Hugsies Blog!

March 28, 2009

OnLive is a pipedream

Filed under: Gaming! — Tags: , , , — Hugsie @ 3:51 am

I don’t even like the concept that Steve Perlman is trying to showoff at the GDC, but more on that later.  In case you don’t know what OnLive is, it’s a new video game service where you don’t need to use real expensive hardware to play your games. Rather than buying an expensive console to render all your pretty eyecandy on screen, OnLive’s servers will do all the heavy CPU/GPU number crunching for you, and simply send you a compressed video feed. All with little to no lag, and HD quality, and it will seem like you’re playing the game on your own PC/Console.

First, everyone who are experts at video compression/encoding are crying “bullshit” about this concept.  OnLive talks about how it has massive super server farms that will handle all of this, so that it can be done in as little as one millisecond from the moment the player press a button on the controller, to when you see those results on screen.  Having a gigantic rendering server farm to do all of this sounds fine and dandy, but how is a small cigarette box sized device going to be decode that video feed that quickly if it took huge server farm to encode it in the first place?

All encoding (video and audio) needs extra buffer time to compress sufficiently to fit though the pipes, before it can start displaying it.  Your mp3 player does this, your DVR does this, your DVD and BluRay player does this.  So how exactly is this supposed to work properly on a “live” video game where things are changing every fraction of a second? A video game isn’t predictable, unlike a pre-recorded video where it’s already known what the next frame will show.  When you run a video codec on a video file, it takes into consideration what the next frame will show in relation to the previous frame, as sometimes many frames in advanced.  Often video codecs only change a small portion of the screen at any given time, since a scene’s background may not change.   Even with live TV broadcast that isn’t pre-recorded still requires several seconds of delay time before it even hits viewers TV screens (not to mention the extra time they put in for censoring)   Also the encoded sound is also another issue that plays into this latency/real-time stuff, and trying to sync audio and video in real time all in one millisecond is mind numbing, considering it’s probably rendering 30 frames a second, minimum 24fps (movie film speed).

What you see during a video game isn’t predictable. The player has control of the so-called “camera” and the player is the director.  This means the codec will have to render the entire screen (full screen) many many times as the player looks around and changes the view.  Or when the player gets shot at, monsters jump at the player, explosions go off, etc. etc.  The codec will not know what the next frame will be to optimize and anticipate the compression method to use next. If it tries to anticipate the next frame, it might be wrong (or show on-screen artifacts) and start a whole new full screen rendering, which will defiantly add to the magical 1ms delay.

This might work if the player just stands in one position the whole time, but that makes a boring game.  The only way this can really work is if they did little to no actual compression of the video, so there is no depredation to the video quality.  After all OnLive states you need at LEAST a 1.5mbit Internet connection to just get standard definition video quality, and 5+mbit connection for 720p HD. So this means we will have to LOSE video quality just to play our games that we are already playing at 1080p.  If they are using very little compression, it makes sense to me since that reportedly OnLive has been in  “stealth” development for seven years.  Seven years ago we BARELY had enough broadband penetration to stream chunky Youtube videos.  So OnLive was probably just sitting on thier hands all this time waiting for ISPs to deliver faster broadband.

<sarcasm>
Ya know I’ve been working on cold fusion for the past 15 years, I’m just waiting on someone else to improve a different technology so my work can be closer to completion.
</sarcasm>

Now back to the real issue I have with this concept.  This will remove the need to buy games on any media.  OnLive is all about a subscription service you pay for (perhaps by the hour) and play your games. It’s not just multiplayer games, but also single player games.  I just don’t like the idea that I’m paying someone a fee for every hour I play a single player game like BioShock, or Fallout 3.  I can understand that for a multiplayer game, but not when it’s a single player game, and I shouldn’t have to be pressured to finish my game because my time is about to run out.  I should be able to buy a game for $50 and play it any time, and as often as I like with out incurring more fees on a game that I already consider to be overpriced.  I also just don’t like being dependent on my ISP, because if my ISP is having a bad day and goes offline, I can’t play any of my games, including SINGLE PLAYER games.

I think the real reason Steve Perlman is promoting this is the fact that if this concept works, video game piracy would disappear.  As a matter of fact so would the pre-used game market would die (like GameStop), since you will not purchase the games on discs anymore.  So you know gaming companies, and publishers will want to endorse, and pump in HUGE amounts of capital into OnLive to get it off the ground, and promote the hell out of it.

Afterall this was introduced at GDC, which is only for game DEVELOPERS, not common game consumers like us.  I’m quite sure he’s hyping this to entice investments from rich people, and game developers and publishers who still have deep pockets even in this economic climate. Feeding promises to game publishers with something that could end piracy, and aftermarket resales of games.  In this economic climate, this could be just an opertunity for Perlman to get a ton of capital he can then sit on, while waiting for ISPs to improve broadband speeds to home users in another seven years.  Sure, by then we will all have 10terabit internet connections, and then this concept will work!

I hope this dosen’t work.  Common sense tells me it won’t, even if broadband became so fast and ubiqudus in the future, it still dosen’t seem right. It’s like big brother watching over everything, and you have no privacy, and closing out indie game developers from ever getting into this market.

I just don’t want to be a gamer slave to OnLive.

Advertisements

Leave a Comment »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: