When I read all that stuff I think I really am happy I didn’t start with crazy hardware like some of you guys did , I believe this could have been a lot of fun, but I definitely prefere todays hardware that allows you for realtime rendering with almost eye quality…
I’m quite satisfied with modest technology what we have now but sometimes I wonder what will “OpenGL guys” be talking about in 100 years…Bumpmapping, per-pixel lighting, shadow volumes seem already to be closed topics , dunno but don’t think many were talking about that say 2 years ago here…
So what’s next? I’m a bit afraid we’re slowly running out of new visual effects to invent.
Do you have any ideas on what will “we” do in not so far future like 2103?
Running out of effects? Don’t think so. My largest problem is that the hardware don’t always have the capabilities to do my effects. Even the day the hardware supports “everything” we will still invent new cool stuff, creativity isn’t going anywhere.
My problem these days is that HW is improving so fast that I don’t have ideas on how to use it… Basically, I don’t have the time to think about what I could do with all this stuff (3D graphics is part of my job but not the main part)…
That being said, some people are really good at using new capabilities (Humus being one of them BTW!!!).
Regards.
Eric
[This message has been edited by Eric (edited 01-10-2003).]
My largest problem is that the hardware don’t always have the capabilities to do my effects
Sure hardware will still be faster and faster, but that only means you can say put real detailful grass with all that shadow stuff into your demo instead of some bumpy fakes, but it’s something different.
[This message has been edited by MickeyMouse (edited 01-10-2003).]
However, for those of you under the age of 35, take a look at copies of IEEE Transaction on on Computer Graphics from 1983 to see what 20 years of development can give you. Then extrapolate exponentially.
theres still a hell of a way to go in cg.
we’re not even close to realtime realism.
eg top of the line supercomputers can churn away for hours creating a single frame of a cg person (for a film whatever) yet it doesnt look realistic.
though cg seems to me to be accelrating recently (over the last couple of years) its very hard to keep up.
the major problem i see is computers cant store all the vast amounts of data that will be needed to create realistic scenes.
solution compress objects.
eg use procedural things, curves etc
So what’s next? I’m a bit afraid we’re slowly running out of new visual effects to invent.
Running out? Everything in cg, has and will always be, a compromise. Correction, everything digital. Just raise the bar a little. When doom came out I remember thinking, wow does it get any better? Many years later I found research websites such as stanford’s and cornell’s, which answered all my questions.
I saw a kind of science + information show yesterday (I had seen it before, long time ago). I don’t know which movie it was, but it had some boats on the sea, all computer generated.
They said it took them 4 months just to do the mist and they used 4 billion particles and they did the modelling using Maya!
I think that what is more required than technics is brute force computation. Instead of using hundreds of high end machines working in parallel for months, it could be done by 1 machine in real time.
Having something like that on my desk would really impress me.
My point of view is a little bit different : I think the real question is not what can we do with so powerfull hardware, but who have the money to produce top class product that take advantage of this power ? The answer to the question “what the future of Realtime CG hardware is ?” is “It is what we do now in CG : global illumination, subdivision surface, global physic engines, etc…”.
The real problem is the amount of data and the complexity of the code to process it. A Quake 1 model is 200-300 triangles, the initial Doom 3 model is build with million triangle and highdef textures. The cost of models, the time and knowledge invested in the engines becomes to high : that’s the real problem of the game industry.
I think the future of the hardware are tools that make a team able to produce rich advanced games for reallistic cost.
Gaby
PS : I’m really happy to have working on very slow processors without graphics cards, because it is the best path to learn how to have a fast, light and clean code, and to know “how” our current system works and how to build it in the most efficient manner. The actual power make a lot of developpers programming in a dirty manner, without searching to be efficient, I think.
There will always be a STAR WARS EPISODE DCXXIV The last Jedi before the Next. To show new great cool effects, that then will be pursued to be made real time…
, I believe this could have been a lot of fun, but I definitely prefere todays hardware that allows you for realtime rendering with almost eye quality…
, dunno but don’t think many were talking about that say 2 years ago here…