The Future of Microsoft Gaming

Ars Technica has an interview up with Matt Lee, a software developer in Microsoft's Xbox division. He's got a lot to say on the subject of the future of MGS gaming. He touches briefly on Xbox Live, Games for Windows, and the powerhouse that is the 360. From the article: "The tessellator in the Xbox 360 GPU is indeed a very powerful piece of hardware, and you're right--most games have yet to take advantage of this. I think you'll see more titles use it in the future. As for procedurally generated worlds, I believe the biggest obstacle to overcome is how to design and build the content for such a system--it can be quite a departure from today's art pipelines. Game studios will figure it out though--it's crucial to generating and delivering ever larger worlds without having to exponentially grow the size of the art team."

Comments

Impressive at Best...

[UberMench]

I do have to say that Microsoft's plan for gaming interCONNECTIVITY is quite impressive. Their press conference at E3 really demonstrated how cool connecting a Vista PC to a 360 to a Cell Phone could actually be. The portability of the GamerTag is truly awesome, but I'm still not sure if it will be enough to take the #1 spot from Sony. (Wii is in a class all its own.)

Re:Impressive at Best...

[Foofoobar]

One virus to rule them all...

Re:Impressive at Best...

[Zediker]

While I can understand where you are comming from, it will be a long time before someone releases a virus that can attack 3 different architectures at the same time. Even viruses that attack two different operating systems on the same harware architecture are rare. One that attacks 3 different operating systems, on 3 different architectures would be very difficult to write at best. Im not saying that it cant be done, but I am saying I dont expect the average virus coder will want to undertake such a challenge.

Re:Impressive at Best...

[stonefoz]

If only I had mod points. Everyone has seen what Micro$ofts intconnectivity does, and if dotnet moves to all three... bingo.

Re:Impressive at Best...

[Foofoobar]

Hardware won't matter much if the software all acts the same. Files are files are files and if they are using NTFS or FAT, the viruses will be able to use the TCP/IP connection or crawl the network and infect files just the same.

HOWEVER... you are right in the fact that the different hardware architecture will limit what it can do. But if they all use a similar language to communicate (as I bet they do), it won't be hard to have the virus use the languages built in functionality to exploit the hardware.

Re:Impressive at Best...

[Anonymous Coward]

You don't have any idea what you're talking about, do you?

Re:Impressive at Best...

[Foofoobar]

Wow. 3rd grade taunts. In case you hadn't noticed, this is a tech forum and criticism without a supporting argument is commonly referred to as 'trolling'; this is a common habit by persons lacking the intellect themselves to understand the conflict at hand. So they lash out due to some inner need for attention.

Is that what you want? Should I give you some attention now? Would that make you happy?

Listen to Will Wright

[Roy van Rijn]

The biggest problem is that the user is expecting bigger worlds with more nice stuff to look at every new game. The problem is creating this huge content.

This is just what Will Wright is solving with Spore for example. What is the thing people like more then watching all this content? Making it! Thats why his games let people create it in a smart way instead of the developers/artists.
And the biggest advantage is that he is letting people share their content creating a huge repository for people to get all their ideas/content.

That is (IMHO) the future of gaming, and hopefully for them, the future of Microsoft gaming.

Re:Listen to Will Wright

[AuMatar]

Some people like making cotent. Others don't. If they did, the biggest games in the world would be Second Life and A Tale in the Desert.

Re:Listen to Will Wright

[Osty]

The biggest problem is that the user is expecting bigger worlds with more nice stuff to look at every new game. The problem is creating this huge content.

I know this is Slashdot, so you obviously didn't read the article. Did you even read the summary, though? This bit was quoted in part right there in the summary:

The tessellator in the Xbox 360 GPU is indeed a very powerful piece of hardware, and you're right--most games have yet to take advantage of this. I think you'll see more titles use it in the future. As for procedurally generated worlds, I believe the biggest obstacle to overcome is how to design and build the content for such a system--it can be quite a departure from today's art pipelines. Game studios will figure it out though--it's crucial to generating and delivering ever larger worlds without having to exponentially grow the size of the art team.

