EXCLUSIVE: EA REVEALS NEED FOR SPEED NEXT-GEN

hey69

i have a monster
Veteran
source : CVG



"Your jaw will drop on the floor!" First amazing details on how the publishing giant is working in tandem with Hollywood to bring its massive franchise onto the next-generation of gaming hardware

18:51 While many of you will already be frothing at the mouth at the prospect of Need For Speed Underground 2, the enormous, Grand Theft Auto-inspired follow-up to one of last year's biggest titles, Electronic Arts has dropped a petrol-powered bombshell, revealing first plans for the next generation of its killer franchise.
Speaking to CVG at EA's salubrious Vancouver headquarters, Underground 2 producer Chuck Osieja confirmed that a Need For Speed title was already in development for next-gen systems, stating: "I think what we're working on is going to revolutionise the way people look at driving games."

In response to a question on what he believed the next-generation of console hardware would allow for in the racing genre, Osieja said: "I'm not saying specifically because we're already working on it.

"If we pull it off, we'll do for driving games what Gran Turismo did for driving games on PS2 and PSone. It'll be that amazing in terms of visual quality and what the experience is for the player." Excited yet?

After revealing the stunning news that development for next-generation systems is already well underway, Osieja began to lift the lid on what gamers can expect from the next serious evolutionary step in the racing genre:

"Since we are getting a head start on that [creating a revolutionary next-gen product], I can't reveal too much but one of the things we think about is, it's not going to be photo-realism, right? The next-generation systems can throw 30,000 polygons into a car and guess what, it doesn't look any better than a 10,000 polygon car with the rendering techniques I can do.

"It's what I can do to gameplay, and what is the experience I can change for the player on a next-generation system to make those people a) want to invest in that system and b) buy that game. So we're spending a lot of time now trying to figure out what those next steps are... Needless to say, even in the lifetime of these systems, Xbox and PS2, there's a lot of things we couldn't get to because the technology couldn't do it."

Having added his own voice to a growing body of opinion that next-gen will no longer place such emphasis on creating visual splendour - since all next-gen games should easily achieve a high level of quality, Osieja gave insight into how his team is using development kits to change to face of the racing genre:

"We're seeing with the specs of the new hardware we can do some pretty amazing stuff; for us it's figuring out how to harness the laws of nature in basically a 2D gaming environment. Even though it's a full 3D environment on screen you're still limited by a TV screen, so how do you harness those things that you experience naturally in the world and translate them to a videogame experience?

"Needless to say it's really, really exciting and I don't know if anyone else is going to do it, but I think EA, well, we're putting a lot of effort into it. I think it's going to be pretty cool." Hell, yeah.

Osieja also revealed how next-generation hardware is already facilitating the unavoidable convergence of games and movies, detailing how EA is working closely with Hollywood talent on its projects: "That's been one of the coolest things about bringing in all the Hollywood talent. I know people probably roll their eyes when we say we're bringing in these guys from Hollywood, but what's cool is the technology of the next-gen systems is allowing them to realise some of the stuff they're doing in the movies right now, and we'll be able to do it real-time.

"We've benefited a lot from things like The Lord of the Rings, and obviously Underground and Bond all benefited from the knowledge of the Hollywood people who have come in and just helped us with tweaking lighting and making sure that things look right to the eye, whether it be [something like] the camera shake, which I think we finally did really well in Underground last year. Now they're being able to take that stuff and apply it to real-time, next-gen systems."

One of the only indicators of the potential power of next generation gaming hardware seen by the public at large has been Microsoft's demonstration of its XNA next-gen software development platform - most famous is the "Crash" demo, showing a car being smashed into a wall, then disintegrating in a unnervingly realistic manner. (You can check the demos out in all their glory right here).

We asked Osieja outright: with Need For Speed next-gen, are we talking the same level of quality seen in the "Crash" demo, or beyond that?

"Easily that. We just looked at a video the other day of a Porsche GT flying up in the air, floating over and basically coming apart - and we'll be able to do that now, with our dev stations. But that's a part of it. At the end of the day the licensor is always very touchy about what you do to their product. I'd love to blow up their car, I'd love to set it on fire, I'd love to do all this stuff and they say you can't.

"Even though we can do a lot of stuff we're still going to be restricted by what the licensors allows us to do. But yeah we can do that stuff."

