Digital distribution is on the way, in a big way

I was interviewed last week by Paul Hyman of The Hollywood Reporter, for his weekly column appearing today, delving into the oncoming wave that is digital content distribution, a la Steam, and new contender on the block, Game xStream.

There's a lot of resistance to digital distribution because of the way Valve handled it with Half-Life 2, which even today continues to annoy more than its fair share of paying customers.  Definitely, there are kinks to be worked out, but rather than condemn digital distribution based on Steam alone, I think it has a bright future.  And if anything, it may be one of the most promising ways for independent developers to bootstrap themselves, and bypass the entire publisher/retailer model, which is closed to most indies.

I have a strong belief that digital distribution is about to catch on like Jar-Jar bashing.  Or, more seriously, like CDs did after Myst hit the scene in the early 90's.

Sure, we all hate only authentication, but that's not an inherent feature of online distribution.  Online distribution, at it's core, is merely a way to buy a game without having to go to a store.  That's it.  Steam has given everyone the impression that it also means an online tether used as copy protection.  But that's something else entirely, and has unfortunately given online distribution a bad name.

3D Realms choose to work with Game xStream because their service is not only highly flexible, meaning that a whole list of features can be either used or not, but because it allows what in my opinion is the most important feature of all, and one that Steam does not support:  A player can buy a game, and within minutes start playing it, without needing to download the entire game.  For example, I can start downloading the Doom 3 demo -- one of the most content intense games currently available -- and within six minutes I'm playing it as if I have the full game, racing like a madman through the opening level, while the game continues to download itself in the background.  (You can not catch up to where new content isn't available.)

This single feature makes digital distribution truly viable for the mass gamer market, because it satisfies the desire to play the game right away (or nearly so).  Without this single feature, you'd need to wait an hour to many hours to begin playing, and that sours the impulse desire of buying and playing right away.

I've been on record for saying that I didn't think digital distribution would ever catch on for the triple-A games, because I didn't think people wanted to wait for massive downloads, but Game xStream jumps this hurdle.  And another nice thing is that there's no effort by the developer at all.  No need to order content in a certain way.  The process is completely transparent to developers.  That cannot be underestimated as a selling point in getting Game xStream (or similar systems) accepted by publishers, and developers.

Comments

[Sludge]

It will be good to have the digital distro option on the table. Even if developers don't use it, the potential to forge a deal with a publisher when digital distro is a viable alternative may go some distance to swinging the negotiation in favor of the developer.

[Jeff Lindsay]

I remember seeing one of Valve's first presentations of Steam at the GDC. One of their major selling points was that it could stream content so you wouldn't have to download the entire game. They even showed a demo of this, but I guess for whatever reason, it didn't make it to the final release.

[Duoae]

"They even showed a demo of this, but I guess for whatever reason, it didn't make it to the final release."

Now, i'm not one to bash valve but: Big surprise there then....

As for online distribution, i really like it. But with the way the whole DRM thing is being handled i'm not sure how much it will take certain companies to overcome the "traditional" online distribution methods that have alienated people from the idea already.

[Jay Barnson]

"And another nice thing is that there's no effort by the developer at all. No need to order content in a certain way. The process is completely transparent to developers."

Truly? This is really scary-cool if this is correct. I can't imagine how it wouldn't require some effort on the part of the developers, unless they give you a specific framework to work with from the get-go.

Still - this is a fascinating opportunity for games to finally make a true break from the brick-and-mortar system we currently have. Not that it won't continue to be an important part of a game's distribution - just not the only option.

I'm anxious to hear how this goes :)

[Scott Miller]

Jay, like I said, developers do not need to put forth a breath of effort to take advantage of Game xStream's pay-and-play distribution. For example, the Game xStream folks used the already released Doom 3 demo and made it work as I described in my post (Id didn't lift a finger). Several other released demos worked in the same way. With internal-only tests, Game xStream has verified that their system works with the full version of many of the most popular triple-A games, including Half-Life 2, which with their system is playable within minutes, rather than hours as with Steam.

