Xbox 2 == Gc2

[PaB]

It cleary doesn't take decades to introduce a new console and get people to switch. If the console is miles better than the last then they will switch.

Laserdisk failed with the public because 'It didn't record'. At the time LD was introduced buying\renting films was not the big feature of VHS. Recording off the telly was the big deal.

DVD hasn't replaced the recording function of VHS it has replaced the moving viewing\collecting\renting function which took quite a few years to take off.

If you are as old as me you'd rember a time when 90% of the public thought the very idea of owning a film on tape was daft. "But I've seen it, why would I want ot watch it again?". It took years for people to get the idea of video software.

I am almost as old as you - almost but not quite .

Yes I can see where you're coming from, and I guess a bit of paranoia is creeping in with my obsession for forward movement. As a Hardware Engineer it is built into my psyche

[Luke]

It cleary doesn't take decades to introduce a new console and get people to switch.  If the console is miles better than the last then they will switch.

Especially if it is fully backwards compatible with their previous console (no matter who had made the old or new console), and if they haven't had to shell out on a new console for a reasonable number of years.

As it is, the next gen of consoles are going to have to offer something pretty impressive if I'm going to buy one - and I doubt they will.

[Kyoshiro]

For a handful of scenes, from very specfic camera angles. It's a movie, not a game. Do you understand the difference?

Oh of course, I do yeah. But what I mean is, the modelling took under a year - the animation etc. The framework was all set after that, they could do what they wanted with it then, 'act' with it (so to speak). The only thing to do after that was the rendering.

And assuming there was the gargantuan amount of processing power in a home console that could render Gollum in real time, what about the 'acting', i.e. a number of animations that would make him a believable character in the context of a game, not simply set, cinematic cut scenes. PoP used, what, 750 for the main character? The mind boggles.

I believe animation is in desperate need of new, more dynamic techniques and is the next key area of evolution just like graphics then physics now.

How can you get more lifelike than motion capture?

The problem with current animation techniques is they aren't dynamic enough. It all looks good in cut scenes and trailers but when you actually play the game you're interacting and doing things which the animation has to compensate for dynamically or it just doesn't look right. And also no matter how good the animation is when three soldiers come running in with exactly the same animation you can't help but think how primitive current techniques are.

I remember reading a news article in EDGE about this animation package that gives your model a brain so he has an innate sense of balance. So the animator can do traditional animation for falling etc.. but when the model would get hit or a force applied on it it wouldn't just flop like a ragdoll but because of it's brain would react with arms legs and body position accordingly to try and keep its balance.

There also needs to be more incidental animation. Anyone who's seen Studio Ghibli's work can appreciate the huge number of little animations that make characters more real, but most developers in videogames just don't pay attention to little details like that.

Hardcoding all these animation is also way too much work and too artificial. There needs to be some sort of technique where the animator only sets animation guidelines which the program dynamically chooses and modifies accordingly. For example the person can have a running animation. But when the model really is running maybe things like footstep placing can change depending on the rocks in the gameworld. Maybe the program can combine the duck and run animations to duck while running under a branch. Maybe when the model is hit by the branch while attempting to duck instead of flopping like a ragdoll the program changes the positioning of the leg and arms to try and balance out the fall.

So many things which can be done, but the big thing now is physics and graphics, but animation is what makes a world feel real, not graphics.

[PaB]

As it is, the next gen of consoles are going to have to offer something pretty impressive if I'm going to buy one - and I doubt they will.

New games?

[Luke]

New games?

Yes, that's probably the one thing (although I'm already building up quite a back catalogue of un-finished/started games).

However, new games don't need a new console - that's just a factor that's forced upon us (that this slowly updated standard would help tackle).

[Apollo]

I think games will never reach the heights of movie graphics unless companies start outsourcing to WETA or ILM etc. It can take a couple of days to properly model a character so when games get on a bigger scope and we stop having things like repeating textures, things are gonna take longer.

Computer's themselves will design the graphics/animations in games eventually using artifical intelligence. There has been a lot of academic research into over the past 20 years.

[Jonathan_Kerr]

That's pretty pre-set though. I mean, what if you wanted to interact with that and, say, STOP them falling halfway... with a big stick. Where's the animation for them now falling onto the stick? Or what about onto something else? You can only compensate to a certain extent, unless you have a motion capture for every possible animation any character could ever possibly do... which just isn't (humanly) possible.

