XNA: the Future of Gaming?

Andromaton · **Andromaton**

Quoted from IGN

March 24, 2004 - I've chatted with Dean Lester a few times now. As the head man for Microsoft's Windows Gaming Group, he's talked with us several times in the past, outlining his plans for the next versions of Windows and DirectX. So when we heard that we'd be talking to him about Microsoft's next big announcement, we weren't too terribly surprised. After all, the Game Developer's Conference is a great forum for Dean to evangelize his games on Windows message. But when J Allard was also mentioned to us, we knew something was up. As head of the Xbox side of Microsoft Game Studios, he's a big name in the console community but doesn't have much of a presence on the PC.

When we were given the chance to interview the two men about the upcoming (and highly secretive) keynote address at this year's GDC, we suspected something was up. After chatting with them both for forty minutes, we knew they had big news.

Without further ado, take it away, J:

J Allard: Basically the game industry has some real challenges on their hands. Game prices haven't changed in 25 years. A game is fifty bucks and that's not going to change. What has changed are the production costs. Production costs, with each successive generation of hardware have just skyrocketed. Pac Man in 1982 sold ten million copies and cost a hundred thousand to develop. Before that there was one guy in a garage who was the designer, developer, artist, composer and tester...and also the publisher of the game.

Now we're into five- to ten-million-dollar games on average. What's worse is that a lot of them aren't coming to life; a lot of these titles are getting killed millions of dollars into development. The reason for that is that we don't have a robust platform. The way that the game industry has evolved has been chaotic in nature.

If you think about the two-year production cycle, the developers are spending the first three-quarters of that trying to get their pipeline up and running in a way that really unleashes the artist to go focus on what's important to us, which is the game design itself. So when you compress a game design itself into such a short period of time, you end up with a bad result. If you spend too much time in that production phase with little to show for it, publishers get wary, particularly if you're not a big name and your last title was a smash success and you're a little bit behind schedule. Publishers loose confidence very quickly.

When you add that up you get this negative cycle going in the industry that leads to more killed or rushed-out titles because of that pressure. When you kill a title, all the pressure for that title to perform gets added to the titles that are going to ship. It's really a tough spot. We're starting a cross-industry initiative to tackle that problem and stabilize the medium. It's our vision that if we stabilize the medium , the craft can take off. Instead of worrying about how to load the camera, and how to process the film and edit it, we can worry about where to point the camera and deliver the games that consumers want.

Dean Lester: In the past we've always been invested, as with DirectX, in making life easier for the developers. But I think the urgency has never been clearer than it is now. Next generation Windows hardware and next generation console hardware have been speculated about. The one thing everybody does know for sure is that it'll be more powerful and significantly more complex. We've seen that with this generation there are developers and publishers on the brink where they have a big bet that they took that didn't pay off due to extended development cycles, and it's damaging to them and it's taken some companies down.

The next generation ups the stakes again and we may head to many, hundred-million-dollar games. At that point there are only a few guys that can play there and, even with those guys, if one or two of those projects don't pan out, it'll take 'em down. The key thing for J and I is to guarantee the flow of triple A titles on our platforms because it's the fuel that drives the engine. With the danger that these titles start becoming few and far between -- one or two major blockbusters while everybody else just dropped by the wayside -- means that gamers aren't going to get the flow of quality titles that they need. All the air goes out of the system. For us, this isn't so much about the tools in and of themselves so much as it is making sure that we've enabled the creative developers and publishers to continue on the current trajectory and have a business that works out for them

IGN: But if you make it easier for triple-A titles to make it to shelves, isn't there a potential to muddy the market as more money gets spent on marketing?

J Allard: I don't know that that's necessarily the case; production costs aren't going to ceiling out as a result of better software. Dean was talking about advances in power and sophistication in hardware, that power and sophistication means more complexity. With each successive generation, we've utilized the hardware less and less.

Look at the first generation PS2 games. They were using 40% of the hardware on Day One. We don't want that to be the case. Look at the new 007 games. Gamers want Hollywood talent in their games. Look at online games like Rainbow Six. People want competitive and collaborative multiplayer with voice in every one of my games. Look at the visual quality in Halo and the depth of story in Half-Life 2; gamers want those as well.

