The Future Of Graphics

[AGD]

Do they then become interactive movies rather than games?

Key components of games are goals, rules, challenge, and interaction. Games generally involve mental or physical stimulation, and often both. Many games help develop practical skills, serve as a form of exercise, or otherwise perform an educational, simulational or psychological role.

Mentions nothing of visual experience.

You're missing the point. People will always use the word "game" to refer to all forms of interactive entertainment. Words can assume new meanings over time due to their usage. Our kids will think that the word game means interactive experience, not something where you get points or have rules and goals.

Graphics will certainly continue to improve towards photorealism but this will put huge pressure on art departments in games companies. Already, this is the highest and fastest increasing cost in the development process and we're gonna need a lot more intelligent procedural techniques to be able to continue creating large scale environments. Don't be surprised to see more companies whose sole job is to produce art assets which can be used in many games. Middleware companies will get a LOT bigger.

 

[FlukeRogi]

If you want to see the future of graphics then take a look at the 3dmark 11 demo video.

Yep, very impressive. The subs looked a bit too rendery(?!) at points, but the scenery - I'll agree. Extremely convincing stuff.

avatar cheated and the character animation is of real people not computer generated, so we still lack the computing power / technology / software to properly animate the real world..

You mean like they've used in games (and movies) for years? Yes, there are alternatives like Euphoria, but motion capture for character animation is pretty standard, or do you consider that games also "cheat"?

When watching something like avatar, would you rather that the rendered characters were animated via software trying to mimick how someone should move, or would you rather the actions were captured from an actual actor for total realism?

Looking ahead for games, characters being able to dynamically react probably is a good thing, but for film (for humanoid characters at least, and where feasible), surly motion capture is more realistic.

 

[craptakular]

What you've got to remember is there are various ways, even now, of rendering real-time 3D scenes. We've all enjoyed the massively adopted polygons wrapped in bitmaps approach, but we're now seeing the long term limitations of churning out scenes based on this approach. We've stuck with it too, promoting features that alleviate some of the growing pains, think bump mapping etc, to get more detail for little extra work from the hardware.

I'm not massively up to date on the scene, but I would be surprised that as our current approach plateaus we don't see different approaches rear their heads. Then again, we now have our new medicine, tessellation, to help us get through.. and the entire industry is built on one approach, so any move is going to be very very very very slow.

25 years is a long time, but I don't see real-time photorealistic 3d environments happening without a dramatic change in the technology (power/cooling/space) or approach. A breakthrough could happen tomorrow, or never..

Isn't point to point rather than polygon the waqy forward? Someone posted a video about it a few weeks back.

 

[AGD]

What you've got to remember is there are various ways, even now, of rendering real-time 3D scenes. We've all enjoyed the massively adopted polygons wrapped in bitmaps approach, but we're now seeing the long term limitations of churning out scenes based on this approach. We've stuck with it too, promoting features that alleviate some of the growing pains, think bump mapping etc, to get more detail for little extra work from the hardware.

I'm not massively up to date on the scene, but I would be surprised that as our current approach plateaus we don't see different approaches rear their heads. Then again, we now have our new medicine, tessellation, to help us get through.. and the entire industry is built on one approach, so any move is going to be very very very very slow.

25 years is a long time, but I don't see real-time photorealistic 3d environments happening without a dramatic change in the technology (power/cooling/space) or approach. A breakthrough could happen tomorrow, or never..

I don't think changes in the fundamentals of graphics programming and hardware will come because of graphical limitations. We haven't seen any evidence yet of a polygon based approach limiting graphics, it's just hardware holding us back right now. I think the real reason we might change to voxels/point clouds and ray tracing is because they are more physically realistic models. This means that you get lots of realistic effects without any extra work. At the moment, as game worlds become more complex, the developers will have to write hundreds of shaders (geometry, and pixel) hugely complicated lighting models to simulate many effects. Furthermore, things like objects breaking fracturing are very hard to code/animate. This complexity could be the limiting factor which forces devs to move to more physically realistic models which give these effects for free.