Well Ragdoll is a step in that direction if you're talking about interacting with the environment.

And yes - it's impossible to have animations for everything. More because how would you be able to do *everything* you wanted in a game, let alone have an animation for it.

ie - I want to try and eat the plasma rifle instead of shoot with it. Designers just couldnt' predict the players wants.

[Jonathan_Kerr]

How can you get more lifelike than motion capture?

The problem with current animation techniques is they aren't dynamic enough. It all looks good in cut scenes and trailers but when you actually play the game you're interacting and doing things which the animation has to compensate for dynamically or it just doesn't look right. And also no matter how good the animation is when three soldiers come running in with exactly the same animation you can't help but think how primitive current techniques are.

I remember reading a news article in EDGE about this animation package that gives your model a brain so he has an innate sense of balance. So the animator can do traditional animation for falling etc.. but when the model would get hit or a force applied on it it wouldn't just flop like a ragdoll but because of it's brain would react with arms legs and body position accordingly to try and keep its balance.

There also needs to be more incidental animation. Anyone who's seen Studio Ghibli's work can appreciate the huge number of little animations that make characters more real, but most developers in videogames just don't pay attention to little details like that.

Hardcoding all these animation is also way too much work and too artificial. There needs to be some sort of technique where the animator only sets animation guidelines which the program dynamically chooses and modifies accordingly. For example the person can have a running animation. But when the model really is running maybe things like footstep placing can change depending on the rocks in the gameworld. Maybe the program can combine the duck and run animations to duck while running under a branch. Maybe when the model is hit by the branch while attempting to duck instead of flopping like a ragdoll the program changes the positioning of the leg and arms to try and balance out the fall.

So many things which can be done, but the big thing now is physics and graphics, but animation is what makes a world feel real, not graphics.

Hmm - it seems that MASSIVE! should be adapted to games. But there's a certain sense of fairness by having enemies who share similar traits and speeds.

Imagine a vertical shoot em up where every enemy had it's own brain - it'd be fuckin' impossible to win! If enemies start having variations in traits too much, games will start getting too difficult to beat. I'm no programmer, but I'd hate to see what sort of impact on CPU that heaps of AI calculations could have.

Mind you, I think some sort of randomiser or noise to animations wouldn't hurt for the variation you speak of.

[Eoin]

TV isn't a fast moving target. The HDTV standard is only 720p (max res.)

Well......there is a proposed 1250p EU HDTV standard, but when (and whether) it ever gets implemented is another question.

Your idea sounds just like the 3DO. And look what happened there.

A significant part of the idea, though, is that we aren't many generations away from "good enough". The 3D0 was quite a bit further away - and was almost immediately followed by more powerful machines.

I cannot beleive people are advocating "less choice" as being a good thing.

If you're just counting consoles, then yes, there'd be less choice - there'd be one less console (barring the entry of another company, and there aren't many who have both the resources and the motivation). However, if there are only two consoles, isn't there a good chance that the competition would actually be greater? Is is not likely that the combined MS/Nintendo machine could force Sony into cutting the price of the PS3 more than they have with the PS2? As it stands, the majority of people this generation own a single games console (likely to be the PS2) - would they not have greater choice if they could own two?

Even if they only chose the hypothetical MS/Nintendo machine, they'd still have a greater choice of games than they would have had if both Microsoft and Nintendo had a machine on the market.

So, I wouldn't see it as being less choice - it could well end up being more.

(BTW, who really complained about lack of choice when it was the SNES vs the Mega Drive, or the PSone vs the N64? Two consoles has been enough for most people in the past).

Mr Cynical USA Gamer: "What? Kiddie Nintendo games on the X-Box2? Sod that.."

Mr Cynical Japanese Gamer: "What? Boring shooting and driving games on the GC2? Sod that.."

MSFT and Nintendo would risk alienating what little markets they already have.

I think you're extropolating the very narrow sample of people posting on games forums onto the public at large. Most people don't really care if there are games on their console that they don't like - just that there are a few games that they do. There are plenty of "kiddie" games on PS2, just like there are plenty of driving and shooting games on PS2. Neither seems to have harmed the sales.