What we're not saying is that people are going to spend less money on games. What we think is going to happen is that less of the money, time and energy are going to be on the basic and fundamentals of getting the game to work and getting a pipeline up. There's so much opportunity -- be it high-def gaming, online gaming, character development, story development, free design -- that it mandates we spend more time in design. What's happening today is that more time in design is turning two-year projects into four-year projects. What we want to do is make the first eighteen months of today's projects as close to free as possible so that the design time and design expense is put into the stuff that matters.

IGN: So how do you do that?

J Allard: The initiative that we're announcing is called XNA. The "X" represents that it's a cross-industry initiative that's rooted in two technology sources from Microsoft. One is DirectX, one is Xbox. The Xbox was grounded in DirectX from the beginning. That's our heritage. The "N" is next-generation. Our view is that software is the fuel for the next generation. While hardware advances are going to happen, the most important and significant advance is going to be in software. The "A" is for architecture. How do all the tools plug in to one another in a way that allows the designers to take the best tools, assemble them into a pipeline, add specific things that they want to add into that pipeline and get going with confidence that that's going to stitch together well. It's more like LEGOS and less like cutting down your own timber and trying to make your own notches fit together.

IGN: You mentioned the chaos of development. It seems like stabilizing this medium is an insurmountable goal. There are people constantly pushing the boundaries of software and hardware. How do you incorporate that into your scheme?

J Allard: That's what the architecture is for. Back in 1992, we announced Win32 on the PC. At the time, there were people saying we want to make digital cameras and color printers and Adobe was doing some very advanced things in graphics. The world was a pretty chaotic place and we didn't have a framework to plug all those things together so that PhotoShop works on the corporate net or drag and drop numbers from an Excel sheet into a mainframe. So we set up an architecture and worked with all these tool and application vendors and rationalized it. To think that it happens overnight is misleading. We announced Win32 in 1992; the first operating system to support it was in 1995. We've evolved Win32 for the last decade.

In a similar vein, this represents a starting point for the XNA initiative -- it's a call to arms to start working together on this. We define some protocols, APIs, interfaces, and formats that we all agree on.

Let's pick out some examples: Imagine a world where you take Tim Sweeney's game engine from Epic. It's the most advanced and applicable to your game. And you can stitch in the Havoc physics engine to that real easily. Half my artists use Max and the level designers use Maya and I can tighten that into the work flow and I have a particle system that I want to create on my own and I want to make sure that that snaps in. And, by the way, when I'm tuning the game and doing builds, I want to make sure that it works on Windows and Xbox. That's a lot of custom code.

Today we want to sit down with all the vendors that play in those spaces and really define some interfaces. If you think of the metaphor that those are the building blocks or bricks, we want to provide the mortar and add a little dose of leadership. If you can make those bricks fit together you'll really have a stable development environment.

IGN: Who are you technology partners then? How do you handle licensing?

J Allard: The licensing thing will be handled partner-to-partner. Some vendors will go Source-code license, some will go binary obejet6 only, some will do a hybrid, and some will do custom deals. I think it's imperative that you leave the tools' system in a very competitive state. It's a meritocracy today. No one game is developed on one tool and that's not going to change. We just want to make it a lot easier. The business model will remain very different depending on who you are and how you're licensing.

I can be clear about our business model. We're not building a tools business -- it's about licenses. On Dean's side of the house, it's about selling Windows licenses; one my side of the house, it's about collecting game licenses. That's our business model. By rationalizing our platforms, we're hoping to increase the flow of triple-A content on both platforms and allow the creators to take the most advantage of their respective platforms.

IGN: How do Sony and Nintendo fit in to this?

J Allard: They haven't called.

Dean Lester: You'll have to ask them. We've come from a generation of consoles where we had the most powerful hardware. People have different opinions of how everybody did, but I think everybody agreed that the Xbox was, ultimately and in terms of potential, the most powerful platform from a hardware point-of-view.

