Unlimited Detail Technology
category: offtopic [glöplog]
i find it sad that:
- their video is so ugly and static.
- they believe the most significant piece of technology of the decade.
- that all games suffer from such problems as tree bases with only 6 sides
- they believe their method is very different to any 3D method that has been invented so far
- they have to compare their low-level algorithm with high-level web-search engines
- their video is so ugly and static.
- they believe the most significant piece of technology of the decade.
- that all games suffer from such problems as tree bases with only 6 sides
- they believe their method is very different to any 3D method that has been invented so far
- they have to compare their low-level algorithm with high-level web-search engines
Surely if their system was as similar to google's search engine as he implies, you could run google on a single PC with a big enough RAID, or you need a shitload of servers and MapReduce (or Hadoop) to make it work ;-)
Does anyone remember of the "infinite polygon engine" claims a few years ago.
http://archive.kontek.net/republic.strategyplanet.gamespy.com/totality3.shtml
Man, how can something be infinite if it has to work on a finite machine, it is beyond me.
http://archive.kontek.net/republic.strategyplanet.gamespy.com/totality3.shtml
Man, how can something be infinite if it has to work on a finite machine, it is beyond me.
that is funny shit - the guy is the most patronising speaker outside of children's tv.
Macaw: And to top it off, the value in a game doesn't actually lie in having hyperdetailed façades on ancient Mayan temples, or whatever those weird gray clumbs of buildings are supposed to be.
i also instantly hate anyone who pronounces data as "darta".
That tkarena link has some comedy comments from all involved:
This is typical of the industry in general. People who come up with cool new technology that might revolutionize the industry get trampled on by THE MAN.
WEATHER CAPITALISM PREVAIL'S OR KNOT, THIS TECHNOLOGY WILL RULE ALL!!!
This message brought to you by my father.
Quote:
I expected some one to point out that all the animals on our pyramid are the same,
and whilst hand making 100,000 different animals sounds like fun I’m busy this week.
So let me try and convince you that the system is not smoke and mirrors, so that I can spend my week
on other things.
*snip*
Weather the animals where all the same or weather they where all different does not effect the amount of calculations that would have to independently done.
Kindest Regards
Bruce Dell
Quote:
By Tom Dell:
Hi I am Bruce Dell’s father and have been fascinated to read the comments written by so many, *snip*
With that the chairman told him that they were far too busy to look at something from a small entity. This company could have licensed to the world. In its original form Unlimited Detail would have needed hardware. It is only now that it is able to be run on software alone that interest is stirring.
This is typical of the industry in general. People who come up with cool new technology that might revolutionize the industry get trampled on by THE MAN.
WEATHER CAPITALISM PREVAIL'S OR KNOT, THIS TECHNOLOGY WILL RULE ALL!!!
This message brought to you by my father.
Yes, and the animals being the same does not at all help keeping identical sub-trees in hot-cache at all. IT'S JUST BECAUSE HE'S LAZY, IT WOULD BE IMPOSSIBLE TO DO SOMETHING SIMPLE LIKE ROTATING THEM!
You're all non-believers!
Men herregud, han har jo ingen tøj på!
I know how he did it! He used one of those "revolutionary" packer algorythms that can pack any amount of data down to, say, 2kb, then he only has to search that small buffer no matter how big the actual scene data!
Sorry, I mean darta ;-)
Sorry, I mean darta ;-)
He could do something not too complicated (like a procedural landscape, with some procedural trees) and add some simple animated animals (he already has the models, some simple animation would be easy enough). That would be enough to prove his point I think. Until that happens (or there's a proper explanation) I'll go back to wishing it was true but believing it isn't :)
I don't think what the guy has done there is all that bad though - for a few coders working on some tech stuff, you have to expect coder art. And if I'd invented that, I surely wouldn't explain how it's done - you're just inviting people to fill in the blanks and write their own version instead of licensing it from you. Patents are horribly complex and expensive unfortunately.
I don't think what the guy has done there is all that bad though - for a few coders working on some tech stuff, you have to expect coder art. And if I'd invented that, I surely wouldn't explain how it's done - you're just inviting people to fill in the blanks and write their own version instead of licensing it from you. Patents are horribly complex and expensive unfortunately.
it just seems like this classic case of a guy working on his own on the next big thing in his basement for 10 years, and not noticing that the outside world has passed him by. and not being too thrilled about it when he finds out.
It actually reminds me most of this story. That guy actually got a venture capitalist interested, then died under mysterious circumstances taking all his knowledge with him in the grave.
IN THE GRAVE.
It's actually quite a neat story, if you have some reading time.
IN THE GRAVE.
It's actually quite a neat story, if you have some reading time.