As for "minnows" I would hardly call 15 million worldwide sales for each a "minnow"

The last figures I've seen were roughly 10 million, give or take, for both Xbox and GameCube. They're a bit old now, but I don't expect them to have suddenly sold 5 million units each in the last few months. Even if it's right, 15 million still isn't a high number by the standards of games industry successes. The Mega Drive/Genesis sold 23 million, the SNES 49 million, the N64 around 30 million, the GBA/SP is hovering around 40 million......and the PS2 is outselling GC and Xbox combined by at least a factor of 2. That's without mentioning the huge sellers like the GameBoy (range), the PSone and the NES.

By the standards of failures like the 3D0, CDi, Jaguar, and so on, the GC and the Xbox are success stories. By the standards of most of the machines that people remember, I'd say that if they're not minnows, they're close to it.

Am I playing "Finding Nemo" in real time? Not by a damn sight.

How far are you away from it though? Really? The PSone struggled with 100,000 polys a second, while the PS2 throws around 10 million of them. The N64 made do with 50 or 60 thousand, while the GC's speclist gives an estimate of 6 million, and realitically the figure is more than twice that. That's a jump of two orders of magnitude in a single generation - how many orders of magnitude are we away from Nemo? How many polys per second will we need? 100 million? Next generation's consoles will do that, probably without much effort. A billion? Smalltalk for the generation after that. If we have that amount of power, then a screen with more than 4 times the resolution of today's best TVs, displaying 60 frames per second, with an overdraw factor of 3, can be completely filled with polys with room to spare....

Sorry, that was probably boring. Is that good enough for real-time Nemo yet? It's certainly good enough that Granny isn't going to be able to tell the difference between Finding Nemo 3 on the PS4 and Finding Nemo 4 on the PS5.

I want to recreate the battle of middle earth in real time and zoom into the body of one of the orc soldiers.

Can I do it yet? No?

Not good enough then.

I don't know. You tell me.

That T-Rex Dinosaur demo thing on the PSone looked convincing to me. Photo-real perhaps? Well I was fooled anyway.

Did I see that same level of detail in any games? Did I hell!

Do you get that kind of detail now?

I'd say yes, you do.

You didn't in PSone games because it was using a significant amount of its power on that single thing, with nothing else. No AI routines, nothing approaching complex animation (just walking), not even a background. Ask one of the current consoles to do something like it, and it wouldn't bat an eyelid.

[Kyoshiro]

How can you get more lifelike than motion capture?

The problem with current animation techniques is they aren't dynamic enough. It all looks good in cut scenes and trailers but when you actually play the game you're interacting and doing things which the animation has to compensate for dynamically or it just doesn't look right. And also no matter how good the animation is when three soldiers come running in with exactly the same animation you can't help but think how primitive current techniques are.

I remember reading a news article in EDGE about this animation package that gives your model a brain so he has an innate sense of balance. So the animator can do traditional animation for falling etc.. but when the model would get hit or a force applied on it it wouldn't just flop like a ragdoll but because of it's brain would react with arms legs and body position accordingly to try and keep its balance.

There also needs to be more incidental animation. Anyone who's seen Studio Ghibli's work can appreciate the huge number of little animations that make characters more real, but most developers in videogames just don't pay attention to little details like that.

Hardcoding all these animation is also way too much work and too artificial. There needs to be some sort of technique where the animator only sets animation guidelines which the program dynamically chooses and modifies accordingly. For example the person can have a running animation. But when the model really is running maybe things like footstep placing can change depending on the rocks in the gameworld. Maybe the program can combine the duck and run animations to duck while running under a branch. Maybe when the model is hit by the branch while attempting to duck instead of flopping like a ragdoll the program changes the positioning of the leg and arms to try and balance out the fall.

So many things which can be done, but the big thing now is physics and graphics, but animation is what makes a world feel real, not graphics.

Hmm - it seems that MASSIVE! should be adapted to games. But there's a certain sense of fairness by having enemies who share similar traits and speeds.

Imagine a vertical shoot em up where every enemy had it's own brain - it'd be fuckin' impossible to win! If enemies start having variations in traits too much, games will start getting too difficult to beat. I'm no programmer, but I'd hate to see what sort of impact on CPU that heaps of AI calculations could have.

