Company invents technology 100k times better than existing ones

they talk about scanning in real world objects, which is an extremely tedious thing to do and I don't see how that will help game designers.

@Anwar said:

they talk about scanning in real world objects, which is an extremely tedious thing to do and I don't see how that will help game designers.

Modeling and texturing is also extremely tedius. Scanning real would objects is likely way faster and produces higher quality, at least in theory.

I remember seeing this video a few years ago. Was interesting.

Seems pretty cool would rather see it from a developers point of view though, also they should hire someone else to narrate that video though...

Edit: also come on folks, technology company showing off their high detail tech doesn't upload in 720p or 1080p? What.

Too bad the method will never catch on because it alienates 75% of the market, as it exists. Will probably take at least 25 years to actually be viable.

That tech seems well and good in a static world but there's probably a reason why it's not implemented in video games.

I would guess that the computation power requirements to have dynamic lighting and physics in an atomic based environment would be nightmarishly high. Sure, you have a camera panning through their "real" island. But if you unleash Nathan Drake to mess things up at your command, you're probably looking at something that consumer electronics cannot do yet.

@Pinworm45:   It depends, I think. I 3d model professionally for architectural renderings and while scanning objects would make certain aspects of it easier, if you have a scene littered with objects that have too many surfaces, polygonals, etc it's going to destroy your machine. There's a technical trade off in terms of what's worth super-realistically modelling.   That said, I'm mostly bothered by their fucking logo. Euclidean? And then a picture of a Doric column framed by two seahorses? What?

I saw this some years ago. If this was real or effective it would've made headlines and blown up like crazy. So fuck this.

Pretty big, if true. Hell, they answered my question of "isn't the application limited by real-world constraints". Yay for that.

@Haru48 said:

Too bad the method will never catch on because it alienates 75% of the market, as it exists. Will probably take at least 25 years to actually be viable.

If you don't know jack, why do you spread misinformation? How exactly does it alienate anyone?

They even said in the video your polygonal models from 3DsMax / Maya can be converted to this format right in the program, so it's business as usual.

If you know ANYTHING about creating 3D art than you know a lot of companies have moved on to Mudbox and Zbrush which allow 3D artists more freedom to sculpt at a much higher polygonal count than what they actually put into a game.

3D artists are able to model at higher levels to evoke a higher detail through mudbox and zbrush, then export it to a lower poly count. They also export normal maps and displacement maps off this higher polygonal model and use them on lower polygonal models in order to make it appear as though it has much more details than the base-low poly model actually has. Like the video suggests, it's all eye trickery, witch-craft and voodoo with today's tech and there would be NO POLYGONAL BUDGET with this tech.

They would be able to sculpt in Mudbox or Zbrush and just convert their high polygonal models to this atom based structure. Less polygonal ripping because of bi-pedal envelopes have more topology to deal with, blending joints much more flawlessly.

There is no alienation here.

@Irvandus said:

@Pinworm45 said:

@Anwar said:

they talk about scanning in real world objects, which is an extremely tedious thing to do and I don't see how that will help game designers.

Modeling and texturing is also extremely tedius. Scanning real would objects is likely way faster and produces higher quality, at least in theory.

Where are we going to find a dragon to scan in?

In the video they say how some things can be scanned, while others can't. They call them fiction and non fiction. A game like crysis could scan a lot of objects for non fiction use, while a game like halo or startcraft would have more fiction items made by artists because they simply to not exist in the real world.

I hope this tech does get bigger, because it really is impressive. I guess we will see in a year after they finish.

"...taken the wings of a swan and put them on a tiger..."

SOLD.

Curious. If their claims are true, this coud be totally awesome, though I have a hard time believing them.
 
Also, that guy's accent and manner of speech kind of makes me want to rip his throat out.

Is atom based modeling the same as voxels? I wonder is they could make so you can manipulate the atoms to have realistic destruction, if of course this is real.
 
The part where they showed the ground with what looked like individual particles of dirt really impressed me.
 
The main problem with something like this imo would be animating and having physics simulation, it would require a freaking super computer to do all the math involved.

Polygons! Numbers! Points! Numbers! SCIENCE! Love it.

Impressive if they can actually pull it off. Heard about that company two years back from my uncle of all people (never did ask how he knew them, business thing probably). But while it's a big statement to make, if it's true then boy it could be pretty amazing. Possible option for future consoles.

Is atom based modeling the same as voxels? I wonder is they could make so you can manipulate the atoms to have realistic destruction, if of course this is real.  The part where they showed the ground with what looked like individual particles of dirt really impressed me.  The main problem with something like this imo would be animating and having physics simulation, it would require a freaking super computer to do all the math involved.

That guy seems like he's doing the most fake voice humanly possible.

@Pinworm45 said:

@Anwar said:

they talk about scanning in real world objects, which is an extremely tedious thing to do and I don't see how that will help game designers.

Modeling and texturing is also extremely tedius. Scanning real would objects is likely way faster and produces higher quality, at least in theory.

Where are we going to find a dragon to scan in?

@crusader8463: Where the fuck is your time machine and how can I use it?