IGN: Not counting PCs.

Dean Lester: Not counting PCs. I'm talking strictly about the console arena now. On the console side, Xbox was the most powerful but, when you looked at it, it was the value of the software that really was the moist important thing. The development environment made a big difference in terms of the success of each of the developers. To come from a place where hardware was one of the areas we could point as leaders to recognize now that it's all about the software is a pretty powerful message. Everybody could invests significantly in more complex and more powerful hardware but if you haven't made the appropriate investment in software, it's almost all for nothing. Those are questions you should ask those guys. We see that: hardware without a software enabler, the hardware sits there unused or development takes forever to get the vision realized.

IGN: Let's ask about that investment then. Nvidia made a big push to tell developers what they could do with the hardware. They felt that performance wasn't being utilized. What's your approach? How are you going to show developers how to use this new mortar?

Dean Lester: J has a developer relations group that are always out and engaging with developers and I have my team of third-party evangelists that are doing the same. In some ways this isn't so much an evangelism job as it is that they've told us about the problems they're facing. This is more of a response. The feedback we've had on this is, "Thank you for listening," and, "Why did it take you so long?" I don't think anybody in the audience sits there and says it's the first time they've ever thought of this. They live with this every single day.

For us to let them know we're working on an end-to-end solution that removes the barriers and lowers the cost of development while keeping the quality high, their reaction is "bring it on." The concern that they would have is whether this would limit the choice of tools. Absolutely not. No one's going to use our stuff exclusively. We're going to work with RenderWare and all these other solution providers. Is this a Microsoft-only initiative? Absolutely not.

The other question is "When am I going to get this stuff?" We've already delivered some of it, this summer will see the next wave and next year we have the next phase, phase three of the XNA deliverables. This is a ten-year investment. It's just going to keep coming.

IGN: When you say you've already delivered some of that, what do you mean exactly?

Dean Lester: You can look at DirectX in its current form as the beginning, one of the foundations, of taking a complex hardware environment and abstracting it down to a virtual single device. You write to Dreict3D as if it was a device and we take care of the abstraction under that.

We don't want to be arrogant and claim that we've invented something brand new that the industry has never seen before. The truth is that we've been working towards goals like this before. The difference here is the alignment between Xbox and Windows is a big part of this. The absolute commitment is the second piece. The third piece is the sense of urgency that, if we don't get this up and running and get everybody participating, the next generation of hardware could be the one to put the smaller publishers and developers out of business. They won't be able to afford to keep their business going. And that has ramifications for everybody -- gamers, developers, publishers and platform providers.

IGN: What room is there in your model for garage development? There aren't as many but there are still a few games being developed by one, two, maybe five or six people. How do they fit into the new model? Do you just assume they'll keep at it and one day catch up with you?

J Allard: I absolutely hope so. I want to see Clerks to use a movie example. In the movie industry we were building eighty-million, hundred-and-fifty million, two-hundred-million dollar films when Kevin Smith maxed out his credit card, stole his mom's camcorder and made a great movie. We want to see more of that. If you combine the tools pipeline with the opportunity to go and have earlier dialogue with publishers and have them checkpoint along the way and really see progress, a lot of these games from these small studios that are ambitious in nature, publishers have to see the breakthrough before they're willing to bet.

If it takes 12 or 18 months to really demonstrate that it's fun and has a hook, it'll get killed. Dean and I were talking at dinner the other night and thought maybe 10 Grand Theft Autos have been killed in the last two or three years. Grand Theft Auto-style breakthroughs in game design have been killed because they haven't got the publisher quickly enough. The smaller developers absolutely fit.

We're seeing a lot of pressure on medium-sized developers today. And it's not just the little guys. We've killed projects internally that have been three to five million dollars in. That's not a little development team.

Today is really a call to arms as Dean said. It's our commitment to bring the platform forward. We've been doing it. Xbox has shipped 25 version of the dev kit. We ship it every month and it keeps getting better and better and we add more and more tools based on developers' needs.