Mind you, I think some sort of randomiser or noise to animations wouldn't hurt for the variation you speak of.

i'm not talking about anything to do with play mechanics it's all just animation like character movement etc... not ship movements and so on. Because soldiersi n medal of honour will scratch their arses differently shouldn't make one harder

[Sprite Machine]

I want to recreate the battle of middle earth in real time and zoom into the body of one of the orc soldiers.

Can I do it yet? No?

Not good enough then.

I don't know. You tell me.

That doesn't look anything like how it looked in the movie.

That T-Rex Dinosaur demo thing on the PSone looked convincing to me. Photo-real perhaps? Well I was fooled anyway.

Did I see that same level of detail in any games? Did I hell!

Do you get that kind of detail now?

I'd say yes, you do.

You didn't in PSone games because it was using a significant amount of its power on that single thing, with nothing else. No AI routines, nothing approaching complex animation (just walking), not even a background. Ask one of the current consoles to do something like it, and it wouldn't bat an eyelid.

Oh yeah, I know you do now. It's easy peasy now, just like a realistic MiddleEarth battle would look exactly like the movie in say the next gen, or the next two?

Anyhoo....

[Kyoshiro]

How far are you away from it though? Really? The PSone struggled with 100,000 polys a second, while the PS2 throws around 10 million of them. The N64 made do with 50 or 60 thousand, while the GC's speclist gives an estimate of 6 million, and realitically the figure is more than twice that. That's a jump of two orders of magnitude in a single generation - how many orders of magnitude are we away from Nemo? How many polys per second will we need? 100 million? Next generation's consoles will do that, probably without much effort. A billion? Smalltalk for the generation after that. If we have that amount of power, then a screen with more than 4 times the resolution of today's best TVs, displaying 60 frames per second, with an overdraw factor of 3, can be completely filled with polys with room to spare....

Sorry, that was probably boring. Is that good enough for real-time Nemo yet? It's certainly good enough that Granny isn't going to be able to tell the difference between Finding Nemo 3 on the PS4 and Finding Nemo 4 on the PS5.

I want to recreate the battle of middle earth in real time and zoom into the body of one of the orc soldiers.

Can I do it yet? No?

Not good enough then.

I don't know. You tell me.

That T-Rex Dinosaur demo thing on the PSone looked convincing to me. Photo-real perhaps? Well I was fooled anyway.

Did I see that same level of detail in any games? Did I hell!

Do you get that kind of detail now?

I'd say yes, you do.

you can always use more processing power to increase not only graphics, but AI and physics. It's not simply about poly count, there are plenty of other processor intensive things to do like lighting, particle systems, destructive environments etc... which have to ensure the high poly environment isn't just a pretty lifeless static world. Finding Nemo doesn't let a player control a fish and swim around and hit other fishes... The lighting algorithm will take very long to compute and even when that can be done in real time without any noticeable improvements it still has to do it simultaneous with the near-perfect particle system. Running alongside the AI which will never have enough processing power until it becomes self-aware or whatever. And even when we have all that it still has to be able to apply to every genre from 2 models in a beat em up to 10000 in a total war game.

We will reach that stage where there isn't any need to upgrade anymore for 25 years and we have some kind of standard but that's not within 2 generations.

[Eoin]

you can always use more processing power to increase not only graphics, but AI and physics. It's not simply about poly count, there are plenty of other processor intensive things to do like lighting, particle systems, destructive environments etc... which have to ensure the high poly environment isn't just a pretty lifeless static world. Finding Nemo doesn't let a player control a fish and swim around and hit other fishes... The lighting algorithm will take very long to compute and even when that can be done in real time without any noticeable improvements it still has to do it simultaneous with the near-perfect particle system. Running alongside the AI which will never have enough processing power until it becomes self-aware or whatever. And even when we have all that it still has to be able to apply to every genre from 2 models in a beat em up to 10000 in a total war game.

Of course it's not just about poly count, but that's hardly the only thing that's advanced over leaps and bounds in every generation. If we're within a few generations of a console that can throw a billion polys a second (or, More than We will Ever Need) around, then I don't think it's in any way unrealistic to suggest that it will also be powerful enough to light source them and deform them according to events in the gameworld. We're not stuck with - many - lifeless static worlds now - why would we be in two generations?

And you're right - enough processing power to do "real" AI isn't happening in the foreseeable future. But why is that a problem? Trick the player.

