s" connection via fedex.

Its not just the kids. I had the corporate IT department tell me they would fix their latency issues by compressing the link, and if that didn't work they would put another compressor in series with the original one.

Re:Head - Desk... (1)

fuzzyfuzzyfungus (1223518) | more than 4 years ago | (#32837852)

If people would just stop using large font sizes in emails, their compressor definitely would have done the trick...

Re:Head - Desk... (3, Informative)

drinkypoo (153816) | more than 4 years ago | (#32838160)

Damn it, kids, Latency and bandwidth are not the same thing and anybody who makes that mistake should be forced to use a "1Gb/s" connection via fedex.

A highly saturated connection will, in practice, have higher latency. Therefore, bandwidth and latency are related. HTH, HAND.

Re:Head - Desk... (1)

icebraining (1313345) | more than 4 years ago | (#32838362)

Latency = transmit time + time to transverse the wire.

Throughput ("having a fat pipe") does affect the first factor. Of course, if the second factor is multiple orders of magnitude larger than the first (like with your Fedex example), reducing the first will be irrelevant, but that's always true with any optimization.

But that's not the usual case in a normal internet connection, and improving the first factor can and does improve the latency visibly.

Not too shabby (1)

kinnell (607819) | more than 4 years ago | (#32837502)

I've played WOW with that kind of latency at times, and while it's not ideal, it's definitely playable without becoming frustrated. It may not work well for some games, especially for people who are serious about PVP, but it seems like a reasonable service. Considering the technology they're using, I'd say they're doing pretty well. As a mac user, I'd give it serious consideration.

Oh For Fuck Sake (1)

QuantumG (50515) | more than 4 years ago | (#32837596)

How many people on this thread are going to confuse network latency with input latency... hint: you have no experience that prepares you to understand these numbers.. just play the god damn game and quit the service if you don't enjoy it.

That's why they're handing out 12 month trials.

Re:Oh For Fuck Sake (2, Insightful)

SharpFang (651121) | more than 4 years ago | (#32837836)

yeah... that's an essential problem.

Network latency: You aim, almost immediately crosshair covers enemy head, you shoot, with bad lag the server will inform you you have missed, the enemy was not there.

Input latency: You aim. It takes 150ms for the crosshair to start following your aim. You finally get to aim at the enemy's head and click. The enemy moves, you move to follow, but since your reaction is delayed by 150ms it's now that your shot (and miss) and you will start following the enemy in 150ms.

ARGH. Totally unplayable.

Re:Oh For Fuck Sake (1)

dtml-try MyNick (453562) | more than 4 years ago | (#32838644)

Anoying yes, unplayable no.

When I started playing fps's online, UT classic mostly, 150ms latency was pretty much the norm. Anything lower was considered a extremely good connection.

Most players just compensated for the lag by aiming slightly off.
Predict / anticipate the direction your opponent is going and adjust you aim according to that direction and the latency. Problem fixed.

Re:Oh For Fuck Sake (0)

Anonymous Coward | more than 4 years ago | (#32838770)

Did you even read the fucking post you just responded to? YOU'RE TALKING ABOUT NETWORK LATENCY. Input latency is not the same experience.

Stupid (2, Insightful)

ledow (319597) | more than 4 years ago | (#32837508)

You had a good business model. A lot of people would be happy to play games that can be played with lag without noticing (I spent hours on Puzzler World, Max and the Magic Marker, Crayon Physics, World of Goo, Age of Booty, all sorts of games that aren't that affected by lag). You could easily have had a Wii-like console in every home that delivered as powerful a game as necessary, against as many players as necessary while needing no fancy installation, discs, etc. and most importantly NEVER needing an upgrade. Specifically, I would compare the system to those arcade machines that let you play, say, 20 minutes of Super Mario World or some other Nintendo games. You pay a flat fee and can swap between games as much as you like during that time without having to install demos, or buy them all. Brilliant idea.

