If PC Gaming Dies, Will it Really Matter??? - Page 8

Edge · 02-12-04, 03:06 PM

I simply think we need hardware acceleration for light reflections. It would make things a LOT easier, and if a 5 year old soundcard can trace SOUND reflections, why can't a modern day videocard be made to trace LIGHT reflections? Though of course, it's much harder to trace light reflections, but you get the idea. Fitting that the "holy grail" of both realistic sound and graphics is the reflections they make off of objects. I think faking radiosity would be easy if someone would actually try. For example, you could have a negative light radius around an object, so the closer it gets to something the darker the surface gets. Obviously, it's far from a perfect solution, but I think it would go a long way into faking radiosity, and therefor "realistic" graphics, especially if it's combined with projection soft shadows.

I'm hoping that at least one graphics company makes the push for good lighting, and others will follow suit. Right now most of the lighting and shadow calculations in games is still handled by the CPU. If we could use actual hardware to replicate it, graphics can make a big jump practically overnight. We honestly haven't made any major graphic jumps since the Geforce 3 was released 3 years ago, and even with that it's rarely used well. Maybe DX10 will offer an prototype system for radiosity...

Oh, it seems we've gotten a bit off-topic, but since the topic creator was banned I say we can turn the thread into whatever we want

Chalnoth · 02-12-04, 09:11 PM

Quote:

Originally posted by Edge
I simply think we need hardware acceleration for light reflections. It would make things a LOT easier, and if a 5 year old soundcard can trace SOUND reflections, why can't a modern day videocard be made to trace LIGHT reflections?

The simple answer is that light has to be traced at a much higher resolution. Anyway, raytraced lighting really isn't that good, and there are tricks available on modern video cards that get you most of the way to raytraced lighting, without the huge performance impact (provided you're not worried about multiple reflections and stuff...).

Quote:

I think faking radiosity would be easy if someone would actually try. For example, you could have a negative light radius around an object, so the closer it gets to something the darker the surface gets.

I'm not sure that has much at all to do with radiosity.

Anyway, lighting is indeed the big problem, and it's not a small one. Doing proper lighting just requires algorithms that are undoable in realtime. People are working on approximations that get you 99% of the way there, but it's just taking time to get the processing power and flexibility up high enough.

One possible bit of excitement is that if quantum computing does start to become viable in the next 5-10 years, it won't just increase processing power, but it will open the way to whole new algorithms for computing that just weren't available before (basically, since a quantum computer works differently, it likes different kinds of algorithms altogether...it would be comparitively poor for running today's programs, but would be excellent for other types of programs...).

PenguinJim · 02-12-04, 09:36 PM

PleeeeaaaAAAaase close this thread

it brings the feeling of unending pain to my head.

Edge · 02-13-04, 12:20 AM

Quote:

I'm not sure that has much at all to do with radiosity.

Well, basicly, the closer an object gets to a surface, the more light it blocks out within a certain area. What making a negative light on an object would do is simulate the blocking out of light it causes against other surfaces. For example, if you hold your hand close to a wall, you'll notice it the wall area right by your hand gets really dark, and since the farther away you something is the less light it blocks, using a linear drop-off negative light on something would have a similar effect (blocks out a lot of light when close, only blocks out a little when it's far away). It's sorta taking the "don't fix the dam, just waterproof your house" method of doing it, but I did test it out one time and it did look decent. Actually it does the opposite of radiosity: rather than tracing the actual light reflections, it would be an estimation of where the light WOULDN'T go. If HL2 or Doom 3 supports negative dynamic lights I'd actually like to try it out and see what the result is. Though it also might end up looking pretty crappy, depending on how it's implimented.

I'm not sure what's going on with quantum computers though. They certainly have potential, but they're so...different...that it would take a long time before we saw any practical consumer stuff out of it (it would work pretty good for things like encriptions and such though). Still, something to look forward to I guess, I just hope it doesn't put me out of a job

Chalnoth · 02-13-04, 02:03 AM

Radiosity does much more than occlude light. Radiosity lighting would also calculate the light that was reflected, or absorbed and re-emitted, from every object.

Edge · 02-13-04, 05:56 AM

Well, that's why it's going to take a very long time to get real radiosity in a game. My idea was basicly a hack to get "radiosity-like" effects without a big performance hit, but obviously it's not implimented the same way at all. Until an engine can acurately bounce light off of every surface, with it's reflection amount and color changed depending on the material, we won't see "true" radiosity. But in the same way, games like Splinter Cell didn't have "true" shadows either. Most of the time the engine only kept track of where the light DIDN'T shine, not where it DID shine, and my idea is basicly the same thing except for radiosity instead of shadows.

Oh, and here, for an example of what I'm talking about, I rendered this scene quick using one of the random Lightwave models I made a while ago. These two scenes are identical, EXCEPT that in the right one I added 3 negative lights by the model (actually I was a bit inefficient, I could've only used 2 for almost the same effect). Now it may not be real radiosity by any means, but you gotta admit that for such a small change it does a pretty nice job of faking one of the effects that radiosity has on shadows.

Chalnoth · 02-13-04, 07:59 AM

That effect may be okay for high levels of ambient lighting (such as you'd get inside, say, a supermarket), it wouldn't be okay for many situations. Even normal, hard shadows would look better.