Make characters in beat-em-ups select moves from a weighted list, and dynamically alter the weights as they succeed or fail. Alter the weights to be biased towards whatever moves the player is using too. That can be done with maybe a couple of hundred lines of code and a text file, and the illusion, while not perfect, would fool most people in a Tekken Turing Test.

10,000 models in a war game? Flock them into groups and use a couple of sergeant AIs to command them. When you're dealing with huge amounts of units, then it's not like the player is going to worry - or even notice - if each individual unit only has the intelligence of a spider.

[Sprite Machine]

Of course it's not just about poly count, but that's hardly the only thing that's advanced over leaps and bounds in every generation. If we're within a few generations of a console that can throw a billion polys a second (or, More than We will Ever Need) around, then I don't think it's in any way unrealistic to suggest that it will also be powerful enough to light source them and deform them according to events in the gameworld. We're not stuck with - many - lifeless static worlds now - why would we be in two generations?

And you're right - enough processing power to do "real" AI isn't happening in the foreseeable future. But why is that a problem? Trick the player.

Make characters in beat-em-ups select moves from a weighted list, and dynamically alter the weights as they succeed or fail. Alter the weights to be biased towards whatever moves the player is using too. That can be done with maybe a couple of hundred lines of code and a text file, and the illusion, while not perfect, would fool most people in a Tekken Turing Test.

10,000 models in a war game? Flock them into groups and use a couple of sergeant AIs to command them. When you're dealing with huge amounts of units, then it's not like the player is going to worry - or even notice - if each individual unit only has the intelligence of a spider.

But these are still 'workarounds', and whilst I find what can be done with that sort of thing clever, you still limit yourself by doing that.

What if the design of your game required individual thinking troops and things, in ways that you couldn't work around?

The T-Rex PSone demo thing I mentioned was a decent example. It was impossible to have that running in-game on the PSX - new hardware wass essential, and now, it can be done. I just think with everything always improving mathematically (AI, physics, the size of this that and the other) you're always going to have 'something else' you can achieve. I just can't envisage a time when that isn't the case.

When the next gen arrives though, we'll see.

[footle]

And as mentioned a million times on this thread already:

You can have as much processing power as you want - but there's already a problem with creating the appropriate resources to make use of it in a commerical manner.

[Kyoshiro]

actually game worlds nowadays are still incredibly static. I think it's one of these things we don't realise unti after it's achieved. It's still just a map in which you place a player and a few objects with only the occasional terrain deforming to try and trick you, the world as a whole just does not "breathe". They make look photorealistic but it still makes it all the more unnatural that a player will run into a wall and just stop, magically turn on the spot without moving its feets and run again into a chair and stop. All in squeaky clean dynamic lighting and shadows.

In fact lots of people are just thinking along a straightforward evolutionary process of better textures, better lighting, better physics, better AI, better polygon count and according to that yes we are reaching a peak but that is still incredibly primitive when you think about other areas that are only now becoming a consideration as a side effect of those benefits. These areas include techniques that need to be improved/discovered to make the world just feel real, like the animation stuff i won't repeat here.

edit: And not necessarily "real" as in realistic, i mean convincing, like Studio Ghibli's animes or Finding Nemo. The problem is videogames are interactive.

[CurryKitten]

I like JP's initial idea on this, it would be wonderful to see, but I don't think it'll happen.... not yet anyhow

Usually I'm dead set against a standard single licensed console design. Creativity / pushing the boundaries / blah blah blah. Eventually this has to be inevitable though.

Lets say we are at the stage several generations down the road, and the 3 console makers are still going. What if these new uber-powerful beasts could do anything you can pretty much imagine. Let's say they can do finding Nemo in realtime, they have super-duper new AI programmed, basically you can throw as much as you like at these things and can't slow them down. For the benefit of Sprite Machine we can even say that everything's in RAM and there's no streaming

If a game is released on all 3 platforms then it would look and play absolutely identical, but for exclusive releases, some people are left not being able to play the game. Could you really see people buying multiple consoles that were to all intense purposes identical ? At this point a single console becomes the only way to go, else it would be the equivalent now of having DVD players that only played movies from certain film studios. I think it's just a matter of when this will be, not if.