Step one is a call to arms. Let's go build an architecture together so these things plug and play and creators can focus on the creative part and not the plumbing. Step two is to get our own house in order. We've got a bunch of tools on the Xbox side and a bunch of tools on the Windows side that game developers love. They're not available on the other platform. Step three is to commit to XNA for the long term and say that, year over year, XNA will be a continuous innovation engine that lets you harness the potential hardware. We don't want to be disruptive.

If you're a PSX programmer and then you're a PS2 programmer, holy crap, you have to throw away all your code and everything you knew and go from 95% effective utilization of hardware to 40%. That's why the first generation games didn't look so good. They looked like great PSX games.

We don't want that to happen when we go through the quantum on our hardware platforms. We want you to be building XNA games today and when new hardware comes out, you're still building an XNA game but there's more to take advantage of. Like Adobe was writing an application for Win95 and then WinNT came out there were special features they could take advantages of -- they didn't have to throw it all away and start again.

IGN: Let me clarify. In your quest to develop XNA as a standard, you're not advocating a one-size-fits-all approach. You're allowing developers to choose a suite of tools?

J Allard: Yeah, you're my general contractor and we're building a house together. What do you do? You go down to Sears and you buy a Stanley hammer, a Black and Decker saw, you're gonna buy a MAKITA drill, you're gonna mix and match tools that are appropriate for you. When you walk out the door, hopefully you don't have to make another trip. If you're remodeling the kitchen instead of building a new house you'll make a different set of selections; your budget is different and the scale of your team might be different. Every GC's trip to Sears looks a little different but Sears arranges everything on pegs and you're guaranteed that your power tools will all work in a three-prong outlet. There are standards in terms of nails and hammers and weights and gauges that allow interoperability between the guys that make the screws and the guys that make the screwdrivers.

If the three of us want to make a game, where's Sears? There isn't. We send you out to the Black and Decker store, and Dean to the Makita store and I go to the Craftsmen store. When we come back, hopefully, we have enough stuff to build the house but we'll probably have to make a few return trips. And then none of the stuff plugs into the same outlets. We have to rewire the freaking house before we even start to remodel. That's what game development is like today.

Dean Lester: The key technologies that we're starting with, in response to customer feedback, is Xact, the audio tool for Xbox. It's been well received because it allows the composer to do the entire job rather than do some work and hand it off to a developer for coding or processing. It's very empowering to the artists. People love it and have used it since the beginning of Xbox. Our feedback has been, "Why isn't there a Windows version of this?" We're delivering that this summer.

The second one is PIX which addresses whether or not you're getting the most out of your hardware. This is a very detailed shader analysis tool that lets you see where the bottlenecks are in terms of performance and what efficiencies can be gained in terms of graphics and pixel rendering. That, again, is something from Xbox and we've been asked why we don't have a Windows version of the same tool. This summer, we will.

The next one is something that you and I have talked about in the past but I can put some flesh around the bones now -- a common control standard. One of the things holding back the Windows side is that keyboard and mouse are great for certain types of games but they're not appropriate for all types of games. Direct input made a good go at it, trying to make a heterogeneous landscape of controllers appear as one, similar to what we did with graphics. The new direction moving forward is that Windows and Xbox will share a family of controllers that will be exactly the same, literally. I can unplug it from one and plug it in the other and it will just work.

IGN: So start to finish, assuming someone hears about this and wants to utilize, when could we realistically see those titles appear?

Dean Lester: Again, in some ways, they've been using some things like Visual Studio, DirectX, or HLSL they've already used some of the existing building blocks that are part of this broader initiative. We almost don't care if it's a definitive use of XNA or not. What we want to hear from the developer is, "Thanks for the tools. Our game got done on time and we were able to deliver everything we need."

It's open to interpretation. I could be very aggressive and say any DirectX game ever developed was a precursor to an XNA solution. Once you start to see Windows games using Xact, once they start using common controllers, once Xbox games start using HLSL, then you can say we've come to a sort of fulfillment of one of the visions. You guys will have to decide what you think is genuine XNA pedigree.

