Unlimited Detail Real-Time Rendering Technology

  • Thread starter Thread starter Raiser
  • Start date Start date
  • Views Views 4,419
  • Replies Replies 36

Raiser

I am mad scientist.
Member
Joined
May 31, 2008
Messages
1,563
Reaction score
42
Trophies
1
Website
Visit site
XP
681
Country
Canada
Take a look at a preview video just released today:

[youtube]00gAbgBu8R4[/youtube]

It's actually not too hard to follow and is very, very intriguing.

Enjoy!
 
If it is real, imagine kinect (or any console with a good camera) just scan you right into the game, atom by atom. O.o

Although, not sure if it really any point.
unsure.gif


How long would games take and how much will they cost?

Is it possible to have better graphics than the irl world itself?
laugh.gif
 
The main question is: will graphic artists take time to create a fully detailed 3D tree instead of a making a quicker, duplicate, less detailed version?
The more time they spend on game, the more it costs to produce. Not all companies have unlimited budget. Though, it's good that objects can be scanned instead of modeled.


I'm more curious to understand how it works on the graphic cards level.
It seems it's still polygons object, so how is working the rendering? Will it work on old/current hardware?

Thank you for the updated trailer.
I saw the first one last year, it's nice to see they are working on it.
 
Well. If this just requires processing power and very little graphics, the PS3 is ready for a longer life.
...If you have enough fans
tongue.gif


But seriously, I don't really get how this thing works.

Is a 2GB Radeon gpu worth £200 mean squat when using this engine? Is the only thing that matters is the processor? Does it just need a huge mother frigin hard drive, with insane loading screens and tons of ram?
I'm at a loss.

It sounds good an' all, but if it needs some expensive shiz to run it...I couldn't care less about the engine.
 
Cyan said:
The main question is: will graphic artists take time to create a fully detailed 3D tree instead of a making a quicker, duplicate, less detailed version?
The more time they spend on game, the more it costs to produce. Not all companies have unlimited budget. Though, it's good that objects can be scanned instead of modeled.That's true.
I guess that, in order to succeed, they should spend a lot of time into making an easy-to-use, quick and flexible programming environment for developers. Otherwise, most of the gaming companies will skip this engine.
QUOTE(Cyan @ Aug 2 2011, 08:20 AM) I'm more curious to understand how it works on the graphic cards level.
It seems it's still polygons object, so how is working the rendering? Will it work on old/current hardware?
For what I understand, still uses polygons but, instead of making an object with a group of polygons, this engine uses a X quantity of polygons, till the point that imperfections are imperceptibles to the human eye.

You could call those atoms "very small polygons"
laugh.gif
 
KingVamp said:
NahuelDS said:
You could call those atoms "very small polygons"
laugh.gif
I still don't see how it is possible without better hardware.
[youtube]Q-ATtrImCx4[/youtube]
watch this. The guy says: "in the future, will we had so much computer power that we will abandon polygons and move to the point cloud dotter system"

So, in short... this definitely needs better hardware

EDIT: my mistake... watch the entire video... it's explained at the end. very clever idea!!
laugh.gif
 
But... How will they make animation? With polygons, it's simply making new polygons to fit the "movement". With this, you basically have to reconstruct the item from the ground up with a slightly different pose.

Imagine how much processing it would take to move 10 in-game characters made like this...
 
I'm not a whore for graphics, so this doesn't mean a whole lot to me, but it is pretty impressive none the less.

Now if only this guaranteed quality games and not just quality visuals.
 
I'm not so much about graphics... but I do enjoy them in games that are TRYING to look real.
This is amazing. But the way I see it, we won't see it for a while.


Posts merged

I'm not so much about graphics... but I do enjoy them in games that are TRYING to look real.
This is amazing. But the way I see it, we won't see it for a while.

AWWWW SCAM?
ohnoes.png
 
Slyakin said:
But... How will they make animation? With polygons, it's simply making new polygons to fit the "movement". With this, you basically have to reconstruct the item from the ground up with a slightly different pose.

Imagine how much processing it would take to move 10 in-game characters made like this...

It'll work exactly the same as normally, you won't move polygons, however you'll most likely be able to use surfaces made of voxels, pretty much "like" using standard model bones.

This also includes destructive envioriments - rather than deleting one model and swapping it for another, using voxels you'll be able to splash them all over the place at will.

I'd also like to point out that Notch is counting each voxel cluster separately, which doesn't necessarily have to be the case. Once a given structure made of voxels is in-memory, it can be spawned on-screen multiple times without increasing memory usage.
 
Foxi4 said:
Slyakin said:
But... How will they make animation? With polygons, it's simply making new polygons to fit the "movement". With this, you basically have to reconstruct the item from the ground up with a slightly different pose.

Imagine how much processing it would take to move 10 in-game characters made like this...

It'll work exactly the same as normally, you won't move polygons, however you'll most likely be able to use surfaces made of voxels, pretty much "like" using standard model bones.

This also includes destructive envioriments - rather than deleting one model and swapping it for another, using voxels you'll be able to splash them all over the place at will.

I'd also like to point out that Notch is counting each voxel cluster separately, which doesn't necessarily have to be the case. Once a given structure made of voxels is in-memory, it can be spawned on-screen multiple times without increasing memory usage.

No it wont.
QUOTE
Another weakness is that voxels are horrible for doing animation, because there is no current fast algorithms for deforming a voxel cloud based on a skeletal mesh, and if you do keyframe animation, you end up with a LOT of data. It’s possible to rotate, scale and translate individual chunks of voxel data to do simple animation (imagine one chunk for the upper arm, one for the lower, one for the torso, and so on), but it’s not going to look as nice as polygon based animated characters do.
 

Site & Scene News

Popular threads in this forum