[Kyoshiro]

I like JP's initial idea on this, it would be wonderful to see, but I don't think it'll happen.... not yet anyhow

Usually I'm dead set against a standard single licensed console design. Creativity / pushing the boundaries / blah blah blah. Eventually this has to be inevitable though.

Lets say we are at the stage several generations down the road, and the 3 console makers are still going. What if these new uber-powerful beasts could do anything you can pretty much imagine. Let's say they can do finding Nemo in realtime, they have super-duper new AI programmed, basically you can throw as much as you like at these things and can't slow them down. For the benefit of Sprite Machine we can even say that everything's in RAM and there's no streaming

If a game is released on all 3 platforms then it would look and play absolutely identical, but for exclusive releases, some people are left not being able to play the game. Could you really see people buying multiple consoles that were to all intense purposes identical ? At this point a single console becomes the only way to go, else it would be the equivalent now of having DVD players that only played movies from certain film studios. I think it's just a matter of when this will be, not if.

i think the matter of when is obvious not the if but it's more like soon or not for a while?

i think not for 4 generations at least. When I will be making the decisions in my own game

[Nick Laslett]

I think a lot of you are over estimating how hard the current consoles are being utilised.

SCEE has recently finished a 2 year study that used the Performance Analyzer to review how much of the potential of the PS2 each title was using.

You can see the results at the following site:

http://www.technology.scee.net/sceesite/fi...arHaveWeGot.pdf

The study included games up to: JakII, ZOE2, Return of the King and Silent Hill 2.

Rendering Analysis:

Average: 52,000 polys per frame

Maximum: 145,000 polys per frame

That equals 8.7 million polys per sec with a title running at 30fps

The best performance was on a title that pushed 7.5 million polys per sec running at 60fps

(These numbers are averages for a game not peaks. The PA counts the polygons rendered on screen, not how many were sent.)

Framerate: 60% were running at 25/30 or less

95% were using full height buffers. (This means most PS2 games could run in prog scan.)

VU0 Usage:

Average: 2%

Best performing games: 8%

VU1 Usage:

Average: 56%

The PS2 theoretical maximum performance with all visual effects turned on is 22m polys per second. There is a publically available PS2 demo that pushes 36m polys per second at 60fps.

So the PS2 is still only being 50% utilized at best. The PS2 can push a lot more polys than it is currently doing and the systems main bottleneck is the CPU for AI and physics calculations.

I think that it is safe to assume that on average the Xbox and GC are also being heavily under utilized. I would expect that their games are only hitting 10m polys per sec averages. But these consoles have much better texture features and therefore the games look better.

The SCEE engineer that gave the presentation reckoned that the average game across all consoles only achieved 5m polys per sec.

If Developers were given more time to R&D with these systems, I would expect we would be seeing much better games in all senses not just graphically.

It is a shame to think that in less than 2 years time these consoles will be obsolete, when they still have so much more potential.

[Eoin]

What if the design of your game required individual thinking troops and things, in ways that you couldn't work around?

It's a good question, but I'm not sure that there are ever going to be many games that call for large amounts of units thinking individually. Typically what you'd want is lots of units reacting the same way - "Charge" or "hold this piece of ground" or "We're outnumbered, let's get the hell out of here", and if that's decided by a flocking routine then it'll look almost (and possibly exactly) the same as each unit deciding that individually. If there's a small amount of units that do need to think individually - let's call them Aragorn units to stay with the loose LotR concept - then that can be handled too. The Aragorns can have individual AIs tailored to individual thinking and decision-making, while the rest of the battle rages around them, controlled by flocking routines.

I'm not sure you'd ever really need ten thousand Aragorn units in a game.

I just think with everything always improving mathematically (AI, physics, the size of this that and the other) you're always going to have 'something else' you can achieve. I just can't envisage a time when that isn't the case.

Of course - something more will always be possible. You might want to make a dino comprised of a million polys, or a billion, or realistically model its entire biological system that the player will never see.

However, doesn't it eventually get to the stage where any improvement won't be noticed by 99% of people? Or, equally related to the thread and possibly a more pertinant question - when are consoles going to be limited by the resolution of the screen they're displaying on? You'll always be able to achieve something else, but are (most) people always going to be able to notice that you've done it.

actually game worlds nowadays are still incredibly static.