I can't explain how they pull this off, as I'm under NDA, but it is remarkable.

[Cebrian]

If I'm not mistaken Steam supports playing a game without a full download. You can start a game that is not ready for "offline playing" and load times are much longer because Steam will download new levels and models as needed. Maybe this is not done in Half-Life 2 because content is bigger, but I'm pretty sure you can play the original Half-Life via Steam this way.

[Jeff Peil]

There are some pretty braindead obvious (at a high level, but devil-in-the-details) things they could do.

For instance treating the game files as page-in-on-demand by using a file system filter driver. So when page x of a file is requested, if it isn't there the network layer prioritizes grabbing it. The harder part would be setting up an initial load order priority. However it seems to me that this wouldn't be all that hard either (in theory) because they only need to record a "download profile" of sorts by running the game a certain number of times, at which point they have a very good idea of the default order to stream the data in.

That's not to say it doesn't sound very cool. Just that I wouldn't call coming up with a way to do this non-obvious.

[Factory]

Erm, what Jeff said, xStream prolly works be pretending to be the filesystem. Of course the downside to this is that loading times will be time it takes to download from the server, as opposed loading from disk. A big FMV at the start of a game wouldn't work so well.

Of course the downsides are that the user still needs to have the bandwidth allowance to dl the game (ie. ppl with only 2Gb download allowance a month), and broadband.
On the good side, eventually ppl will have Fiber-to-Home, and games will take a few seconds to dl, and this prolly won't be needed so much.

[Ben]

Seems good, I have been running a series on my own blog about how I would like steam redesigned on my blog, and that feature was high up on the list aswell.

But the thing that worrys me about xStream, is can it handle an AAA game release? Streaming content is all well and good, but if everybody is only getting 5kB's per second because the servers are getting hammered, then it's not a whole lot of use.

I will save my criticisms and praises of xStream untill it handles the launch of a title like HL2 or DNF and the subsequent patches.

The amount of money valve must spend on bandwidth just makes my mind boggle. Does xStream have 15gBs of bandwidth behind them? And can it overcome some of the problems that are out of it's control like a poor internet infrastructure in lots of countrys like Australia?

[Marty]

Another option for small companies is to simply handle the distribution themselves. A nice success story is http://www.liveforspeed.net/

[mono]

The whole streaming aspect of xStream is not interesting. It continues to be emphasized, but really, its only relevant for the 1st 2-3 hours of a game release. There isn't a single PC gamer that I've ever met that doesn't want the entire game downloaded to his drive, burnable to disk, and reinstallable a bijillion times on every iteration of their homebrew PC's over the coming years.

A simple key code, ala the Quake 3 engine games, that authenticates against a master server during online play should be enough. Limiting re-installs, or requiring online activation to play single player will quickly sour all gamers against any digital distribution service.

[gf]

I'm a bit confused as to what exactly xStream is. I get the impression it's a service you subscribe to (and pay for). If that's the case then I think that's a bad thing. To be really appealing to the mass market, I beleive buying something online should be as simple as walking in to a store and grabbing a copy, no strings attached whatsoever.

I might have just misunderstood it though, I think their site is a bit fuzzy about what exactly it is they do. Will they be an online shop type of thing, where you can browse games and buy, or do they just provide the back-end (servers and tools) and the developer/publisher will deal with the rest.

[Mr.DJ]

I haven't done any research on xStream, and therfore doesn't know more than what you've(Scott M.) told us.

But, like others have mentioned, I am curious to know if they have thought of the possibility of not having enough bandwith, to serve the massive downloads at launch, at acceptable speeds?

And what if a disconnection occurs, while playing early downloaded content, and there's suddenly no more data?

The situation you descriebed, sounds a little like taken from an ideal enviroment.
Where the user has a stable 2Mb connection with low ping, and has max download bandwith speed from xStream.
If the eksample you descriebed, only works on above mentioned enviroments(or close to).
I fear there may be a considerable amount of dissapointed users complaining.

