New Tech To Revolutionize Gaming?

[Sunbury]

http://www.technewsworld.com/story/OnLive-Promises-Hard-Core-Gaming-Minus-the-Hardware-66605.html

Basically video compression, so both client and server are on the server, client is 'dumb cheap' display.

Much of the media buzz surrounding OnLive revolves around the partnerships it is announcing at its debut. Those attending Tuesday's press conference will be able to watch demonstrations of games from eight of the industry's major software companies, including Electronic Arts, Atari, Ubisoft, Eidos, Take-Two Interactive and Warner Bros.

The company is promising no lag time on its streams.

Wat?  Dawn Lives!

[Salamok]

saw this earlier as well, seems like if true it would eliminate most hacking in multiplayer games (except macro based stuff).  Also might be the answer to mmog + destroyable environment w/accurate effects (everyone sees the same shrapnel from that hole you just blew in the ground).

[HaemishM]

I smell a Phantom. It sounds like it's using a Citrix like tech, only with video instead of stillshots streamed out. And Citrix, while cool for what it does, blows monkey ass for speed. Short of negative ping code, I remain skeptical about the viability of this service.

[schild]

When I saw this on NGaf, my first response was "Servers will be on Ebay within a year."

[Salamok]

I'm thinkin it might be doable at low resolution 480x320/15-20fps or somesuch, if so it would be pretty sweet if they released a mobile phone client.  Then again that would be the only decent use for that level of performance. 

[Samwise]

This might be okay for MMOGs or other games that don't have to be pretty or demand good reaction times.  I can't see it taking off for FPSes any time soon.  Even if the tech actually works and the average user has enough bandwidth for it.

[Venkman]

This has been kicked around for a number of years. The theory is that if you've got a connection to watch streaming HD video on your PC, why can't you "watch" the game be rendered for you and then just streamed to your PC as a video. It's an interesting idea, but isn't going to work for FPS games right away, for the same reason MMOFPS games aren't really MMO in the way PS tried to be right now.

The other problem of course is that all it really does is put the onus of both server and client overhead onto the developer/publisher. I don't know what that does to the resource and maintenance costs, but it would seem like it's at least adding a cost of $900 per user on your system. Right now MMOs rely on hundreds of thousands of those consumer-owned PCs doing the work for them.

So while it's cool, is it really worthwhile? What's the benefit to the company?

I really want to know. I've been wondering that for years.

[Yegolev]

Are game servers only $900?

[Fabricated]

Hey guys I have my own secret psychic netcode, can I get shitloads of venture capital to spend on marble desks and air hockey tables!??

[Sheepherder]

You would have to send input to the server for it to be reacted upon.  I think most people will shortly say "fuck this noise" and upgrade their PC when they realize the response time of their character is limited by their round-trip ping.

[Quinton]

Oh gods... Steve Perlman...

[Yoru]

Game timesharing.

Everything old is new dumb again.

[schild]

I need to come up with something amazingly stupid and amazingly awesome and get incredibly dumb fucking investors to just hand over money to give myself a fat paycheck and come back in 3 years with dismal failure.

This is, in fact, so fucking beyond stupid that I wish I was at GDC just to publically ridicule these retards.

[Yegolev]

I have had this idea on a tech to replace ethernet.  Instead of shooting out a packet at random, I think it would be far better to have each computer take turns transmitting on the wire.  This would eliminate packet collision and lag would disappear!

[Venkman]

Are game servers only $900?

No. That was just my guesstimate for the extra cost per account holder, based on building a good enough gamer PC to drive the graphics that would then be streamed.

[schild]

Are game servers only $900?

No. That was just my guesstimate for the extra cost per account holder, based on building a good enough gamer PC to drive the graphics that would then be streamed.

This is reason #13728318273 that this idea is just fundamentally stupid.

[Sky]

This might be okay for MMOGs or other games that don't have to be pretty or demand good reaction times. 

Wut? I know this is an old adage, but I don't think it's exactly true. When I have to time a feign death to just before I get taken out, or I have to time stuns/interrupts to the beginning of a casting animation...Maybe not fps razor's edge speed, but not ye olde 'autoattack + sandwich'.

[schild]

It's not even really worth discussing at length, it won't work for fucking anything, it's goddamn stupid.

[Venkman]

It's not even really worth discussing at length, it won't work for fucking anything, it's goddamn stupid.

Yea, but here's the thing: lots of business people are talking about it. They see "WebTV" and "Quicktime" (with the obvious allegory to $$$ Appleomgrichesiphone!!!1/) and are all wondering about it. Telling them they're too stupid to see the truth doesn't prevent them from wanting to spend their money investing in it.

[schild]

It's not even really worth discussing at length, it won't work for fucking anything, it's goddamn stupid.

Yea, but here's the thing: lots of business people are talking about it. They see "WebTV" and "Quicktime" (with the obvious allegory to $$$ Appleomgrichesiphone!!!1/) and are all wondering about it. Telling them they're too stupid to see the truth doesn't prevent them from wanting to spend their money investing in it.

Print up a flier that says "You are an untrained, knowledgeable douchebag."

And then hand it out to anyone that looks interested in OnLive. On the back put "You can thank me later."

[Sky]

It's not even really worth discussing at length, it won't work for fucking anything, it's goddamn stupid.

I agree with this statement. 

[FatuousTwat]

I actually saw a segment on "Up to the Minute" about this... WTF?

[NiX]

Oh man! I'm hijacking this thread. I have to write a thesis paper on video game consoles and how they will drastically change how we interact with media. Is there any new tech coming about? Stupid stuff like Head2Go and all that jazz.