 

[mame]

CPC 464 was my 1st also. Beast of a machine. Funny having to wait up to 40 minutes sometimes for a game to load from a cassette tape!

40 minutes the longest one I had was Sultans Maze which was 17 minutes

 

[gord]

We haven't seen any evidence yet of a polygon based approach limiting graphics, it's just hardware holding us back right now.

The hardware has been developed to drive the polygon approach, so is it not fair to say they are one and the same?

 

['Wes]

Isn't point to point rather than polygon the waqy forward? Someone posted a video about it a few weeks back.

I'm guessing you mean the guy who didn't sound too convincing and went on about how much better his point to point program is compared to the polygon way of things and apparently was in the line up to get his program used in games but was later turned down or something along those lines.

 

[rexehuk]

We will never need more than 2GB of onboard video ram, as graphics have reached the best possible and no one will want more - Rob 2010

Remind you of anyone?

 

[ShaolinDreams]

http://en.wikipedia.org/wiki/Ray_tracing_(graphics) what about ray tracing?

 

[.one.]

Forgot about Booty!

First game I ever bought for my Amstrad CPC464 (full colour monitor)

Will still have it somewhere and when I last booted my the old girl about 10 years ago, it still worked.

AH brilliant Booty was my first game i bought for the specy which we all know was better than the Amstrad ............walks off whistling ..........

ps.It was 1.99 and i still have the game

 

[AGD]

The hardware has been developed to drive the polygon approach, so is it not fair to say they are one and the same?

No, that's like saying current CPUs can't calculate a billion decimal places of pi in a second so we must have the wrong approach. The physics and technology behind minimising transistors and increasing frequencies has nothing to do with a polygonal based approached to graphics rendering. The point is that cards are getting better and so are graphics. You could maybe say you think progress would be more rapid using hardware optimised for voxels/ray tracing methods but that would just be pure (and by most accounts faulty) speculation on your part.

 

[gord]

No, that's like saying current CPUs can't calculate a billion decimal places of pi in a second so we must have the wrong approach. The physics and technology behind minimising transistors and increasing frequencies has nothing to do with a polygonal based approached to graphics rendering. The point is that cards are getting better and so are graphics. You could maybe say you think progress would be more rapid using hardware optimised for voxels/ray tracing methods but that would just be pure (and by most accounts faulty) speculation on your part.

It certainly wasn't my intention to make that claim. I'll try and clear things up.

Correct me if I am wrong, and I may be, as I said, I'm not fully up to speed with the scene but:

You are saying the entire pipeline based rendering architecture that has been developed for modern GPU's hasn't been influenced by the polygonal approach? The shader increases hasn't been influenced by the workarounds of bump mapping. And now, tessellation isn't the new buzz word, bringing us some more hardware changes?

It's indicative that you mention transistors and frequencies because this has been the methodology of increasing performance in applications generally. Graphics is much more specialised due to the approach adopted during its infancy and its an unsustainable approach. Eventually scene complexity will outweigh the physical die capabilities, so we need workarounds like tessellation, otherwise real time photo realism will never be possible.

 

[AGD]

It certainly wasn't my intention to make that claim. I'll try and clear things up.

Correct me if I am wrong, and I may be, as I said, I'm not fully up to speed with the scene but:

You are saying the entire pipeline based rendering architecture that has been developed for modern GPU's hasn't been influenced by the polygonal approach? The shader increases hasn't been influenced by the workarounds of bump mapping. And now, tessellation isn't the new buzz word, bringing us some more hardware changes?

It's indicative that you mention transistors and frequencies because this has been the methodology of increasing performance in applications generally. Graphics is much more specialised due to the approach adopted during its infancy and its an unsustainable approach. Eventually scene complexity will outweigh the physical die capabilities, so we need workarounds like tessellation, otherwise real time photo realism will never be possible.