Regarding the issue of enough available bandwith at launch, I would imagine something like a Torrent distro. system, would help on the bandwith cost.

[Jerane Alleyne]

Hey Scott,

It's really a pleasant surprise to see you've warmed up to digital distribution. I remember the discussions we had on Garagegames a few years ago :)

I remember that Steam was initially supposed to have a game-streaming system similar to what GamexStream will be doing...I think Yahoo has a system like that as well, but they just offer older games. I hope they were able to get around the isues that Valve never got to do. It would be nice to see a game service that offers current/new games and becomes a major force in distributing games.

Good luck Scott, I'll be looking forward to seeing this work.

[Tom Edwards]

I'm with the blog all the way: it's good to see someone higher up the chain than me give the message that DD (or Steam for that matter) doesn't always have to be used as overzealously as it was in HL2. When Valve get their new publisher and re-release the game they have a chance to redeem themselves - let's hope they take it.

One other thing to say: Steam fully supports content streaming. It doesn't work because of poor, no scratch that awful implementation. I uploaded just one of the lists of files HL2 needs before it will launch to the main menu at http://www.btinternet.com/~varsity_uk/HL2/engine.lst - take a look, it's terrible, auto-generated illogical rubbish. If it were even vaguely optimised you'd see a huge improvement. Another example of a perfectly good feature being put to a poor use that unfairly brings its reputation down. :-(

Ben, could you post a link to your blog? I wouldn't mind comparing notes.

[Ben]

Sure, it's a series and I just finished up the "visual and usability" aspect of it, I should have the rest up within a week.

Wait a second........ Varisty ........

Small world, halflife2.net - ^Ben

I completly agree on the oversealousness.

http://tripstertoo.blogspot.com/

[Scott Miller]

-- "I get the impression it's a service you subscribe to (and pay for). If that's the case then I think that's a bad thing."

We'll be using xStream merely as an alternate method to buy the game. There is no subscription fee, there is only a one-time purchase fee, exactly as with buying the game from retail stores.

-- "The whole streaming aspect of xStream is not interesting. It continues to be emphasized, but really, its only relevant for the 1st 2-3 hours of a game release."

Well, I strongly believe in people's desire to begin playing a purchased game asap. The shorter the period between pressing the Buy Me button and the Play Me, the more successful online sales will be. Satisfying the impulse buying decision is the key.

BTW, streaming also works with game patches, allowing a game to be patched while you are playing it, such that it is entirely transparent and hassle-free. Wouldn't this be a great feature for those of us who almost weekly have to put up with WoW patches that can take 30 - 120 minutes to patch, keeping us from playing the game while the patch is in progress?

[Tom Edwards]

While you are playing it? Does that include code? If it does, I'm very impressed indeed.

[Niobium]

--"streaming also works with game patches, allowing a game to be patched while you are playing it"

That would be terrifying.. They tweak a "balance" issue and all of the sudden the totally whoop weapon you've been using, all of the sudden, does half as much damage.... Right in the middle of the battle with the last boss.. You're boned, have a nice day. :)

[Scott Miller]

Niobium, you are pulling a worst-case scenario out of you butt. Calm down your fear. ;-) You can always elect to patch the old fashioned way.

Tom, yes, code, content, you-name-it...patched while you play.

[Anon]

More to the point you've perhaps confused two issues Niobium.

Most likely the game will still require a restart for the patch to start applying. What you would be able to avoid is the wait for the patch to finish downloading before you can start running the game (granted their are potential exceptions to the needing a restart, such as patching a map that isn't currently loaded, but the odds of this working reliably without changes to the game code where they don't get a chance to intercept the game from startup to ensure that it only sees data from the new version of the file are low.)

Similarly in a WoW like game they'd need to bring the servers down, patch them, then bring them back up.