Out head's duly spinning at the prospect, Osieja wished to leave us in no doubt as to the quality of what EA is working on right now in secret: "If I was able to show you some of the demos we've seen, your jaw would literally drop on the floor. We're going to be able to do that stuff in real-time and it's going to be amazing."

Bring it on.
 
Oh dear. Same old shit. Every new generation brings out the best in publisher's PR skills... Ever since the NES-SNES days...

It's always the same "Well, we did a lot this gen but there are some things we wanted to do but couldn't because the technology will not allow us, but with the next gen of hardware u'll be amazed"..

I mean, OF COURSE we'll be amazed! That's why they MAKE new generations of hardware...

Anyway, it will be lovely to see what kind of pre-release demos EA will show us.... :?
 
hey69 said:
"Since we are getting a head start on that [creating a revolutionary next-gen product], I can't reveal too much but one of the things we think about is, it's not going to be photo-realism, right? The next-generation systems can throw 30,000 polygons into a car and guess what, it doesn't look any better than a 10,000 polygon car with the rendering techniques I can do.

30 000 Polygons per cars?? That would mean the game will be under 50/70M polys/sec (if we assume there's 8/10 cars on screen, no LOD, and no multipass techniques for the cars that force to resend the geometry, tracks with complex geometry (and a lot of traffic) and of course 60fps).

That's really underwhelming, IMHO.
 
Well, 8-10 cars is too much of an assumption, I don't believe that number you threw out, shoot just look at the unreal engine 3 models, many were under 10k, naturally where the geometry is really being thrown around is in environments, and if this is supposed to be anything at all like GTA, I imagine they'd have some massive environments with the quality we'll come to expect out of next-gen products...so the final number may be far highers.
 
GwymWeepa said:
Well, 8-10 cars is too much of an assumption, I don't believe that number you threw out, shoot just look at the unreal engine 3 models, many were under 10k, naturally where the geometry is really being thrown around is in environments, and if this is supposed to be anything at all like GTA, I imagine they'd have some massive environments with the quality we'll come to expect out of next-gen products...so the final number may be far highers.

Actually, Chuck Osieja is insinuating that the cars could be made of 10 000 polys since he says we can't tell the difference between a 10k and a 30k model when its well "rendered"(shaders), in which case that would leave a lot poly for the environement, especially with a good LOD...

We heard that next-gen consoles could push something like Billion polygons, therefore listening someone telling that he's developing a racer game with car made of 10k/30k polys, is somewhat puzzling, but that's just me... :D
 
Again, that unreal engine 3 is using less than 10 for many models, if this game is anything like GTA, imagine crowds of people made up of a few thousand polys, buildings, birds, stray dogs, etc, what have you...its too early to speculate on how many polys they're really pushing...shoot what if the game had as many cars as F-ZeroX, upto 30 on screen...what if they incorporate traffic..etc I can throw whatever at ya...or they could be doing far less, who knows lol.
 
I think it's long overdue that racing games move beyond 8 cars in a dead lifeless track environment.

I want to see hundreds of cars and thousands of spectators.
 
aaaaa00 said:
I think it's long overdue that racing games move beyond 8 cars in a dead lifeless track environment.

I want to see hundreds of cars and thousands of spectators.

Darn tootin. If its a street racing style racer, I want spectators to be eating hotdogs, drinking sodas, people messing with each other, some taking photos, videos, some real life.
 
You won't get that from EA, especially not with any NFS: Underground game. I really do hate the direction they have gone with the series and hope they bring it back to its roots in terms of extreme cars with arcade style racing. Not riced or pimped out cars that anyone would destroy in real life.
 
So basically, they didn't reveal shit.

This is the same website who a week ago had that article claiming "xbox 2 specs locked down!"

I'm sick of this misleading garbage.

But it's noone here's fault, not even you hey69; don't get the wrong idea.
 
I think the interview shows two things really. First, software development mirrors hardware development. As the latter begins to spiral in complexity (becoming more expensive and more time-consuming in the process), so does the former. ;)

Secondly, there's a point of diminished returns ... and next generation hardware may be approaching it. :oops: As the number of polygons continues to grow, it's taking a larger and larger geometric increase in order to see the contrast. The difference between, say, one polygon and 10 is obvious, but 100 billion polygons and 500 million may not be...
 
My 2c

I think that next gen titles will look dramatically better.

However from my standpoint the exciting part is what they are going to let us do from a CPU standpoint. Consoles have been woefully under powered in the CPU department, what we think we know so far seems to indicate this may be changing.