@onarum said:

Is atom based modeling the same as voxels? I wonder is they could make so you can manipulate the atoms to have realistic destruction, if of course this is real.  The part where they showed the ground with what looked like individual particles of dirt really impressed me.  The main problem with something like this imo would be animating and having physics simulation, it would require a freaking super computer to do all the math involved.

That's what I was thinking too. Panning around the landscape and viewing it is one thing, but make that shit interact with things like.. a hand grenade and you will have a melt-down on your hands.

@crusader8463 said:

I remember seeing this video a few years ago. Was interesting.

A few years ago? with footage of Bulletstorm? .. How old is that game?! >_<

Either I see something functional like a game demo, benchmark or something, or they can just shut up. My bet is that this will never pan out.

I'm not qualified to say if this is going to replace 3D graphics technology as we know it (which is basically what the company claims), but it's gotta be at least useful in limited cases. We'll know in a few years, i guess.

The tone of that video is needlessly smug though.

A low res demo of a very limited scenario that could be pre-rendered providing no proof whatsoever? It looks really interesting and sounds like it could change video games forever, but I wouldn`t hold my breath for it.

I will be so happy when this kind of tech eventually catches on. The issues with 3D graphics that are mentioned in the video have really started to irk me recently - photos on the ground with bits of grass poking out, low poly objects, model swapping, etc...

With luck, I'll be playing Crysis 9 with this technology when I'm in my 40s. But for the forseeable future? Nope.

With luck, I'll be playing Crysis 9 with this technology when I'm in my 40s. But for the forseeable future? Nope.

Crysis 9 will be awesome :D I can imagine being on a beach and shooting into the sand, actually causing grains of sand to fly off in all different directions. Hm, maybe I should cryogenically freeze myself and wait until it's released...

I would be more inclined to believe this stuff if they explained how it's possible and why it works instead of just telling me it does.

@Pinworm45 said:

@Anwar said:

they talk about scanning in real world objects, which is an extremely tedious thing to do and I don't see how that will help game designers.

Modeling and texturing is also extremely tedius. Scanning real would objects is likely way faster and produces higher quality, at least in theory.

Where are we going to find a dragon to scan in?

I think I read about this years ago :P 
Looks pretty great, could be too good to be true though.

I saw almost this exact thing about 2 years ago and I basically call bull shit again. If this technology actually was as amazing and revolutionary as this weirdly accented guy says it is, it would have already become an industry standard by now but clearly no one is using it or talking about it. So if it actually is real than that's awesome, but until game companies actually start using it then I dont really care.

@Irvandus said:

@Pinworm45 said:

@Anwar said:

they talk about scanning in real world objects, which is an extremely tedious thing to do and I don't see how that will help game designers.

Modeling and texturing is also extremely tedius. Scanning real would objects is likely way faster and produces higher quality, at least in theory.

Where are we going to find a dragon to scan in?

Special effects studios use clay models.

@drewbert: Thats only because WETA keeps all of the real dragons hidden in New Zealand

That is absolutely amazing technology.

I still can't decide if this is cool until I see what a game company does with it. Hand it over to Naughty Dog and lets see what they can make.

I hate his voice, very smug sounding but i like his ideas. if this pans out could be good

Hurry up and standardize this.

This looks fantastic, but think of the system requirements that you'd need to play this type of game.

First you create elaborate game world complete with characters using this tech. Then you'll have to have AI for the non playable characters, some physics for the game world and perhaps realistic weather effects.

On screen fps would have to be at least 30-35 to make it playable. There's simply not enough computing power available to run a game like this.

From John CArmack's Twitter: "Re Euclideon, no chance of a game on current gen systems, but maybe several years from now. Production issues will be challenging."...  if i am willing to listen to anybody saying anything about computer graphics is John Carmack.

Grains of dirt? More like bits of gravel, clearly the scale was completely different.

This doesn't show shit. If they want to impress people, they need to show them that that wasn't just an average engine. Nothing about that rock impressed me. It looked like a lot of texture and normal mapping, and not a lot of "omg real". If they want to impress, they need to show the models, they need to move some dynamic lighting around, have some actual effects going on. This is no good if all the engine can manage is simple lighting and models. Stop running it in truebright mode and it'll probably tank. And again, why should we believe that there was an unlimited amount of detail involved? All we were shown that even suggested that was the "grains of dirt" which anyone could see was a bunch of brown rocks. Dirt particles are quite small, much smaller than that. If it was real, they would be giving hands on with this or something, not hiding on youtube. I doubt it's as impressive as they claim. Crysis 2 looks good because of all the post processing and the effects on top of relatively good assets, not because it was pure assets. Assets are, if anything, less important than the method used to render things. If I render something in blender with shadow stencils and no reflections, bloom, or motion blur, it won't look terribly real. But even if it's a simple Monkey mesh bouncing around with soft shadows and some AO, motion blur, maybe a bit of bloom if a light reflects off of it, then it looks a great deal more convincing. Even if both renders use the same number of polygons.

Consider me unimpressed and unconvinced.