Don't think of it as "magic" downloading in the background or "magic" patching. Think of it as a streaming extension to your filesystem which would allow for earliar access to files (because you're good to go as soon as the parts you initially need have finished streaming.) But much like a traditional filesystem, don't imagine that they can safely change the contents of a file while the game is running (unless the game is specifically designed to support it.)

[Soren]

What happends if you have to re-install? Surely, it can't just let you download arbitrarily whenever you want...

[Ben Harris]

What you guys are describing is already being used in Guild Wars, the difference being that Guild Wars has been designed to be able to take advantage of streaming technology - games like Doom3 and it's ilk have not.

I would be extremely interested in learning how xStream plans to implement their technology with games that were not designed with this distribution method in mind, especially considering that my testing of xStream some months back left me quite disappointed as to its performance.

Considering that Guild Wars is the only successful implementation of content streaming (that I know of), I would say the future is very bright indeed.

[Tadhg]

Question Scott (If you can talk about it that is):

One of the main advantages of downloadable distribution, it seems to me, is not its ability to replicate the data delivered on a disk via the interweb, but the ability to break out from some of the unnecessary conventions of disc-based gaming to begin with.

Such as the need to be thinking in terms of lots of linear content. Rather than having to release a whole lot of Max Payne 2 and then wait a couple of years for the next one, like a novel, online distribution affords the ability to serialise games, like Max Payne 2 the continuing series of levels and stories etc. This means organising studios more around continuous production, more like comics. The main advantage of it being that you really get the value out of the engine development and ultimately facilitate a process through which far richer content (in terms of depth) can be delivered if punters are interested.

My question is whether you think that this is possible, probable, lunacy, and whether or not this is the sort of thing that you're thinking of with online dist?

[Manofgod]

Ben....

>> I would be extremely interested in learning how xStream plans to implement their technology with games that were not designed with this distribution method in mind, especially considering that my testing of xStream some months back left me quite disappointed as to its performance. <<

That is the brilliance behind the Game xStream system. Games that are not designed for streaming can be streamed with Game xStream, with no developer coding. All Game xStream requires is a gold master or a CD from the retail shelf. Average start times are about 12 minutes, depending on the size of the game and the bandwidth the user has. This way the user gets instant gratification for being able to play the game, while still downloading the full content. Absolutely no part of the gaming experience is lost during the streaming transfer. If the game supports online gaming, the technology monitors the users bandwidth, and allows them to get the best ping rate, while still downloading the game. The system even supports save games, and resume if the streamer or download process is interrupted.

Applying the technology to the master files can even turn the CD or Gold Master into a streaming device. Turning a PC into a console. This means that the Game xStream technology can turn any retail sold CD into a Tray-n-Play CD. Imagine putting in a game (CD) and playing it within seconds, that would normally require upwards of 45 minutes to install!

The benefits to the consumer become very apparent, and takes the normally boring install time and turns it into play time, giving the consumer the instant gratification they desire.

Also remember that the Game xStream streaming system is designed for at the very least, a 1 megabit broadband connection. Below that, the download option is still available to the consumer, and then can be switched into streaming mode once the slower connection has downloaded enough data to get the game going.

This is a good time for users to test the beta system of Game xStream and provide input to improve itself prior to commercial launch. This puts the gaming community in perfect position to relate what they would like to see. If they provide no comments, then they have no reason to complain about it when the commercial launch is done.

http://community.gamexstream.com/

[Tom Edwards]

Applying the technology to the master files can even turn the CD or Gold Master into a streaming device. Turning a PC into a console. This means that the Game xStream technology can turn any retail sold CD into a Tray-n-Play CD. Imagine putting in a game (CD) and playing it within seconds, that would normally require upwards of 45 minutes to install!

Now THAT is cool. ;-)

[Scott Miller]

Tadhg,

I'm not sold yet on the idea of episodic content, at least for the kind of game we like to do. It just seems better to release a full game of x years, rather that eight game episodes of the same period.