Instead you didn't listen to the only criticism of the idea (enormous lag is inevitable - yeh cannae break the lawsa fisics...), wouldn't heed it, denied there was any problem, etc. and thus in the first, purportedly "ideal" real-world test, your founder's press statements were found to be orders-of-magnitudes out. As such, you've killed the interest from people who *knew* that all along and who would be asked their opinions on it by other people. If you'd just said "the affect won't ruin the majority of games", or "the latency isn't something we can do anything about but we don't expect it to affect the titles we offer, and the kind of customers we're aiming at", then nobody would have cared and if their granny bought the system they would have played on it too. But the stupid claims did not hold up and, thus, we're waiting to discover what the next lie is... *do* you have an accord with BT to get onto the UK broadband backbone? Do you have top-name titles properly licensed and ready-to-go? Do you have the capability to scale the service with the number of users? Do you have the hardware ready? Do you have something that you can sell if the system was to go live as quickly as possible?

You spoiled your image with bullshit. On an ideal test, a quite basic but fast-paced game that plays well locally gets up to 250ms of lag. Optimised or not, ideal conditions or not, that's just never going to sit well with people, even if they have a 60ms lag on their TFT monitors and don't realise it (http://www.tftcentral.co.uk/images/input_lag_graph.jpg). All I see is the "250ms" and think - damn - when I play CS online I think of anything over 80ms as "laggy". And that's just a one-way property, my lag to the server. God knows how a server performs when ALL players have a few hundred milliseconds of lag. I think 90% of your CPU time in that case must be input smoothing and path prediction.

It's just a pity that your failure to be honest will tar the rest of your business' life and that of any similar systems that might arise in the future.

Re:Stupid (1)

Velex (120469) | more than 4 years ago | (#32837558)

I don't think you understand how capitalism works. It's the engineers who are trying to make this happen who are the incompetent dumbshits, and so stupid and uneducated compared to these CEOs. The CEOs are geniuses who are only hampered by how illiterate their engineers are.

Pointless (0)

Anonymous Coward | more than 4 years ago | (#32837514)

I still don't get what OnLive offers to the consumer instead of just catering to industry participants. They seem to only be trying to push "See? We're on par with what we have currently!" instead of "We're actually offering something more to the consumer than existing solutions". The whole low-end devices argument is crap, wireless is going to be slow and unreliable, pretty much making OnLive irrelevant for small devices. Having it on today's computer seems kinda pointless, leaving only set-top box type solutions which seems fairly niche' in the first instance seeing as you're only going to get these latency numbers in certain areas anyway, that combined with existing solutions/competitors makes this failware.

Re:Pointless (2, Insightful)

SharpFang (651121) | more than 4 years ago | (#32837740)

Cheap, dead-simple "Game rental", buffet-style. Pay per hour, not per game.

You can play any of hundreds games, now. No purchase, no download, no install, no cracking, no registering, easier than torrents. You just start a game and play it. And if you don't like it, switch it off and play another, you lost maybe half a dollar trying it out, not fifty bucks at a store, not thirty bucks and three hours downloading and installing from Steam, not three hours downloading and installing from piratebay. You just have it as if it was already installed on your PC, all included in rental fee.

Re:Pointless (1)

fuzzyfuzzyfungus (1223518) | more than 4 years ago | (#32837838)

What I don't really understand(though what might be part of why they don't seem to offer anything all that compelling to the customer) is how the economics of their operation stack up.

With something like storage, or web hosting, there are fairly large and obvious gains to centralization and specialization. In the case of storage, the economics of not having a local copy are stupid, when a 1TB drive is $99; but unless your data are super extra secret, especially enormous, or you are atypically skilled, getting backup by buying 1 1 millionth of some professional data warehouse's capacity and sending incrementals over the internet is clearly sensible. Web hosting even more so. A tiny slice of some gigantic datacenter with multiple redundant enormous pipes is way cheaper than replicating that at home.

