Ubisoft's Lighting And Physics Tech For Splinter Cell

I know it said that of element-to-element comparisons, but 1 pair comparisons does not equal a solution for a given point. And now that I look back at it, I do see that I was mixing up my figures -- numbers came out 510,000 results actually computed per second on a 6800, not 150... which is still total crap, but not as bad as I had in my mind. It still doesn't change the real problem it had when I tried it (and I haven't really thought about it since, so I'd totally forgotten what sort of throughput I got out of it) -- it could not maintain that sort of throughput beyond scene complexities of a certain size. While it supposedly does scale O(n log n) in the algorithmic sense, it doesn't scale well at all when that hierarchy gets big and deep because you'll end up memory access bound really fast. You might have noticed that the demonstrations don't go beyond the point of a scene containing even 20k verts, and that's probably also true of the Fantasy Lab demonstrations.

The only major problems which he claims to have fixed with the newer version include not needing to do multiple passes to get multiple bounces and working in conjunction with the displacement mapping technique.

Thanks for the insight, that's good info. It'll be interesting to see if and when Fantasy Lab ever comes to demonstrating the game they're supposedly developing, to get an idea of scalability now.
 
Duno if you've checked the Fantasy Lab website of late, but they have got a new screeny. Still not very involving, but more complex than the original rhino in armour. Nice looking game too. Scalability isn't proven, but then they'll be really up the swanny if they're producing this simple still now with their game engine, and then find the engine can't actually render the game they've got in production!

Perhaps it'll be very small and local areas, like screens in an old adventure game?
 
the shots with the guy in, looks to use the same method ive explored (video on my site btw) where the geometry is stored as a low frequency representation
 
There's a trailer at GameTrailers. The lighting's definitely very good, but it's hard to tell what's dynamic. Something I looked at for a clear example was the explosion, and yet there's no light cast. So the lighting solution certainly isn't 'complete.' There's a snippet of fight in a cafe/restaurant, but the editing's so jumpy it's had to follow any bit to look for lighting changes.
 
Thanks for the insight, that's good info. It'll be interesting to see if and when Fantasy Lab ever comes to demonstrating the game they're supposedly developing, to get an idea of scalability now.

Sorry to diverge the disscussion of SC:C again but I'm intruiged by this Fantasy Labs Engine..

On their website they claim that the scene with the Sub-division surface modelled Rhino in the desert could be rendered using the GI solution at 300 fps on a GeForce 7900 GTX.. I sounds pretty good although I understand that the scene complexity here isn't the largest it could be..

I was wondering though, how suitable would the engine be for games which don't particularly require vast open environements? (a fighting game a la Tekken for example?)

Granted scalability of such an engine would be problematic if you tried to make GTA, but surely there are alot of game type which don't have large scale proxy requirements and therefore could allow developers to still utilise the technology to some degree?

Just curious..?
 
I was wondering though, how suitable would the engine be for games which don't particularly require vast open environements? (a fighting game a la Tekken for example?)
We don't know whta their engine really does, but in theory, yes, some simpler environment titles should benefit. As I've said before, one game that should be photorealistic by now is Snooker/Pool. Nothing but sphere and baize, you could do all sorts of GI and reflection calculations with the simplest of primitives to work with. I'd like to see more devs target a minimalist game scope and focus on the quality of it. Another titles that could look awesome would be Eye Of Judgement. They could design it around an advanced, realistic lighting model and tried to get everything looking real and solid, especially at the interactive AR view.
 
Right ive just seen the 720p footage from Ubi Day and i must say im completely dissapointed. The game while it looks OK it is'nt mind blowing like all the previous SC releases, some of the textures remined me of PS2.
 
On their website they claim that the scene with the Sub-division surface modelled Rhino in the desert could be rendered using the GI solution at 300 fps on a GeForce 7900 GTX.. I sounds pretty good although I understand that the scene complexity here isn't the largest it could be..
Actually, in some interviews, they clarify that. What the statement says is that their simulation alone for that scene can run at 300 Hz. The actual rendering still runs at what they claim to be "greater than 30 fps."
 
Back
Top