I think a more workable system is the one Valve seems to be pursuing, where they release a blockbuster AAA game, and then as quickly as they can they release expansion packs for digital download. I think you need that full, blockbuster game to make that giant splash, and then you can switch to an episodic-style approach.

[Ben]

Maybe a possible route you can take with eposodic content.

Release maybe 4 episodes each with about 3-4 hours of game play contained within each episode on to digital distrabution. Once you amass a "full games length" then bundle them together and release it to retail.

Month 1 - Release Episode 1, holds about 3 hours of gameplay.
Month 2 - Release Episode 2, holds about 6 hours.
Month 3 - Package Month 1 and 2 together and sell them at retail.

[Michael Dragojlovic]

If you buy a game that can be streamed via Game xStream at a shop will you still be able to use Game xStream to patch the game at no additional cost?

[Tadhg]

"I'm not sold yet on the idea of episodic content, at least for the kind of game we like to do. It just seems better to release a full game of x years, rather that eight game episodes of the same period."

It just seems sensible to me. I played Max Payne 2 a year ago. In the intervening year you could have put out further instalments online or whatever and I'd buy them. I think that the secret to character franchised games is actually to have a continuing character in that sense. More like a cop TV show than, say, Lethal Weapon.

"I think a more workable system is the one Valve seems to be pursuing, where they release a blockbuster AAA game, and then as quickly as they can they release expansion packs for digital download. I think you need that full, blockbuster game to make that giant splash, and then you can switch to an episodic-style approach."

Yes, but at what cost? I seriously wonder, given the dev costs of Steam and the 6 years of HL2, whether Valve have even started to see any real profit out of doing things that way. I thik that they could have easily released a lot more HL installments and side stories and so on than they did, possibly through a secondary studio or whatever, while burrowing away on HL2.

[Tom Edwards]

Yes, but at what cost? I seriously wonder, given the dev costs of Steam and the 6 years of HL2, whether Valve have even started to see any real profit out of doing things that way.

I think you are a little confused. Firstly HL1 was before any of this started and isn't relavant, and secondly not every developer is going to go out and make their own distribution system so you can hardly take the cost of setting one up into account.

[Tadhg]

"I think you are a little confused. Firstly HL1 was before any of this started and isn't relavant, and secondly not every developer is going to go out and make their own distribution system so you can hardly take the cost of setting one up into account."

HL2 took six years or so to develop. In that time, they produced a couple of expansion packs for it, and Counterstrike essentially fell in their lap. Why, in the intervening time, they weren't focusing at least some effort on putting together further Half Life episodes (whether to release over the internet or at retail) is a mystery, because there would be nothing physically stopping them from doing so.

As for the distribution method, for the moment I disagree. Steam would appear to have been built very much with HL2 in mind, not just as an overall general portal. It's part and parcel of the product for the moment, and I see no reason to assume that its continued existence will not be tied to HL3, HL4 and so on as Steam 2, Steam 3 etc.

Which is fine, of course, so long as you're making money off of it.

[Tom Edwards]

We're talking about episodic content delivered over digital distribution platforms. Whether or not enough xpacks were made for HL1 really has no bearing on it. ;-)

Steam is not built purely for HL2, or even purely for Valve's games (what makes you think it is?). There's a bunch of titles coming out on it from other teams, though the only one we know of right now is the MMORPG Pirates of the Burning Sea.
Even if it was, people are NOT going to spend millions of dollars and two plus years of their time making proprietary systems when existing ones are waiting for their phonecall.

[Tadhg]

"We're talking about episodic content delivered over digital distribution platforms. Whether or not enough xpacks were made for HL1 really has no bearing on it. ;-)"

Indeed. But I'd diverged into talking about episodic content generally.

"Steam is not built purely for HL2, or even purely for Valve's games (what makes you think it is?). There's a bunch of titles coming out on it from other teams, though the only one we know of right now is the MMORPG Pirates of the Burning Sea.
Even if it was, people are NOT going to spend millions of dollars and two plus years of their time making proprietary systems when existing ones are waiting for their phonecall."