Admittedly there's still a long way to go in terms of how much of the environment can be interacted with (if we actually want to go down that route), but I still wouldn't describe most of them as static, and certainly not lifeless.

The thing is, we're still looking at gameworlds like people who've played a big pile of games. We know they're maps with textures plastered over them with scripted setpiece moments triggered by where you stand, and sometimes it's difficult to look past that unless the game does something we haven't seen before or something that tricks us. For us, Monaco in GT3 is a grey-textured road with some boxes painted as buildings to the side and a sky that's really a bitmap postcard. Most people (and I kind of envy them) don't think like that - they see it as Monaco, like the one on the telly once a year, and now they can drive cars around it. From that point of view, it's no more static than it should be, and I think that applies almost right across the scale.

[Cheeko]

It is a shame to think that in less than 2 years time these consoles will be obsolete, when they still have so much more potential.

Yes, but the next consoles will have even more potential, out with the old I say.

The quest for photorealism and the discussion here can't be decided either way. If your game involved a blue box in a plain white room, then on modern consoles yes, you could make a game on todays consoles that would look photoreal.

As with a point that Jpickford made earlier, I've been playing a racing game before and turned it off to watch the F1 grand prix on a sunday and my mum will walk in and say 'is this still the game?' but then you can afford to throw more poly's and effects into a driving game, PGR2 looks spookily real in some places.

So what is it you're after as your pinnacle of technical achievement, reality simulators? Don't think that comes from the graphics, look at The Sims.

[DJ Sack]

Nothing will ever be photorealistic, mainly due to our crap TV systems, but also everything is digital in a gaming world, such as pysics, in the real world its analogue.

[___]

Nothing will ever be photorealistic, mainly due to our crap TV systems, but also everything is digital in a gaming world, such as pysics, in the real world its analogue.

The mind boggles...

[Luke]

As with a point that Jpickford made earlier, I've been playing a racing game before and turned it off to watch the F1 grand prix on a sunday and my mum will walk in and say 'is this still the game?'

A couple of my friends were playing a football games once (must have been PS1).

Another friend walked in and sat down to watch thinking it was a real game. After a while he said "eh? what's going on there ... oh, you're playing a game".

[Daley]

A couple of my friends were playing a football games once (must have been PS1).

Another friend walked in and sat down to watch thinking it was a real game. After a while he said "eh? what's going on there ... oh, you're playing a game".

My dad does that all the time. Only when Maradona went round everyone and squared it to Pele to tap in did he realise I was playing PES3.

[JPickford (retired mod)]

Nothing will ever be photorealistic, mainly due to our crap TV systems, but also everything is digital in a gaming world, such as pysics, in the real world its analogue.

That means nothing DJ

If it looks like a photo it's photo realistic.

If moves like real life. it's realistic.

My whole point is that we'll be 'good enough' soon. Not perfect.

[acidforblood]

Instead of MS teaming up with the Number Two in the world why not go straight to source, and team up with Sony?

[kensei]

You have sucked into the "Bigger, better, more" mentality that is an absolute cancer on gaming, This game is better because it lasts 150 hours. This game has 40,000 levels, this game had 97 character and 48 billion unlockable cars. This game has real time shadows and super-duper refelctions, so it must be the most superior experince around. All of these things are of themselves, completely irrelevant. It's how they effect the game that matters.

You are completely missing the point.

No I'm not. The ideas you are arguing for are part of the same mentality.

Essentially what you and Sprite Machine are arguing for is this: a machine that produces the EXACT same game, with the difference that it can zoom into the skin of orcs where the previous gen couldn't. To all intents and purposes, and to 99% of the population, there is no discernable differnce. And you are asking people to shell out more money for it. Why? Some stupid "bigger, better, more" USP.

Let's take Sprite's ideas and apply it to something else: bump-mapping. Bump Mapping is cheating! You should use the extra polys and lighting calculations to do it properly! Of course, in the end it will look identical, but you ahve the satisfaction of knowing that it's all being done properly.

[JPickford (retired mod)]

Instead of MS teaming up with the Number Two in the world why not go straight to source, and team up with Sony?

1) There's nothing in it for Sony.

2) That's missing the point of why they will team up with Nintendo.

[squirtle]

That dinosaur thingie on the PS1... Dino Crisis 1 and 2 did better than that, and they were in game with me shooting them. Pah!