Video games, though, don't seem to offer the same arguments. Unlike, say, making good backups, which is a comparatively rare skill, console gaming is pretty seriously accessible. Worst case, you bribe some local 16 year old with a six pack to make it work. So there is no argument from skill specialization. Consoles have already carved out a "gaming appliance" market, and(particularly for casual games, flash games, and the like) using the PC that already needs to be working for internet access and word processing for a bit of gaming on the side isn't much harder.

There is also no efficiency of scale argument: In something like storage, a fancy data-deduplicating, zippy-special-compression, etc. storage setup, as available to the pros, can handle rather more customers than a simple comparison of its total capacity to the sum of all the customer's data would suggest. Most home-user data are either in the form of unique; but fairly tiny and compressible, stuff like text documents, or fairly large; but far from unique, things like downloaded movies and songs. With games, though, that effect is much smaller, if present at all. Console games are generally designed to be at the limits of their hardware, since those limits are fixed. You might, if buying in gigantic bulk, convince the console maker to provide you with their console in "processor card" format, allowing you to aggregate things like PSUs and mass storage; but, even in that ideal world, you are still buying as much silicon as joe gamer.

Games are also relatively "bursty" which is bad. Because of latency/speed of light issues, you cannot aggregate demand across the globe, or even across more than a handful of time zones. So, everyone your datacenter serves will be on almost the same schedule. During peak hours, like shortly after kids get out of school, you'll need to be able to support almost as many instances as you have customers. During off-hours, you'll just have a few inverted odd-shift workers and the like. Unlike batch number crunching, an hour starting at 4am is nearly worthless to most of your customers.

Then, of course, you come to the fact that "cloud gaming" inevitably incurs certain additional costs: bandwidth and video compression hardware. You'll see some bandwidth savings, since none of your customers will be downloading game or demo binaries from you, and because you will be able to keep multiplayer games, in some cases, occuring between multiple users within your datacenter; but the fact that you are sending 720p video constantly, to each one of them, will erode that pretty quickly. You also have to pay for, and power, whatever silicon is pumping out that video.

This, I'm assuming, is why OnLive is charging a subscription fee just for the right to show up and buy stuff, and why the games they are selling access to are not, generally, forecast to be much of a discount over their retail counterparts.

Bandwidth != Latency (1)

msauve (701917) | more than 4 years ago | (#32837538)

Latency was also reduced still further simply due to the masses of bandwidth FiOS offers compared to bog standard ADSL: in our case, 25mbps.

No, it wasn't (at least not significantly). The difference in latency between a 1 mbps link (ADSL upstream) and a 25 mbps link (which is due to serialization delay) is about 12 ms for large (1500 byte) packets. Since the vast majority of packets sent in this sort of application would be small ones (having only to convey simple info like "button 1 pressed"), it would actually be well under a millisecond. Compared to the measured results, this is insignificant.

Re:Bandwidth != Latency (3, Informative)

SpazmodeusG (1334705) | more than 4 years ago | (#32837682)

Remember they don't just send the inputs there. They have to get the display back again.
If each frame is 100Kilobytes and they need 30fps to look smooth that's approaching the limit of 25Megabits/s (=3.125Megabytes/s).

Re:Bandwidth != Latency (1)

msauve (701917) | more than 4 years ago | (#32837960)

It doesn't work the way you think it works.

From Onlive's site [onlive.com] : "OnLive recommends a wired 5 Mbps connection to the Internet..." They haven't released any technical info on their proprietary video compressor, so it's not clear where your numbers come from.

In any case, even if the full difference in serialization delay is considered (~12 ms), that is minor in comparison to the measurements.

Re:Bandwidth != Latency (1)

msauve (701917) | more than 4 years ago | (#32838636)

I'll just add that the video is on the downstream channel, where the difference between ADSL and FiOS is 12:25 mbps. That's a difference of about 0.5 ms in latency for large packets. I was talking about the upstream channel, where game inputs would be sent. These are the small packets, so still under 1 ms.

Re:Bandwidth != Latency (0)

Anonymous Coward | more than 4 years ago | (#32837776)

Wouldn't the majority of the packets be the video frames that OnLive sends as a response to your button presses? Those are quite large...