What makes me think that it is is the lack of other Steam based titles to this point. It's like opening a shop with one product. Gives you the impression that it's really a one-product thing. Don't get me wrong, I'm all for developers getting out from under the yoke and I wish Valve all the success in the world. I'm just saying that according to my current impressions, Steam looks more like a part of HL2 than a product in its own right. Look at www.steampowered.com, for example.

This may change with time, of course, but I suspect that Steam thus far would have been better served if it had much higher levels of presence and some other large titles ready to go when it launched. Almost like a console manufacturer would do it, a la xbox live.

Here's what I think is actually going to happen with online distribution btw; It's going to become another retail channel. Some large corp, probably Microsoft, is now doubt currently barreling away on their own high profile PC distribtion system, maybe to go with Longhorn. They'll cheerfully allow Valve etc to spec out the marketplace for a while, proving its worth, and then do what they always do, which is slam in like an 800lb gorilla and do key deals with everyone,and control the distrubution as a corporation would.

I think that online's usefulness is therefore real but limited, and it isn't going to be the gateway to riches that developers might think. My hope is that it survives enough to help build that indie-ethic side of distribution that has been sorely missing from games for a long time, but as a mainstream content-provision channel it's got limited value until MS (probably) make it theirs.

[Scott Miller]

Tadhg, it's a lot harder to develop high-level, AAA episodic content than maybe you think. Look at Max 2, for example, it took two years to make a 6-8 hour game!

Also, when you're making episodic content, do you freeze the engine/tech for years, because that's really the only way to make it work. IOW, your only chance to release new content yearly (or more often than that even), is to not touch the engine or tools, so that content can be created without worry of redoing it or chasing an evolving tech spec. Thus, episodic content will look dated after the first year or two, falling behind the look of more advanced engines.

Next, with expansion packs, the buy-in ratio is always 25% or less of the full version. Will episodic content follow this same dismal pattern? And similarly, will anyone that doesn't get in on the first in the series even think about buying follow-up episodes?

So these are just a few reasons I do not think we'll see the episodic approach used anytime soon for triple-A releases.

[Jeff Peil]

Scott,

I think a notable exception is the mmorpg space,

Certainly the somewhat episodic content is one of the carrots that keeps subscribers around (be it new dungeons, new types of gameplay, new parts of the world, etc.)

If it wasn't for content like Dire Maul, Battlegrounds, etc coming. I know of many that would already be done with the game.

In fact the high end guilds in all of these games (which in my opinion act in many ways as their own dangling carrot for a large portion of the non-end-game population) would almost certainly have moved on if they weren't constantly preparing for the next episode of content (they've beaten molten core, they only keep going back to prepare themselves for Blackwing Spire, when they beat BWS the same will be true until the next raid dungeon, etc.)

[Tom Edwards]

[i]Also, when you're making episodic content, do you freeze the engine/tech for years, because that's really the only way to make it work.[/i]

Back to Valve again - welcome to the Source modularity era. ;-)

[Tadhg]

"Tadhg, it's a lot harder to develop high-level, AAA episodic content than maybe you think. Look at Max 2, for example, it took two years to make a 6-8 hour game!"

:)

Ok, but breaking that down between prototyping and actual production, what time did it take to actually produce the levels themselves rather than the engine, physics, game mechanics balancing and so on?

"Also, when you're making episodic content, do you freeze the engine/tech for years, because that's really the only way to make it work. IOW, your only chance to release new content yearly (or more often than that even), is to not touch the engine or tools, so that content can be created without worry of redoing it or chasing an evolving tech spec. Thus, episodic content will look dated after the first year or two, falling behind the look of more advanced engines."

But again, are we in the business of character/franchise/IP creation as you often preach, or are we in the business of technology novelty?

It's important to make the distinction because if we're in the first business then what you're doing is building loyalty and love, a model that has served (among others) Nintendo very well.