Edge · 02-13-04, 04:36 PM

Well, it's not supposed to be a real shadow. Remember, there are NO shadows in that picture, only a couple of negative lights. The effect is best combined with actual stencil shadows or something like that. Used properly, I think it's one of the best tactics to use to simulate radiosity shadows without a big performance hit (keeping in mind that it uses the opposite approach to drawing fake radiosity).

And actually it does work pretty good in almost any lighting situation. Even with minimal environment light, it still creates a nice effect as long as you lower the intensity of the negative light. Anyway, here's a pic showing what happens when you use the negative light idea with stencil shadows (both pictures are using basic point-lighting):

02-13-04, 02:03 AM	#89
Chalnoth Registered User Join Date: Jul 2002 Posts: 1,293	Radiosity does much more than occlude light. Radiosity lighting would also calculate the light that was reflected, or absorbed and re-emitted, from every object. __________________ "Physics is like sex. Sure, it may give some practical results, but that's not why we do it." - Richard P. Feynman

02-13-04, 07:59 AM	#91
Chalnoth Registered User Join Date: Jul 2002 Posts: 1,293	That effect may be okay for high levels of ambient lighting (such as you'd get inside, say, a supermarket), it wouldn't be okay for many situations. Even normal, hard shadows would look better. __________________ "Physics is like sex. Sure, it may give some practical results, but that's not why we do it." - Richard P. Feynman

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Why Buying A Cloud Gaming Company Could Be The Smartest Move Sony Has Made This Gener	News	Latest Tech And Game Headlines	0	06-01-12 07:50 AM
Rumor: Cloud Gaming Deal with Sony 'Close to Being Signed'	News	Latest Tech And Game Headlines	0	05-30-12 05:00 PM
Nvidia On Cloud As The Future Of PC Gaming	News	Latest Tech And Game Headlines	0	05-30-12 02:40 PM
Product Review ' SteelSeries Scope Gaming Eyewear by Gunnar	News	Latest Tech And Game Headlines	0	05-18-12 09:30 PM

02-12-04, 03:06 PM	#85
Edge 3d animator for hire! Join Date: Oct 2002 Location: Minnesota, USA Posts: 3,564	I simply think we need hardware acceleration for light reflections. It would make things a LOT easier, and if a 5 year old soundcard can trace SOUND reflections, why can't a modern day videocard be made to trace LIGHT reflections? Though of course, it's much harder to trace light reflections, but you get the idea. Fitting that the "holy grail" of both realistic sound and graphics is the reflections they make off of objects. I think faking radiosity would be easy if someone would actually try. For example, you could have a negative light radius around an object, so the closer it gets to something the darker the surface gets. Obviously, it's far from a perfect solution, but I think it would go a long way into faking radiosity, and therefor "realistic" graphics, especially if it's combined with projection soft shadows. I'm hoping that at least one graphics company makes the push for good lighting, and others will follow suit. Right now most of the lighting and shadow calculations in games is still handled by the CPU. If we could use actual hardware to replicate it, graphics can make a big jump practically overnight. We honestly haven't made any major graphic jumps since the Geforce 3 was released 3 years ago, and even with that it's rarely used well. Maybe DX10 will offer an prototype system for radiosity... Oh, it seems we've gotten a bit off-topic, but since the topic creator was banned I say we can turn the thread into whatever we want

02-12-04, 09:36 PM	#87
PenguinJim Join Date: Oct 2002 Location: I live in my chair in the UK Posts: 440	PleeeeaaaAAAaase close this thread it brings the feeling of unending pain to my head.

02-13-04, 05:56 AM	#90
Edge 3d animator for hire! Join Date: Oct 2002 Location: Minnesota, USA Posts: 3,564	Well, that's why it's going to take a very long time to get real radiosity in a game. My idea was basicly a hack to get "radiosity-like" effects without a big performance hit, but obviously it's not implimented the same way at all. Until an engine can acurately bounce light off of every surface, with it's reflection amount and color changed depending on the material, we won't see "true" radiosity. But in the same way, games like Splinter Cell didn't have "true" shadows either. Most of the time the engine only kept track of where the light DIDN'T shine, not where it DID shine, and my idea is basicly the same thing except for radiosity instead of shadows. Oh, and here, for an example of what I'm talking about, I rendered this scene quick using one of the random Lightwave models I made a while ago. These two scenes are identical, EXCEPT that in the right one I added 3 negative lights by the model (actually I was a bit inefficient, I could've only used 2 for almost the same effect). Now it may not be real radiosity by any means, but you gotta admit that for such a small change it does a pretty nice job of faking one of the effects that radiosity has on shadows.

02-13-04, 04:36 PM	#92
Edge 3d animator for hire! Join Date: Oct 2002 Location: Minnesota, USA Posts: 3,564	Well, it's not supposed to be a real shadow. Remember, there are NO shadows in that picture, only a couple of negative lights. The effect is best combined with actual stencil shadows or something like that. Used properly, I think it's one of the best tactics to use to simulate radiosity shadows without a big performance hit (keeping in mind that it uses the opposite approach to drawing fake radiosity). And actually it does work pretty good in almost any lighting situation. Even with minimal environment light, it still creates a nice effect as long as you lower the intensity of the negative light. Anyway, here's a pic showing what happens when you use the negative light idea with stencil shadows (both pictures are using basic point-lighting):