[Salamok]

I have had this idea on a tech to replace ethernet.  Instead of shooting out a packet at random, I think it would be far better to have each computer take turns transmitting on the wire.  This would eliminate packet collision and lag would disappear!

Old solutions for new problems, I bet you could even sell the idea to IBM if you could just come up with a jazzy new name for the talking stick network protocol.

[raydeen]

Just seeing 'WebTV' would be enough for me to flee in terror.

From one of the articles I read, the video is super compressed and has a stream time of around 1 millisecond. Not bad there, but what is the time for the set top box to transmit the control signal to the remote server? The video may be fast but I'm betting the remote control isn't. You're basically remote desktop'ing into a server which might be nice on a LAN but over the internet? They're gonna need negative ping code especially if/when they have several hundred/thousand/million clients all connecting to their own virtual system.

[Venkman]

One hour video on GameSpot.

Check out the footage of the games in action. Laggy even ven on what I'm sure is a highly tuned presentation tech. Now put the entire internet in the way 

[Sunbury]

PC Worlds take on it -
http://www.pcworld.com/article/161930-2/gdc_09_6_reasons_onlive_could_be_a_bust.html

[IainC]

PC Worlds take on it -
http://www.pcworld.com/article/161930-2/gdc_09_6_reasons_onlive_could_be_a_bust.html

I'm not saying they're wrong but they did pull some pretty weak reasons out. There are a lot of much better reasons to slate the concept on.

[Murgos]

Just seeing 'WebTV' would be enough for me to flee in terror.

From one of the articles I read, the video is super compressed and has a stream time of around 1 millisecond. Not bad there, but what is the time for the set top box to transmit the control signal to the remote server? The video may be fast but I'm betting the remote control isn't. You're basically remote desktop'ing into a server which might be nice on a LAN but over the internet? They're gonna need negative ping code especially if/when they have several hundred/thousand/million clients all connecting to their own virtual system.

50 ms for the stream to go from their server and reach your box at a best case.  200 ms for your brain to process it and hit a button at best case. 50 ms for the command to reach them, 1-5 ms for processing and then 50 ms return time seems like a reasonable model for gaming with, basically thats only 105ms delay between you pressing a button and seeing the effect.

BUT

The 1 ms time to receive the packet is meaningless and arbitrary without knowing the bandwidth and the packet size.  Even assuming a 5 Mbs dl speed and 32 bit words this gives a payload somewhere around 150 words.  How the fuck do you turn 150 words of data into acceptable graphics?

You can do it with a full client because the words just say move pc pov to x, play animation y, change numbers[z] and all the data and processing on doing that is on your side of the pipe.  It doesn't make sense to say you can do that with nothing doing processing on the receiving end.

[Severian]

I need to come up with something amazingly stupid and amazingly awesome and get incredibly dumb fucking investors to just hand over money to give myself a fat paycheck and come back in 3 years with dismal failure.

This is, in fact, so fucking beyond stupid that I wish I was at GDC just to publically ridicule these retards.

You'll need to include Dave Perry in that group now, too. Although I'm not clear on whether said group is supposed to be retards or VC-fleecers living large.

OnLive video games on demand service shakes loose a competitor: Dave Perry

"...He plans on raising a round of venture money..."

[CadetUmfer]

Great idea, wrong implementation.  This uses existing executables.  Net win = 0.

I want an MMO with server-side rendering, designed to draw many views of a single scene.  From a tech standpoint, getting the vertex/texture data to the card is harder than the actual drawing.  Right now, every client has to go through the whole rendering process.  Imagine if that was only done once per zone.  The server has all the data needed for an entire zone on the GPU at all times, and can render your view better than your high-end gaming PC, and stream it to any PC that can watch Hulu.

[Trippy]

Great idea, wrong implementation.  This uses existing executables.  Net win = 0.

I want an MMO with server-side rendering, designed to draw many views of a single scene.  From a tech standpoint, getting the vertex/texture data to the card is harder than the actual drawing.  Right now, every client has to go through the whole rendering process.  Imagine if that was only done once per zone.  The server has all the data needed for an entire zone on the GPU at all times, and can render your view better than your high-end gaming PC, and stream them to anyone that can watch Hulu.

[Hindenburg]

Ookii, for shame, you could've just gone with

Great idea

You're wrong.

[Murgos]

Great idea, wrong implementation.  This uses existing executables.  Net win = 0.

I want an MMO with server-side rendering, designed to draw many views of a single scene.  From a tech standpoint, getting the vertex/texture data to the card is harder than the actual drawing.  Right now, every client has to go through the whole rendering process.  Imagine if that was only done once per zone.  The server has all the data needed for an entire zone on the GPU at all times, and can render your view better than your high-end gaming PC, and stream it to any PC that can watch Hulu.

Just for clarity are you suggesting that the server should maintain in memory every pixel needed for the entire zone and the perspective of every client in the zone and then just stream the result at their native resolution directly to each client at 60 FPS?

[CadetUmfer]

Just for clarity are you suggesting that the server should maintain in memory every pixel needed for the entire zone and the perspective of every client in the zone and then just stream the result at their native resolution directly to each client at 60 FPS?

That's exactly what OnLive is doing.  It runs a version of the game for each client, on their servers, and streams you the output.  N scenes, N copies of the graphics data, renderered N times.  Lots of server hardware, minimal win.

I'm saying that's not the right way to do it.

1 scene, 1 copy of the graphics data, rendered N times.  Exponentially faster, and capable of doing things even the best gaming PC can't (because the server controls the hardware).