Its like FV opened a graphics company. No other explanation.
Name:
Anonymous2010-03-10 6:49
I'm surprised Mat Dickie hasn't thought of this first.
Name:
Anonymous2010-03-10 7:49
>>3
Mdickie focuses on content, not on the technicalities of presentation. He is an artist, after all. Did you know that the initial mock-up of the "You Testament" was roguelike?
Name:
Anonymous2010-03-10 7:57
>“Unlimited Detail algorithm”
I want to believe.jpg
Name:
Anonymous2010-03-10 7:57
>>5
I can only see the file name but not the image itself. Is this bad?
Name:
Anonymous2010-03-10 8:03
>>6
Its too much detail for your eyes, so your brain filters it automatically
Name:
Anonymous2010-03-10 8:17
>>5
It uses procedural algorithms. You can describe anything at any level of detail as you want with the power of procedural asset generation.
Name:
Anonymous2010-03-10 8:28
How does it work?
If you have a background in the industry you know the above pictures are impossible. A computer can’t have unlimited power and it can’t process unlimited point cloud data because every time you process a point it must take up some processor time. But I assure you, it's real and it all works.
Unlimited Details method is very different to any 3D method that has been invented so far. The three current systems used in 3D graphics are Ray tracing polygons and point cloud/voxels, they all have strengths and weaknesses. Polygons runs fast but has poor geometry, Ray-trace and voxels have perfect geometry but run very slowly.
Unlimited Detail is a fourth system, which is more like a search algorithm than a 3D engine. It is best explained like this: if you had a word document and you went to the “SEARCH” tool and typed in a word like “MONEY” the search tool quickly searches for every place that word appeared in the document. Google and Yahoo are also search engines that go looking for things very quickly. Unlimited Detail is basically a point cloud search algorithm. We can build enormous worlds with huge numbers of points, then compress them down to be very small. The Unlimited Detail engine works out which direction the camera is facing and then searches the data to find only the points it needs to put on the screen it doesn’t touch any unneeded points, all it wants is 1024*768 (if that is our resolution) points, one for each pixel of the screen. It has a few tricky things to work out, like: what objects are closest to the camera, what objects cover each other, how big should an object be as it gets further back. But all of this is done by a new sort of method that we call MASS CONNECTED PROCESSING. Mass connected processing is where we have a way of processing masses of data at the same time and then applying the small changes to each part at the end.
The result is a perfect pure bug free 3D engine that gives Unlimited Geometry running super fast, and it's all done in software.
This is pretty cool. I wanna start a free implementation of this.
>>11
His point does still remain. The background is s static, albeit very awesome, composition of trillions of dots; the only thing that's moving is the viewport. Does the processing efficiency of of UDT hold up when everything has been turned into trillions of chunks of jello during an earthquake?
Unlimited Detail leverages search algorithms and MASS CONNECTED PROCESSING through enterprise-quality voxel patterns to provide clients worldwide with robust, scalable, modern 3-D Engine implementations of flexible, personalized, cutting-edge futuristic graphic application product suite e-solution point cloud architecture that accelerate response to player and real-world CGI demands and reliably adapt to evolving technology needs, seamlessly and efficiently integrating and synchronizing with their existing legacy renderers, enhancing the visual capabilities of their desktop environments across the enterprise while giving them a critical competitive advantage and taking them to the next level.
Name:
Anonymous2010-03-10 10:53
>>13
Apparently, it is quite cheap to display a static scene using the CPU. I would bet that this system doesn't take significant cycles to include the sort of movement you'd find in any current 3D game.
Well, looking at the video, that wasn't quite as retarded as I thought it would be.
If it delivers it sounds like a good match for procedurally generated content. They also suggest laser scanning on their web site. Both nice ways to generate lots of detail without significantly increasing development costs.
Thin on technical details, though.
I think it's safe to assume that it takes quite a bit of memory, and that it needs fast random access. How well does it parallelise? How well does it deal with effects like shadow? Anti-aliasing? Animation?
You could probably do a hybrid system where you paint the detailed background first, and then lay over moving parts drawn with polygons, but how far does that get you when nowadays, every straw of grass is expected to sway in the breeze?
Name:
Anonymous2010-03-10 11:42
trollface.jpg
Did anyone notice the "Unregistered version" watermark that shows up for few seconds?
Spinning fractals/procedural generation as something new?
I've once seen some impressive game fit in 96K using such tricks, but loading/generation took forever, also high memory and CPU usage - those GPUs are there for a reason - so you could offload stuff to them. I'll excuse that demo from the GPU rules as it was made a long time ago when it wasn't as popular to offload computation to GPUs for less traditional tasks.
Name:
Anonymous2010-03-10 11:56
How does point cloud geometry work in the first place? When you talk about a trillion points, you still have a trillion points being manipulated in data form. There's probably some form of cohesive nature being simulated but how do you define a unique element of the scene? Moreover, are all the points strictly defined, do they behave in a temporary quantum nature, or are they simulated as needed from a series of control points using some method similar to vector fields (or is that too polygon thinking)?
Name:
Anonymous2010-03-10 12:20
>>21
They're not being manipulated, just displayed (by the search algorithm they've vaguely described), which is why there is no animation (as the rest of the thread have pointed out - manipulating trillions of data points to animate them would be very expensive). A unique element is defined as with any other 3D scene, using 'models' (or the equivalent) and placing them in places. I imagine the point definition could be many things, like cartesian floating-point (which would easily allow for scaling and such) or indeed vector fields.
This is fucking ridiculous. It's just a glorified raytracer, Epic games and iD Software are already working on this and they have actually working implementations (if you look really hard you'll find some videos). IIRC iD software's implementation easily hit 60fps on a GTX280. They mentioned stuff like recursing the traces that hit certain stuff in order to generate as much detail procedurally as you want, and also the almost-constant render time. Another big advantage is automatic resource management, so you can put as much detail as it'll fit in the disk and load it smoothly as needed, based on distance. Current compression ratios are about one bit per "texel"/"voxel"/whatever you want to call it, which is pretty reasonable. Also this doesn't work for dynamic stuff, so monsters, characters and even fucking doors have to be drawn as polygons still.