Look at the Pokemon games, for example. Regular releases = loyalty even if they all look the same. Look at EA and their annual franchises. Same gig. Episodic mentality false into the same sort of thinking. There is no shame in actually getting the value out of all that engine development work.

On the other hand, if we're in the business of technology novelty, then there is no loyalty and we're trapped playing the endless-chase-the-dragon game of trying to impress journalists, hardcore and media types in the hope that that translates into game sales. That is deeply flawed in my opinion because we're chasing imaginary enemies and what other people *might* do. It's that attitude which produces poor rates of return and vastly inflated development costs and games that seem to take forever to get out there.

"Next, with expansion packs, the buy-in ratio is always 25% or less of the full version. Will episodic content follow this same dismal pattern? And similarly, will anyone that doesn't get in on the first in the series even think about buying follow-up episodes?"

I would imagine that, yes, the episodic content would likely attract between 25 and 33% of people. However, if you're aware that you're doing no engine redevelopment and you're aware that as an 'episode' it doesn't have to be 40 hours long, and if the price is bite-sized enough for people to get into it without feeling that they're spending big money (or if its subscribed) and pushed over the net so your take and your production of disks etc is lower etc etc etc then it can work. 25% of your original per episode take can still be highly lucrative if you're not spending for Jesus.

Would anyone who missed the first one feel like they couldn't get in? Possibly. That depends on how you structure it. 24 is very difficult to get into if you haven't seen all the previous episodes that season because of the huge story arc. The West Wing, on the other hand, is not because it's story arcs are self contained over a short number of episodes (or single ones) with more general loose threads carrying it along. It all depends on how you approach it.

"So these are just a few reasons I do not think we'll see the episodic approach used anytime soon for triple-A releases."

Well not if you keep chasing those imaginary enemies it won't. On the other hand, all the developers that seem to be doing that also seem to be very slowly going out of business anyway. Nobody has a solution to the fact that the geometrically increasing costs of development are not being matched by geomatrically inceasing sales, and the end of that road is ruin. As many developers have already found out, and as the more premium developers are beginning to.

Looking over the landscape of development today, in the West, I think the only developer that seems to have their financial head screwed on is Bioware. The rest of us appear to be stealing from Peter to pay Paul, fighting fires and convincing ourselves that it will all end well. Somehow.

[Jerane Alleyne]

I remember playing Guild Wars, which uses streaming tech, and I remember it killing my Ram, which was at 512mb at the time. I would like to know, that with constant streaming of info, what kind of affect would this have on a system playing a live game, in the case of Game xStream particularly.

[Charles E. Hardwidge]

The rest of us appear to be stealing from Peter to pay Paul, fighting fires and convincing ourselves that it will all end well.

I've given up worrying about it. Right or wrong doesn't come into it when you're dealing with ego. Sure, there's been a lot of achievements but denial doesn't mean there aren't better alternatives.

Whether we're talking about delivery and content, as with xStream and episodic games, or Blogs and comments, the fundamentals are the same. How does the small and different challenge the big and uniform?

People blame the Hollywoodisation of industries, of which British charities are the latest victim, but I'm not so sure. Big spin is a frightening monster, but little truths are an effective sword.

[thwidra]

Hello Scott,

what you describe in your posting is still used in this new Game "Guild Wars". You can buy a key online, and then the download is running. After five minuts waiting you will be in the game, you can make your charakter and than you're in the first town. In the backround the game continues to download itself, while you are playing.
But I think this streaming must be inside the game-engine still from the beginning of the game development.