Re:Bandwidth != Latency (1)

Krahar (1655029) | more than 4 years ago | (#32837788)

They also have to send you back frames of video.

Re:Bandwidth != Latency (1)

msauve (701917) | more than 4 years ago | (#32838344)

Your point being? ADSL provides on the order of 12 mbps downstream bandwidth, so the difference in serialization delay is less than 1 ms, even for large packets.

Re:Bandwidth != Latency (1)

KiloByte (825081) | more than 4 years ago | (#32838660)

12ms -- for every single hop. Sure, the routers upstream have a better link than you so it will be less than 12 for those hops, but you still suffer a separate delay for every routers on the way.

gn4a (-1, Troll)

Anonymous Coward | more than 4 years ago | (#32837614)

tto many rules and

moBd 0p (-1, Troll)

Anonymous Coward | more than 4 years ago | (#32837692)

]cuntwipes JordAn

200ms input latency (3, Insightful)

Tei (520358) | more than 4 years ago | (#32837760)

Wen you see "I am connected to a server, and I have 200 ping with this server", that is not input latency. You can have 10ms input latency with a server that give you 200 ping. Things are computed clientside. So this will be much less playable than your average 200ms server.

Re:200ms input latency (1)

nmg196 (184961) | more than 4 years ago | (#32838708)

> You can have 10ms input latency with a server that give you 200 ping.

I presume you meant those numbers the other way round? There is no way to get a lower input latency than the physical one the connection is running over, but it's easily possible the other way round.

Forget the strategy games. (1)

Tei (520358) | more than 4 years ago | (#32837812)

Lag with any "mouse" cursor is horrible, so all strategy or table games that need a cursor will be painfull to play with this.

150ms doesn't mean 150ms in this case (2, Insightful)

Krahar (1655029) | more than 4 years ago | (#32837880)

This number of 150 ms latency may be true but it very likely does not mean at all what you think it means - it is not to be compared to the network latency that multiplayer games often report. The latency you normally see in a game is just the network latency - the amount of time it takes for a small packet to go from your computer to the server. The 150 ms latency includes the time it takes for a packet to go to the server, for the game to process that packet, and then send a frame of video back to you. So the server has registered your action long before the 150 ms are up. Also, normal lag does not include the time it takes for the game to process your command, which can be even more lengthy than your network latency, but that time is included in the 150 ms. Unless you are aware of these things, then the 150 ms number is completely meaningless to you and if you compare it to the latency number from some game you've played before then you are doing it wrong.

What they should have done to get a meaningful comparison is to do the exact camera setup thing they did, but also do it for a game running locally and then over the net. Only then can you meaningfully compare the numbers and know that you got it right.

Beta Tester (1)

se7en11 (833841) | more than 4 years ago | (#32837966)

I was able to get a beta testing account setup about 3 weeks ago. Most of the games play surprisingly well. But for the ones that require fast response times - Dirt, Unreal Tournament - there was just too much lag. Unreal Tournament was practically unplayable since it felt like you were running at 10fps. But for games like Lego Harry Potter, it works great. :)

Can't see this taking off... (1)

Fross (83754) | more than 4 years ago | (#32837998)

200ms input lag is huge. Really huge. The sort of amount that makes you feel like your computer is about to die. Bear in mind this isn't network lag, this is the amount of time it takes to react to your mouse moving, to change your direction with mouselook, etc.

Supposedly it might get better the bigger connection you have. However, if you have a 5-10M connection as they recommend, it's simpler and probably quicker to download the whole game off Steam, or a torrent or whatever.

The only way I can see this working is for people with very high quality net connections, and no decent rigs to run the games on (just enough to stream the audio/video). I'm not an expert but that seems a very specific and counterintuitive demographic.

Fret not! (1)

McTickles (1812316) | more than 4 years ago | (#32838172)

Steve Jobs is going to come and "fix" the laws of physics so as to have 0ms displayed.
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?