Right now, the best example of using procedurally-generated worlds is a third-party bit of middleware "SpeedTree," which was used to generate all the trees in Oblivion, although this was not done on the fly. This cuts down a lot of work for the artists, who then don't have to spend days making dozens and dozens of slightly different tree models. The idea of using on-the-fly procedural synthesis is one that even Microsoft hasn't explored yet.

In other words, Microsoft and third-parties already have ways to procedurally generate textures and models so that an artist doesn't have to toil over "dozens and dozens of slightly different tree models", for example. Adoption of such techniques will only increase as games get ever more complex.

This is just what Will Wright is solving with Spore for example. What is the thing people like more then watching all this content? Making it! Thats why his games let people create it in a smart way instead of the developers/artists.

That's all well and good, but it's a limited approach. The obvious issue is that it works best for PCs, where people have the full power of a PC (and thus developer- and community-provided tools) to build mod content. That's not going to work nearly so well on a console. More importantly, IMHO, is that the developer still has to provide some content to get you going, or most people won't stick around long enough to create their own. Finally, this doesn't necessarily work for all genres. It'll work with Spore because that's a sandbox game. It wouldn't work so well with Half-Life or Halo, because the single-player portions of those are too important (you can get community involvement for multiplayer, but the developers are still going to have to produce huge amounts of high-quality art for the single-player experience).

That is (IMHO) the future of gaming, and hopefully for them, the future of Microsoft gaming.

I disagree. This is part of the future of gaming, but I really hope it's not the whole future. That's like saying blogs are the future of book publishing, or YouTube is the future of movies. It's great to give your users/fans/players a way to contribute back, but unless every game is just another Sims sandbox (boring!), we'll still need the "masters" (as in, professionals who are dedicated to their art, with the resources to produce outstanding work) to provide us with compelling stories, graphics, AI and behaviors, etc. Some of that may be procedurally generated, but at the root of that should always be a professional artist.

Re:Listen to Will Wright

[Roy van Rijn]

Well true, of course I didn't read the article :)
And of course I meant its a part of the future of gaming, you can't say what the future of gaming is.

There will always be people who love creating stuff, and others that don't, but the biggest advantage of Spore for example is that you can easily use stuff made by other people.

You can already see that in almost all parts of gaming, shooters have map-editors, some just play in those maps, some make them etc.

Re:Listen to Will Wright

[Emetophobe]

I'd say the biggest problem is the push for better and better graphics, neglecting the gameplay. Some games still manage to produce gorgeous worlds with awesome gameplay, but most seem to neglect the latter. I've enjoyed UT2004 for over a year because the gameplay is awesome and has nice graphics to boot. I was disapointed with Doom3 and HalfLife2 which had arguable much better graphics but lacked multiplayer gameplay. HalfLife2 didn't even come with a CTF mode built in for multiplayer, only deathmatch! Welcome to 1990! Doom3's multiplayer support was even worse, like a last minute addon (which it pretty much was). Some might argue that it's hard to compare UT2004 to HalfLife2 and Doom3 since UT2004 is primarly a multiplayer game (singleplayer mode is just bot fighting), and Doom3 and HalfLife2 focused more on the single player mode. Maybe that's why I find UT2004 has more replay value, once I finished doom3 and halflife2 a couple times I don't really have the desire to play it again (the story is decent at best).

4 more years! 4 more years!

[Traiklin]

most games have yet to take advantage of this. I think you'll see more titles use it in the future.

all 4 years of it? or will they decide that 4 is far to long for the system and knock it down to 3 years? then the one after that would be 2 untill we are buying a new console every single year! :P

Re:4 more years! 4 more years!

[Osty]

all 4 years of it? or will they decide that 4 is far to long for the system and knock it down to 3 years? then the one after that would be 2 untill we are buying a new console every single year! :P

Well, the 360 is designed to eventually break-even or even make money, which the Xbox was not. So, Microsoft will not be as pressed to get the Xbox 3/720/Next/whatever out in another 4 years. And if all goes well (poorly, that is) with the PS3, Microsoft will be in a much more comfortable position and not have to fight for first-mover status like they did with the 360.