Now and again that guy in his basement comes up with something stunning though. I reckon smash is very likely right (it looks no different from gigavoxels and the rest, and there's no evidence of it solving any of the drawbacks of voxels/raytracing etc., plus those big companies likely rejected this for a good reason rather than the lame excuse offered), but it's also easy to knock something like this when you don't know the full story.
I'd like to point out that it'd make an awesome demo-part, though! My only problem with it, is that it's presented as the largest break-through in 3 dimensional technology since computer graphics began. And I'm not thinking of that one phrase where he actually says that, I'm thinking of the entire tone of the video. I mean... it's cool tech, but it's certainly not a the largest break-through yet.
I DO believe that sparse voxel-trees will at some point in the not-too-distant future become a popular rendering technique, though. And I think there's a lot of potential for innovation in such technology moving forward (animations are somewhat tricky, for one). But I'm not convinced by this video that this is anything bigger than just another static voxel-raytracer.
I DO believe that sparse voxel-trees will at some point in the not-too-distant future become a popular rendering technique, though. And I think there's a lot of potential for innovation in such technology moving forward (animations are somewhat tricky, for one). But I'm not convinced by this video that this is anything bigger than just another static voxel-raytracer.
now that's an interesting story psonice... All movies ever made in just 1 kb!
It reminds me of when I was very young (probably around 12-13) and I thought I had discovered the ultimate sorting algorithm. It went like this:
you had an array H[0..65536], filled with zeros. For every number X (0-65535) you got through scanf you did H[X]++;
Then you looped through H and printed all non-zero elements as many times as indicated. SORTED!
The other story was of an ultra efficient compression algorithm me (or somebody else, can't remember) did back then: it was only a joke, it would create a random stream of data 20% smaller than your original file. The catch was this: the "decompressor".exe was also compressed as such.
People in my local bbs (nobody noticed that critical issue) would start compressing things down only to come back after a couple of weeks demanding a way to decompress the decompressor itself...
It reminds me of when I was very young (probably around 12-13) and I thought I had discovered the ultimate sorting algorithm. It went like this:
you had an array H[0..65536], filled with zeros. For every number X (0-65535) you got through scanf you did H[X]++;
Then you looped through H and printed all non-zero elements as many times as indicated. SORTED!
The other story was of an ultra efficient compression algorithm me (or somebody else, can't remember) did back then: it was only a joke, it would create a random stream of data 20% smaller than your original file. The catch was this: the "decompressor".exe was also compressed as such.
People in my local bbs (nobody noticed that critical issue) would start compressing things down only to come back after a couple of weeks demanding a way to decompress the decompressor itself...
That was sagacity's link. Interesting story though.
A lot of stuff like this is clearly impossible (like the dutch guy's compression scheme). Then again, I've come across a few things recently where I've thought (and been advised by people much more knowledgeable too) that something is totally impossible, only to come across some research paper showing it's possible and how it works. It's shaken my belief in such things enough that I wouldn't write off the dutch guy even :)
A lot of stuff like this is clearly impossible (like the dutch guy's compression scheme). Then again, I've come across a few things recently where I've thought (and been advised by people much more knowledgeable too) that something is totally impossible, only to come across some research paper showing it's possible and how it works. It's shaken my belief in such things enough that I wouldn't write off the dutch guy even :)
iq : you can bore me with links if you want i don't mind :)
Also,
Quote:
the system isn’t ray tracing at all or anything like ray tracing. Ray tracing uses up lots of nasty multiplication and divide operators and so isn’t very fast or friendly.
Unlimited Detail is a sorting algorithm that retrieves only the 3d atoms (I wont say voxels any more it seems that word doesn’t have the prestige in the games industry that it enjoys in medicine and the sciences) that are needed, exactly one for each pixel on the screen, it displays them using a very different procedure from individual 3d to 2d conversion, instead we use a mass 3d to 2d conversion that shares the common elements of the 2d positions of all the dots combined.
so maybe it's more than raytracing after all, but oh well...maybe we'll never know!
Quote:
Q: How long have you been working on Unlimited Detail?
A: Unlimited Detail has been around for four years but in fact taken thirteen years to develop. There are many problems to overcome when you are going to completely rewrite the 3D system from the ground up! It’s only now about to move in to the commercial scene because it’s finally found a way to run fast enough.
His linkedin profile says he's been CEO since 1995. By my research, he founded the company when he was 12 years old. His website is registered to a caravan rental company in Australia. Also this technology, as everyone has said, is old crap that suffers the same issues as every other voxel technology out there.
Atomontage are doing something similar. Their tech looks more credible but still suffers the same issues.
In the youtube video he compares his algo with doing a search in google for every pixel. Maybe he somehow adapted the google page rank algo to 3d search ?
Ahhh nadreamia. Funny when insanity or arrogance or both compels people to produce this shit.
Yeah. Nadreamia. That was something.
"Louis, I think this is the beginning of a beautiful friendship."
