As I said earlier, animated voxels have already been solved by other people (with enough integrity to publish their research) -- it's no longer a difficult/impossible problem.

I'm not saying it's not possible for UD to animate their scenes, just that they need to be honest about their current capabilities when it comes to animation.

It's not voxels!

You're really saying that this has more detail than current game characters? What? Really?[/quote]
I think i saw a different one that was not made by them and got confused :-/ I was wondering around youtube for a while on point clouds and it was early in the morning.

notch both ignored any sort of compression or optimization on voxels in his data size analysis and didn't realize that the demoed tech IS NOT VOXELS. It's point cloud based.[/quote]This has absolutely nothing to do with their lack of honesty...


Anyway, uncompressed sizes of the data is still important during production (and drives home the point that their island is not uniquely modelled, but is repeated instances of the same objects -- something the CEO fails to mention), [/quote]
Why are uncompressed sizes important? They could be very important to their method. The way they talk about it indicates that the way the compress the data IS the important part of the technology. Why would that not be important? That still doesn't make notch's point accurate, as he assumes that the entire volume is entirely filled by atom sized voxels, which the tech is not, rather than voxels of varying sizes only representing geometry that actually exists in the game world. His estimate is overblown to say the least, which is exact problem he's complaining about.

and seeing their tech is top-secret, with no real descriptions beyond marketing-hype, there's no way to know if it's point-clouds, or voxels, or even just *bruce dell voice* tiny little atomic polygons.[/quote]

He says in the video that it's point clouds. It's in the first minute of the video and talks about laser scanning in objects to get pure point cloud data for the engine to work with.

[quote name='way2lazy2care' timestamp='1312504876' post='4844802']He says in the video that it's point clouds. It's in the first minute of the video and talks about laser scanning in objects to get pure point cloud data for the engine to work with.

He also says that it allows us to put unlimited amounts of geometry into games, which is obviously a distortion. ...Unless he's making "Mandelbrot, the game".

And again, he dumbs stuff down so much, only ever touching on the edge of any real terminology (which he calls "technobabble"), that you can't really expect him to be using the correct terms for anything. For all we know, he's converting point-cloud data into a voxel representation, or turning voxel data into a point-cloud representation, or polygons into point-sprites into distance-fields --- the way he talks, you really can't tell.[/quote]
It's essentially an ad for his company. Why would you expect him not to promote or be excited about a future product or go into technical details of an unfinished product?

Uncompressed data sizes are important when authoring data. They have an impact on the pipeline (of which there is little to no information), so without further information, there's uncertainties about some pro's/con's in their art work-flow.[/quote]
If uncompressed data sizes were that huge an issue zbrush wouldn't even exist. We already have the problems you are pointing out with his system with traditional methods except with his you probably won't have to make your artists retouch a model 20 times to cut out extra polygons or do LOD work.

Again, I was talking about Notch's 'honesty' argument though, why are you pointing out the assumptions he made in a different point?
[/quote]

Because notch's argument doesn't use "honest" examples in the first place. Everything should be viewed through the scope of advertisement sure, but just because someone is being a salesman doesn't mean he's selling snake oil.

...and that's not what I'm saying. I actually pointed out earlier how mundanely feasible it all is...
The problem isn't that he's selling snake oil; he's selling common vegetable oil while telling you that it comes from secret space plants and that you've never tasted anything like it.

I have no real attachment to the tech. I am just dumbfounded that you think it wouldn't have an impact on art quality and productivity.

If it is what he says it is it would have a huge impact the same way voxels could if they met their potential. You said that with megatextures and applying similar methods to polygons we no longer have texture or polygon budgets, but clearly there is a need for fewer performance bottlenecks for higher detailed art or we would have more detailed art. Drawing a model in zbrush, exporting it to a poly mesh, and optimizing it at 3-10 LODs is not an optimal solution for artists. The fact that you can use unmodified high detail point clouds straight from the scanner is a huge selling point alone. Send dave to a cathedral for an afternoon and have a high quality textured model of it by the time he gets back. That would be awesome.

I'm excited about it for the same reason I'm excited about light field technology. It's something cool that is revolutionary rather than evolutionary. There isn't a crap load of evidence that light fields are awesome that couldn't easily be manufactured without light-fields, but I still think those are awesome too. Excuse me for being excited about the possibilities of the future instead of being content to just keep throwing more horsepower at the tech we have today.

edit: light field stuff more related to light fields being awesome for the average consumer.

Raycast shaders (POM)

*cough* QDM *cough* There is a lot of hate. I won't stir the pot.