I personally think there are a lot of interesting gameplay avenues that can be explored, that have been unavailable to this point.

Give me 1000x the power or a 1000000x the Power and I'll find a way to use it. I really think we have an awfull long way to go before we reach the point of significant deminishing returns.
 
ERP said:
Give me 1000x the power or a 1000000x the Power and I'll find a way to use it.

But will it be cost-effective? ;) A geometric increase in the polygon count, for instance, is liable to mean an exponential increase in the workload.
 
Pepto-Bismol said:
ERP said:
Give me 1000x the power or a 1000000x the Power and I'll find a way to use it.

But will it be cost-effective? ;) A geometric increase in the polygon count, for instance, is liable to mean an exponential increase in the workload.

Not necessarily, in the end artists work with models made of sometimes million of polygons, then reduce the polygon count until they're manageable to work with in real time. So the work is always already there. It's just a matter of decreasing the polygon counts less than we do today. ;)
Not sure if that is always the case though.
I do still think that the main problem will remain, and for a long time too: RAM.
 
london-boy said:
Not necessarily, in the end artists work with models made of sometimes million of polygons, then reduce the polygon count until they're manageable to work with in real time. So the work is always already there. It's just a matter of decreasing the polygon counts less than we do today.

Eventually, the developer is going to have to work with what the game will be manipulating. :| And gobs of polygons won't be just the artists' problem -- I think the brunt of it will be borne by an army of technologists (engineers, mathematicians, physicists), who are trying to squeeze elaborate models into a more general physical universe.
 
Pepto-Bismol said:
Eventually, the developer is going to have to work with what the game will be manipulating. :| And gobs of polygons won't be just the artists' problem -- I think the brunt of it will be borne by an army of technologists (engineers, mathematicians, physicists), who are trying to squeeze elaborate models into a more general physical universe.


Well i'm not really sure how video game studios work, only have experience with how CGI studios work, but i guess developers and coders only need to worry about how the hardware will deal with what the artists will throw at them.
The "exponential" increase in work you were talking about is more likely to affect the artists.
The increase stems from the fact that the artists themselves will have to create more and more highly detailed environments, and the "issue" will be that next gen(s) there will be less to "hide" therefore they will have more to show and more to create.
Today they can "get away" with some things for the fact that today's hardware can only display so much, come next gen(s) they won't be able to hide behind hardware capabilities. Do you know what i mean?
 
london-boy said:
The "exponential" increase in work you were talking about is more likely to affect the artists.

All I was saying was that software development mirrors hardware development, particularly as it becomes increasingly complicated and interconnected. So a change in one area is liable to have ripple effects in others (and vice versa) -- and THIS, I am theorizing, is what will increase a developer's workload exponentially.

Sure, if artists have a higher polygonal limit, they're going to have their work cut out for them -- I'm not arguing with you there -- but what about the blokes who toil behind the scenes (audio, software tools, game physics, AI, etc.), trying to fit more detailed models into larger, more interactive worlds ... convincingly? Aren't they deserving of sympathy too? :?

london-boy said:
Today they can "get away" with some things for the fact that today's hardware can only display so much, come next gen(s) they won't be able to hide behind hardware capabilities. Do you know what i mean?

Well, I -- I haven't the foggiest idea what you mean actually. Sorry. :oops:
 
I haven't done very much complex modelling work myself, but I've always thought that it must actually be quite time-consumung to make a low poly object look like it has much more detail than it actually does.

For example a car in GT2. I'm sure the artists had to think hard where to save the polygons, but still make the car make and model recognizable.

With higher polygon budget, they can just take the car manufacturer's CAD modelling data (if they allow them to use them, I don't know), and use it with much less modification that was needed for PSOne game.
There'll be less work for texture artists to make textures that 'fake' the details look like they are actual geometry.

I'd think that high-res modelling would still be easier than finding short-cuts and trying to 'patch' the missing geometry with texturing.

I'd be happy if someone with more modelling and texturing experience would comment (those who actually have done modelling + texturing in console games).
 
rabidrabbit said:
I'd be happy if someone with more modelling and texturing experience would comment (those who actually have done modelling + texturing in console games).

Might as well put a padlock on this thread...I mean who dares speak on topic now?! :oops:




P.S. I think the next generation experts that you are soliciting are probably bound, head-to-toe, by non-disclosure agreements! ;)
 
Back
Top