Our goal isn't to create a tools business. Our goal is to anticipate an upcoming crisis and provide a runway that allows developers to just keep on going. It's a good question. In our mind, we'll know we failed when we see developers going out of business and they say they couldn't get the game going fast enough to reassure publishers. When we see the problems getting addressed and people thanking us, that's when we know it works.

J Allard: I totally agree with everything Dean just said. XNA isn't XNA 1.0. It's not a product release; it's a constant evolution. This holiday titles will show that Microsoft listened to my needs and helped me out -- particularly those that are moving between Xbox and Windows or trying to ship on both platforms. Next year people will say that XNA has helped bring the tools pipeline up a lot quicker. It feels more like Sears; all the pegs might not be full but it really has helped integrate people who are familiar with different toolsets. The year after that you'll have developers up to speed on it and, as the hardware technology is shifting, they'll thank XNA for helping them weather that storm and take advantage of more capabilities on the hardware.

IGN: Not to call your motivations into question but why are you being so generous? Why not keep this pipeline just for Microsoft Game Studios?

Dean Lester: We kind of summed that up when we described how we get paid. Ultimately the Windows business is about the customer seeing enough value to upgrade their machine and their version of Windows. J's business is about making sure great games are being sold ion his platform. That's the point at which either of us get paid -- when the customer decides there's enough value on this platform to warrant spending more money.

They won't spend money on a platform in isolation. People don't buy platforms and, frankly , they don't buy hardware. What they buy is a solution to get to the software experience that they want. If those games aren't available on console or Windows, the business model doesn't work. No one's going to upgrade to Longhorn or buy the next Xbox if they don't see great content available. That's the motivation here. If we can't find a way to bridge today's developers and publishers to the next generation, the games won't get delivered, the gamers won't be buying and the platforms will stagnate. That's not just true for Windows and Xbox. That's true for everybody.

Do we have our own motivation here? Absolutely. But it's not to create a fabulous giant tools business. It's to create content that makes people say it was worth buying a console. We've heard people say many times, "If I never play another game on this, I got my money's worth out of that triple-A game." If those aren't coming out, the platforms are worth nothing.

J Allard: I'll get a little greedier. 40% of America and a smaller fraction of the rest of the world plays games today. Dean's talking about keeping the business where it is. I say, "Who cares about the 40% that play games? Let's go get a chunk of the 60% that don't."

It's the next Myst. It wasn't the highest costing game. I seem to remember they pitched the idea on HyperCard and it was a super cheap game to develop and get an iteration going. Then they threw some money at art. If we can get the game development environment to be more HyperCard like and get those real creative breakthrough ideas on our platform and catalyze growth, great! Not only area we doing a great thing for the industry, but hopefully that content that catalyzed that growth is exclusive on our platform. That can help as well.

Dean, did you tell him we're bringing Live over to the PC?

Dean Lester: No, I didn't get to that. Since we brought Live to the Xbox was when are we going to be able to play across Xbox and Windows? The XNA initiative covers tools, technologies, peripherals and services. Live is the first of those services. It will allow Xbox and Windows users to have equivalent functionality and lots of interesting scenarios between them.

IGN: Will they be compatible across platforms for the end-user?

J Allard: We're going to work with our partners to figure that out. We're responding to developer needs. They say this Live thing is big. The voice across all games, the single log-on, the security, the billing and hosting work, it's awesome. They want to build on top of that for their PC games. This is one of those areas where XNA is evolutionary. We sit down with developers to ask about the requirements for their next ten games. One thing people have talked to us about, and I think it's great, is the notion of having the Xbox and PC play together and play to their strengths. Imagine the RTS game on the PC where you're the general and you're deploying troops and managing resources. The guys on the console are the troops, playing a real-time action/adventure game. They're going out and beating each other with clubs or storming the castle.