That said, even during the Xbox's four years, you could visibly see developers making advancements in technology. A good example is Halo vs. Halo 2. Not only did things get shinier/bumpier (adding more detail through bump mapping techniques rather than geometry), they also were able to render in widescreen while the original didn't. If you compare Project: Gotham Racing to PGR 2, you'll see similar changes -- better trackside textures and models, better textures all around, better car models and environment maps, etc. And then compare PGR2 to Forza, and you'll see even more differences (though some are under the covers, such as the 4-wheel independent suspension physics simulation used in Forza, compared to the standard of Pacejka's Magic Formula). It's certainly not as much of an improvement as we saw through the life of the Playstation (compare Final Fantasy VII to Final Fantasy IX), but it's still there and once would expect similar advances on the 360 as developers get used to the technology.

MGS4

[ZakuSage]

From TFA:He's got a lot to say on the subject of the future of MGS gaming.

Yeah, Guns of the Patriots is going to rock! :P

Re:MGS4

[poot_rootbeer]

He's got a lot to say on the subject of the future of MGS gaming.

He does, but how much of it will turn out to be accurate?

I tried to read the article all the way through, but kept getting distracted by something that was in the first couple pages. Matt Lee got his Masters degree in 2001, only five years ago. And he's already the go-to guy for pretty much the entire Microsoft gaming platform as far as graphics optimization goes, at least going by his description of his job duties.

Even if he is a technological genius, which I have no reason to doubt he is, should someone with such limited experience be in such a multiscient, guru-like role?

I dunno, if I was seeking advice on 3-D graphics, I'd prefer to talk to someone who was around in the days where it took an hour to render one frame of flat-shaded polygons. They're simply more likely to be familiar with more approaches to problem solving.

Re:MGS4

[Nossie]

I do agree...

but you also get the flip side where the real gurus become stuck in their ways and/or wont touch a 3D workstation unless its SGi.

I wouldnt buy an xbox.... but then I'm a Nintendo fanboy and hate large conglomerates (sony, microsoft) trying to monopolize all the pies... but this guy with his lowly five years commercial experience could bring a new twist to the world where a guy with 20 years experience might have went stale 10 years ago.

I'm not trying to diss the experienced here at all... if an ex member of (the real) bullfrog wanted to develop games for my company then I'd be more likely to hire him than someone with less experience... but if the new guy impressed me more then why the hell not?

Procedural gaming, in demo form

[SirSlud]

http://games.slashdot.org/games/04/04/15/1239203.s html?tid=127&tid=186&tid=204

Under 100kb of code creates a fairly rich, neat demonstration of procedural game content.

Procedural is definately one way the industry is leaning, but its not the end all be all. Testing collision related bugs in games that has procedurally created collision requires some concessions to be made in terms of the game design. Its tough to create a game where content is created dynamically, but doesn't create situations where the player can get stuck, or produce other similar 'progression stopper' kind of bugs.

SpeedTree works in wide open environments, but indoors, in tight quarters, procedural content is a whole different bag. I think the biggest potential is in creating procedural textures that ensure no two places look exactly alike. But as with any new approach, procedurally generated content provides a whole new set of challenges and cons.

Re:Procedural gaming, in demo form

[twistedsymphony]

While there are problems now It can only get better the more it gets used and more techniques and tools become available.

If anything I think it works best for ORGANIC items. That is to say it might not be the best idea to make a building; though a procedurally generated texture layer to generate a weathered effect on the shingles might be cool. Then you could use the same small generic repetitive texture for all the shingles and simply procedurally generate the effects that make them look unique. Procedural generation for the bushes outside or the potted plant inside might also be cool. You could even go as far as to procedurally generate the population of the town where no 2 NPCs looked alike. I don't know if we'll ever see it working properly for level design where ridged play tested and static boundaries need to exist.

I think we'll see the best use of this when they start using it (well) for animation and sound effects. Never see your character move in the same fashion twice or notice the visible hiccup between animation routines. Also using it for sound effects would make for much more realistic repetitive noises like footsteps and machine-gun fire. I've seen some examples of animation and they look OK, some of them look like drunk puppets but some look pretty damn good. Using it for sounds and textures however would be fantastic for spaces savings.

Re:Procedural gaming, in demo form

[SheeEttin]

You know, it's extremely possible. In fact, it's been done.

Re:Procedural gaming, in demo form

[SirSlud]

Jesus, you have to be kidding me.