Now I'm confused and I really don't get what point you are trying to make. I'm not saying that modern GPU architecture hasn't been affected by the polygon approach - obviously the complete opposite is true, the architecture is dictated by the approach. What I am saying is that the main limiting factor right now is not that architecture or the polygon based approach. If you're saying that polygons won't be suitable in the future then I agree with you and I've already mentioned a few times why I think this will be the case. What I am saying though is that polygons will continue to be the best option for a while yet. It doesn't really make sense to pick out graphics for having an unsustainable approach - all CPU and GPU architectures and paradigms will change and improve in the future.

 

[gord]

Exactly, so my original post was what you've just posted but with a 25 year time-frame on it. That's long enough for a technological or methodological change to graphics, its just the probability of one of those changes happening that I questioned.

Basically, if you asked me whether or not in 25 years we would have real-time avatar graphics, I'd say "not unless something changes dramatically".

 

[soundb0y]

If you want to see the future of graphics then take a look at the 3dmark 11 demo video.

Watched this last night and I have never seen anything look so good. The images were film quality. It is of 2 deep sea submersibles viewing a wrecked deep sea drilling platform.

I was blown away by how good it looked. At some points I was convinced it was an actual film and not a pc rendering this.

Thought this was some kind of joke about BP at first

 

[LeJosh]

I don't think graphics are at the forefront of developers minds now, I think it's all about physics now. More and more games having better physics, BC2 has it, it's premature I felt but it still has destruction. It will be sought after in MMOs I feel aswell.

 

[bhindi]

having been playing with computers from commodore vic 20 to what i have to today i never would have thought i be writing this on a LAPTOP !

For me god knows where things will be in 10 years never mind 25 years . I think one thing for sure laptops will be gone and so will be desktops . The future pc will be on a contact lense and works on eye movemet and brain signals. Getting back to games and graphics how good was the game " DIZZY" i was amazed by it as a kid on my speccy.

 

[Chojin]

If we drew a linear line from here, in 25 years time the current gen console will be something like an Xbox 6. I think by that time you could easily expect real time graphics surpassing the quality seen in 3d animated films today. The reality is that technology is improving exponentially and the rate of progress over the next 25 years will be a steep curve compared to the last 25 years. I think the popularity of the Wii gives an indication of where people want to go with gaming, and any stagnation in the market in the coming years is unlikely. Plenty of innovation to come. I expect telepathy and holographic displays in this time period but wouldn't be surprised if we ditch the physical world and start using nanobots.

 

[mushroomsoup]

http://www.youtube.com/watch?v=THaam5mwIR8&annotation_id=annotation_54092&feature=iv

anyone heard of this? 'unlimited detail technology'?

"Unlimited Detail is a fourth system, which is more like a search algorithm than a 3D engine. It is best explained like this: if you had a word document and you went to the SEARCH tool and typed in a word like MONEY the search tool quickly searches for every place that word appeared in the document. Google and Yahoo are also search engines that go looking for things very quickly. Unlimited Detail is basically a point cloud search algorithm. We can build enormous worlds with huge numbers of points, then compress them down to be very small. The Unlimited Detail engine works out which direction the camera is facing and then searches the data to find only the points it needs to put on the screen it doesnt touch any unneeded points, all it wants is 1024*768 (if that is our resolution) points, one for each pixel of the screen. It has a few tricky things to work out, like: what objects are closest to the camera, what objects cover each other, how big should an object be as it gets further back. But all of this is done by a new sort of method that we call MASS CONNECTED PROCESSING. Mass connected processing is where we have a way of processing masses of data at the same time and then applying the small changes to each part at the end." - quote

i don't really know much about graphics, any opinions anyone who knows about it?

 

[Devrij]

Yeah that was posted up a while back, but the problem with that is animating it apparently. Notice that everything in that video is standing still. Otherwise it's a good idea.