But you're taking advantage of the best of both. The PC guy has a 16X12 screen with a keyboard and mouse. He thinks he's god but he has to motivate the troops now. He has to communicate and have clarity of strategy. From the troops' point of view, it's not some scripted action. It's different based on who's playing general and what their intentions are. I look at that and think that's a huge gaming breakthrough. I want to be part of that. Live and XNA in general unlock those kinds of breakthroughs in thinking and game design that will advance both platforms. Create the genre I just described and have two or three runaway hits along those lines? Damn, that's good for Windows; damn, that's good for Xbox; damn, that's good for gaming.

Dean Lester: I love that idea because the Windows guy is in charge of the Xbox guys. That really plays to my sensibilities.

J Allard: It's always been that way at this company, baby.

Dean Lester: I issue the orders and the Xbox guys do it.

J Allard: I know where my bread is buttered.

IGN: So you guys are just doing this so you can play this game?

Dean Lester: J's a frustrated game designer.

IGN: Aren't we all?

Andromaton · **Andromaton**

Don't forget to check out the XNA demos here

Martin_the_Wise · **Martin_the_Wise**

Its a gamble, but if it works, gaming will never be the same.

And Microsoft will truly rule the world.
_________________
Sometimes, we forget who we are.

counterparadox · **counterparadox**

...Sounds like a shitload of fluff to me.

Here's what I heard "We're trying to make things simpler."

Ok. So? What, no one else tries to do that?

No, I'm not being cynical about Microsoft. Just listening to what they said, they spat out the same one statistic 3 times, the gave it a catchy, edgy name (XNA), it's all a big marketing ploy. Seriously, all I heard was "we're using buzzwords to enforce our marketing plan."

I honestly don't see what the big deal is. Sure, it's nice to think that smaller developers can be successful, and nice to think Microsoft will help them (hell, if they're right, that's fantastic) but it really seems like they just put a marketing spin on what they were already doing.
_________________
anime is teh s uck

Play City of Heroes/Villians? Look me up, Pinnacle server, @C Paradox

John_Bono_Smithy_Satchmo · **John_Bono_Smithy_Satchmo**

Can anyone give me the cliff notes?
_________________
This space left intentionally blank.

Goldfinger2K · **Goldfinger2K**

dougisfunny · **dougisfunny**

i tried reading it, but there was too much poppycock.
_________________
The problem with America is stupidity. I'm not saying there should be a capital punishment for stupidity, but why don't we just take the safety labels off of everything and let the problem solve itself?

ToonamiL · **ToonamiL**

Wait, could someone summarize this because I don't want to read anymore today (blame Antigone)?

FinalDivineDragoon · **FinalDivineDragoon**

"We are Microsoft. We will make the gaming world better by monopolizing it. Give us all of your money as you are swayed by these newfangled monachers, anagrams, and analogies."

Good enough summary? /end sarcasm
_________________
There is limited intelligence in the galaxy, but the stupidity of the universe is infinite.

ToonamiL · **ToonamiL**

Not really. Is it going to be on the X-box?

overdrive535 · **overdrive535**

I don't see any sarcasm.

As far as I read, it's only for the XBox, ToonamiL

Nobuyuki · **Nobuyuki**

dougisfunny · **dougisfunny**

Nobuyuki · **Nobuyuki**

overdrive535 · **overdrive535**

So it's just a ploy to get designers who normally just build games for the GC/PS2?

Or am I reading too much into this...
Just wondering...
_________________
"If you've ever had your nipple bit off by a beaver, you might just be a redneck..."
Jeff Foxworthy


	Disclaimer: Toonami Infolink, its creators and their kin are not owned by, affiliated with, or bossed around by Williams Street, Cartoon Network, Turner Broadcasting, AOL Time Warner, Long John Silvers, Tremont Corp (they do something with titanium), or the Corporation for Public Broadcasting. All opinions expressed on Toonami Infolink are those of the writer, and do not necessarily reflect those of Dave Coulier, George Stephanopoulos, Selma Hayak, Mark Mothersbaugh, or Ron Santos. PHP-Nuke Copyright © 2005 by Francisco Burzi. This is free software, and you may redistribute it under the GPL. PHP-Nuke comes with absolutely no warranty, for details, see the license. Page Generation: 0.10 Seconds