[Manofgod]

>> But I think this streaming must be inside the game-engine still from the beginning of the game development.<<

You are exactly right about Guild Wars, it was built for streaming from the ground up. However, as stated before with Game xStream, developers don't have to think about incorporating streaming into their code. Doom 3, nor Need For Speed Underground 2 demos had them incorporated into the code, and yet they stream from the Game xStream demo system. On a standard 1.2 DSL connection start times average 12 minutes. As stated before, all Game xStream needs is a copy of the Gold Master, or a retail CD (off the shelf) to profile the product for streaming. The Game xStream system as several advantages over all other "streaming" systems:

1. No code needed for streaming process.

2. Monitors serve speeds and throttles the servers during online gaming.

3. No virtual CD-ROM loaded to begin streaming process.

4. Less than 1% game overhead.

5. Monitors users game play and intelligently delivers assets the gamer needs during game play.

6. Supports any DRM tool.

7. Can interface with any other delievery system, and be completely transparent.

8. Can Stream any type of game, including MMO's.

9. The Streaming system is "switchable" meaning that even if a user has a slower connection, they can still download the game, and then "switch" from download to streaming and back again at will.

10. Supports Save Game, and resume.

The list goes on, but I think this is enough to keep your mouths watering. ;)

[Tadhg]

Charles, I'm not so sure where your hollywood point fits in.

On a more whimsical note, does anyone else think the EXTREME name is kinda silly?

[Scott Miller]

Tadhg, I agree about the name. It's too generic, and thus not memorable or catchy.

For example, back 6-7 years ago when some company paid $7 million for Business.com, I said it was a total waste of money. Too generic. Has this once touted URL caught on? Nope. Computer.com was also once hugely sought after. And Sale.com. Buy.com still can't make money, except on ads. Anyone remember ValueAmerica.com, which had $100 million is seed money, and still couldn't buy a customer? What a horribly generic name! (I recommend the book, Dot.Bomb, which covers the behind-the-scenes story of this internet flame out. They never once considered the crappiness of their name as a contributing factor to their failure.)

Would Yahoo have caught on with a name line Search.com? Would eBay have caught on with a name like Auction.com? Would Bookstore.com have been a better choice for Amazon? Why hasn't some game company used the name VideoGame Studios? Steam is actually a great name, as it's not tricked up (like Game xStream), it's not too long (just one syllable, thus it's easy for people to type out without using a short-hand version), and it's non-generic. It also ties in well with the parent company's name, Valve.

Game xStream is making their life more difficult by sticking with a non-cool, generic, too long, tricked up name. If it weren't for the existence of Steam, I'd recommend Stream as a name, but it's only one letter away from Steam and therefore should be avoided due to likely confusion. Anyone have any name recommendations?

Names can be so important. Is it any wonder that a boxing movie named Cinderella Man is a giant flop? Shouldn't be.

[Tom Edwards]

DiStream is a good name. Definitely something based off that.

[Mark]

Damn you Scott for making my poor brain think of a name.. ;)

You need to come up with a name thats one word, easy to remember and shows what the program does.. what about: InFusion?

..my head hurts. *explode*

[Robert Howarth]

How about calling it "Stream"

;)

[Mark]

Bad example, Scott already mentioned why :)

I like your site Apache, am always dwelling on it for news.

[Robert Howarth]

ty :)

I was more kidding about the Stream thing, it would be akin to some cheesy generic grocery store soda branding like Dr. Piper or Mountain Mist.

>>A player can buy a game, and within minutes start playing it, without needing to download the entire game. For example, I can start downloading the Doom 3 demo -- one of the most content intense games currently available -- and within six minutes I'm playing it as if I have the full game, racing like a madman through the opening level, while the game continues to download itself in the background. (You can not catch up to where new content isn't available.)<<

Scott,

Are you saying that if you download and play a demo, it will download the full game in the background while you're playing, or that the games are modular and allow you to download while you play [the full version]? Maybe the wording is tricky and I'm not getting it.

[Manofgod]

Robert,

>> Are you saying that if you download and play a demo, it will download the full game in the background while you're playing, or that the games are modular and allow you to download while you play [the full version]? Maybe the wording is tricky and I'm not getting it. <<

Whether you play a demo or the full version of a game, you Play the game as it streams to your computer in the background onto your hard drive. Simple as that. No install needed, totally bypassed, but the game resides on your system.