Unlimited Detail Technology?

category: offtopic [glöplog]
Stunning? I'm playing the original Far Cry nowadays and that looks about the same up close, except it's almost 10 years old now. If anything, that realtime demo made me even less impressed.
added on the 2011-08-12 13:01:17 by Gargaj Gargaj
Far cry is certainly impressive too, and has way more impressive environments, but for detail? Nowhere near, surely. It's running a limited number of polygons and textures for the detail, not extremely detailed voxel objects.

You can't zoom in and see a 3d leaf resting on 3d grains of dirt in far cry. That kind of detail I think is totally stunning. If they can make it work in a real environment like the far cry ones, it'll be utterly mind-blowing.. but I'll believe that when i see it. ;)
added on the 2011-08-12 13:10:18 by psonice psonice
the demo running on his laptop in this video is actually quite cool...
added on the 2011-08-12 13:53:40 by tobé tobé
Blocks of predefined objects, If this is the only way he can get it to work, then I'm not sold. As psonice said, if it can be put into real environments, then yeah, they will have won the internets. But if it's limited to blocks, they'll have a hard job trying to get someone to use it.
Even if it is just blocks, there's a lot that can be done with blocks and a decent artist. And even if it was just pretty much cubes, would you complain about mine craft with graphics like that? Ok, so I'd still be complaining hard about their marketing claims :D
added on the 2011-08-12 14:12:58 by psonice psonice
Nothing is too certain until we get a binary - hopefully with more varying objects and animation. But it sure as hell feels fishy.
added on the 2011-08-12 14:17:18 by msqrt msqrt
would you complain about mine craft with graphics like that?

BB Image
added on the 2011-08-12 14:25:07 by Gargaj Gargaj
He should do a universe where you can fly from galaxy to galaxy, zoom in to planetary systems and where you also can watch the stars/planets in them ;)
added on the 2011-08-12 14:28:03 by seppjo seppjo
...and it looks like a texture pack, and would look even more like a texture pack in motion. Hi-res photo textures aren't impressive these days, seeing that in full 3d sure as hell would be :)
added on the 2011-08-12 14:28:58 by psonice psonice
seppjo: with planets perfectly aligned in a grid? :D
added on the 2011-08-12 14:29:48 by psonice psonice
even more important wirh realistic physics :D
added on the 2011-08-12 14:31:21 by seppjo seppjo
So to reiterate, which one looks better again?
BB Image
added on the 2011-08-12 14:32:14 by Gargaj Gargaj
I love how matter-of-fact-ly you state that. As if you've thoroughly studied their source code, understood all implications and side effects and came to that conclusion in a matter of minutes.
I don't need to, because this company has had ample time to file patent applications and protect their technology and show the world how it works, and yet: they don't. They release half-assed videos on YouTube and brag about the amount of hits they are getting. They make fun of known 3D technology geniuses and make outrageous claims to get press -- press they've used to pat themselves on the back and get grants and financial support to keep developing something that inn all likelyhood will end up as a useless piece of tech that (most likely) shines in one very limited usecase, or not at all.

Why do I know this? Because we've seen this a gazillion times before, and all the pieces fit. If you have brilliant technology? Show it to people, get it peer reviewed. If you don't do that, you don't have shit.

I don't like the ha-I-know-this guys, who have little to no credentials on the topic, prematurely jumping to conclusions, just like notch.
And I don't like cocky people who do whatever they can to get press and yet produce no proof what-so-ever. I guess both of those types of people have equal right to speak up, so.

Fact is, nobody really knows what's up. We can speculate about SVOs and their "search engine" and "atoms" (all those term might be repackaged and slightly misleading), but until they explain themselves a lot clearer and address the obvious challenges, everything is more or less speculation.
Correct. Until they explain themselves everything is speculation. Therefore they should explain themselves, something they deliberately will not do, therefore nothing is proven, therefore I reserve the right to state that they have nothing.

And as it is, they simple have no reason to explain themselves. Their only crime so far is to make videos that make them look unprofessional (and by extension, slightly scammy).
If you mean they look like the Uwe Boll of computer graphics, then yes. Also: they _do_ have a reason to explain themselves if they want to be believed. The fact that they refuse to do so tells me that they have a reason _not_ to explain themselves, and therefore -- again -- have nothing.
added on the 2011-08-12 12:37:25 by tomaes

I guess there's a chance that their claims actually hold, and that this is not just marketing clumsiness, but marketing BRILLIANCE. Get everybody all riled up :-)
The only good marketing is the marketing that results in money earned. Yes, they are getting a lot of press because of their outrageous claims and general dickishness, but if the day comes that they finally show off this tech of theirs and it doesn't hold water, all that press was for nothing. :)

Yeah, no sign of anything non-block based (he even refers to being "between bricks" at one point) still. Even so, the realtime demo was pretty stunning. I'd be happy with that kind of rendering even if it is totally block based with fixed lighting and the characters end up rendered with polygons in the traditional way :)
Sure, and so would a lot of people. However, "Unlimited detail!" isn't the same as "Unlimited detail! (...as long as this VERY LONG AND COMPLICATED LIST OF THINGS YOU CANNOT DO IS FULLFILLED!)" :)
added on the 2011-08-12 14:33:39 by gloom gloom
second one from my point of view
added on the 2011-08-12 14:35:39 by seppjo seppjo
Unlimited details means infinity... Those guys are gods!
Still i wonder, is this discrete or continuous infinity?
added on the 2011-08-12 14:43:07 by baah baah
You should write them a mail and ask. I'm sure that they would be more than happy to tell you all about it.
added on the 2011-08-12 14:46:57 by gloom gloom
is infinity a successor of itself ?
added on the 2011-08-12 14:47:14 by tobé tobé
Well, i've read the thread and already lost enough of my, sadly, finite sparetime!
added on the 2011-08-12 14:51:16 by baah baah
gargaj: show them both at the same scale, then definitely the first one ;)

More interesting would be to show the infinite detail demo and the far cry screenshots, both distant and up close. You'd have a different winner for each case, and in each case one of them would look like yesterday's tech ;)

For the vast majority of games, the distant view is more important than the close up one, so unless they fix this whole tiling issue this is a totally nice product.
added on the 2011-08-12 14:51:17 by psonice psonice
For the vast majority of games, the distant view is more important than the close up one, so unless they fix this whole tiling issue this is a totally nice product.

added on the 2011-08-12 14:57:00 by Gargaj Gargaj
For the vast majority of games, the distant view is more important than the close up one,
I would say it's the exact opposite. Most games look good from far away anyway, and ALL games look crap up close. :)
so unless they fix this whole tiling issue this is a totally nice product.
"so if they fix this whole tiling issue" you mean? Well, since it looks like the tiling-technique is the _basis_ for their whole "technology" then I seriously doubt they'll be able to "fix it" :)
added on the 2011-08-12 14:57:46 by gloom gloom
But tetris looks good up close too :(

And 2D games miss all these detail problems anyway, why doesn't the game industry see this!
added on the 2011-08-12 15:00:24 by msqrt msqrt
Lol! NICHE product :D
added on the 2011-08-12 15:01:15 by psonice psonice
baah: Aleph 0 is an upper bound here. However, it is so encouraging to witness the actuality of one's achieving such a thing out of inventing some hyper-wheel. Leibniz wrote at some point that hobbyists/non-professionals do often make quite novel insights, because they question what others take for granted. cf. this: http://www.pnas.org/content/early/2011/05/10/1008636108.full.pdf
added on the 2011-08-12 15:03:12 by Kais Kais