Not to take anything away from nethack, which is cool and fun and was groundbreaking, but are really comparing creating 'collision' in the form of a 2d small-res grid with creating fully 3d worlds with collision including jump, crouch, object interaction, varying other mechanics that may include riding horses, swimming, climbing, etc?

What do expect me to say? Oh shit, nethack! I guess we might as well go home and wait for BF2, Quake, Oblivion, etc to stop selling because clearly customers dont recognize that nethack made them all obsolete 20 years ago?

And before you tell me that those games didn't produce procedural content, spawning any objects with any physics programmatically can count as 'prodedural' ... I was pointing out that producing procedural level data using todays physics engines is a FAR more complicated task. I think nethack's creator would find your assertion overly simplistic.

Re:Procedural gaming, in demo form

[SheeEttin]

Actually, I was trying to be overly simplistic. After all, the Wright Brothers didn't invent Boeings, now did they?

nice interview. Could almost make out the slurping

[Anonymous Coward]

sounds.

I've heard that one before.

[KDR_11k]

(Realtime) Tesselation was a feature in Ati cards since the 8500. How many games made use of it and how many don't look ugly when they do? Tesselation reduces the artist's control over the mesh and only makes sense if you have a severe bottleneck in the system preventing you from having models use this many polygons right away without any post processing beyond what's found in the file. From what I heard the 360 does suffer from bottlenecks like the relatively slow DVD drive, overall tesselation is not the future, it's a workaround.

Re:I've heard that one before.

[apoc06]

you seem to be the first person i found thats agreed with me on the subject.

its a workaround. a workaround used to speed up the development process. if time were not a factor, they would still require procedural content. in a perfect world the artists would have control over all content and asset minutiae. due to the increasing focus on deadlines and the limitations of the hardware, this is basically the only way the x360 can advance.

due to the space constraints of HD material on DVD9 and the lack of ability to require a harddrive for games, the xbox360 will need advancements in generating procedurally generated content in order to advance. MS is pushing for this. because the bottlenecks as you put them are firmly in place and have been tested already within the first couple of months. not a particularly good sign.

the PS3 seems to want to set its bottlenecks slightly higher. they want to give developers space for content storage [blu-ray]. they want to give them the ability to count on having a harddrive [ to cache data and speed up loading times]. the want to give us supposedly cutting edge superpowered nuclear cell chips and video card so powerful that its nickname should be HAL9000, just to make sure we can crunch and process all the graphical/physical data. [ excuse the sarcasm] thats great, but that takes time and comes at an expensive cost. one that's dirtied the playstation name in the eyes of many.

the wii on the other hand sees this as a losing battle. they can wait for procedural techniques to advance more before they charge ahead and push the line. they are going in the opposite direction. unfortunately this is only setting their bottlenecks lower. they are gambling on using a new form of gameplay control to offset that.

who will it work out for best? i guess we will have to see what happens in two years' time.

Ala 'Second Life' perhaps?

[Anonymous Coward]

I'm not sure if I get the gist of this conversation but is the player created content used in the MMO "Second Life" something along this line? In other words, is a game like that the expected future of gaming?

Powerful GPU?

[Draracle]

The article talks about procedurally generated worlds and even references Speed Tree used in Oblivion as a pseudo example. When Oblivion came out I was impressed with the graphics but quickly I realised that this game had been dumbed down for the Xbox. First, I kept reading reports of Xbox users having fps problems when riding the fastest horses. Second the whole world was rendering far below my PC's abilities. So I tweak the game out, maxed out draw distance, increased the tree count, and retextured the landscape and the sky. When I finally added dynamic indoor lighting and night lighting in the towns my system took a noticeable dive in fps -- an my system isn't that great (with the exception of my 7800 (agp) card and 512 stick, everything was purchased years before the 360 release). Maybe Oblivion's code wasn't optimised for the Xbox's gpu but based on that limited experience I think Oblivion has already tested the limits of the Xbox's GPU.

I've thought for a while now...

[QJimbo]

that procedral gaming was the future. Imagine if GTA generated it's own cities for example, and countryside inbetween etc. it would be really cool. This sort of thing would really open up the exploration aspect in games.