Atomotage Working to Replace Polygons With Voxels

R

rgMekanic

[H]ard|News
Joined
May 13, 2013
Messages
6,943
Gamasutra is reporting that Atomontage Inc. is now a thing that exists, and it's core team want to create scalable, interactive 3D environments using voxels instead of polygons. Co-founder of Atomontage, Branislav Siles, has been working on this project since at least 2011, and the company claims that Atomontage coxel tech has been in R&D for over 15 years.

The statement of the tech being in R&D for over 15 years is believable. This isn't the first time that a company has come up trying to bring voxels to the forefront. Back in 2011 we had an exclusive interview with Euclideon, a small company based in Brisbane Australia attempting to do the same thing with its Unlimited Detail Technology. Thanks to cageymaru for the story.


"Running our software, common computers can now process enough voxels to make their traditionally blocky appearance almost vanish, while still being fast enough for VR and AR. We have been able to break through long-standing technical barriers to create more richly detailed simulations at a scale and resolution way beyond what’s previously been thought possible."
 
?u=https%3A%2F%2Fih1.redbubble.net%2Fimage.111120879.2822%2Fflat%2C800x800%2C070%2Cf.u1.jpg
 
Voxels remind me of Wankel engines - they're a unique, elegant and functional piece of engineering that never had enough evolutionary superiority to force natural selection.


P.S. I am releasing the above sentence under the GNU General Public License, and grant everyone the right to use this sentence whenever they want to sound superior to other people who happen to be talking about voxels. However, if I ever catch any of you saying something similar to, "I like XML because the code is so reusable," I will find you and beat you to death.
 
Last edited:
Nice video but boy are they light in the details. “Common computer” can run it. What does that mean? Why not give at least a video card spec for what the video “ran” on?

It’s almost looking like the Euclideon sham all over again. Marketing speak, tech demos, no products.
 
So... kind of like Minecraft?

My first experience with voxels, that I can remember was Master of Orion3. Devs talking about how they would use them for the space combat... hot load of crap that was. But I guess I should have known being as they bragged about 16 unique ship styles and I asked was it 16 or 8 with different color palettes and they had no comment on that
 
Way better than what eucilideon showed. If they can make it work games would look a lot better. But we're not there yet.
 
sfsuphysics said:
So... kind of like Minecraft?

My first experience with voxels, that I can remember was Master of Orion3. Devs talking about how they would use them for the space combat... hot load of crap that was. But I guess I should have known being as they bragged about 16 unique ship styles and I asked was it 16 or 8 with different color palettes and they had no comment on that
Click to expand...

Considneringg the unumbered unique spaceships you could make in master or orioen it seems liek every other iteration ever really cathed the full spirit of MoO

Need a huge bombder.. desing one
need a small bomber desing one
need a missile battery ship desing one
need a qucik striking missile ship design one
need a long stnading tanking/survivle ship desing one
need huge spaceship with tons of firepoweer desing one.
need a swam tiny spacpship desing one.

that was very unique for MoO and still seems to be pretty rare to get that kind of tactical options
 
Every once in a while someone comes out with a new (or old) rendering primitive that's going to change the world.

I wondered what happened to metaballs, quadratic texture mapping, quadrilaterals and Surfels.
 
Euclideon wasn't/isn't a scam. Their tech is legit and it works. They opened up a holographic demo centre here, Holoverse, and I've been there and seen it working myself. It's not super high framerate (though apparently they've optimised it better and it's faster now), but experiencing VR while seeing your own hands in front of you is a bit of a trip. The "False Eden" game was pretty sucky though, they clearly need to get some real game devs involved to make it actually fun. Most of the enemy motion in the game was very primitive straight lines or rotations, nothing complex, so I'm not sure if that's a limitation of their tech or just because they put so little effort into into making the game.
 
  • Like
Reactions: blkt
like this
RIUM+ said:
Euclideon wasn't/isn't a scam. Their tech is legit and it works. They opened up a holographic demo centre here, Holoverse, and I've been there and seen it working myself. It's not super high framerate (though apparently they've optimised it better and it's faster now), but experiencing VR while seeing your own hands in front of you is a bit of a trip. The "False Eden" game was pretty sucky though, they clearly need to get some real game devs involved to make it actually fun. Most of the enemy motion in the game was very primitive straight lines or rotations, nothing complex, so I'm not sure if that's a limitation of their tech or just because they put so little effort into into making the game.
Click to expand...

They're not a scam but hubris is their undoing. They think they can do everything better so try and do it themselves but haven't got the breadth and depth to do it.

The problem with all this stuff is tool and product ecosystems. If people can't work with it then it's just a nice tech demo.

Got to be able to create assets, create scenes and render it all on a heap of target platforms, all whilst having a compelling story over both customer experiences but also development efficiency. That's tough. Look at how much money companies like Autodesk have put into software (not enough if you actually try and use it ) or how much something like Unreal Engine 4 costs to develop.

It takes a brave company to experiment in new areas as it can be an infinite time and money pit. Look at Cloud Imperium and their CryEngine customisation. It's only due to the unique way they were funded that they've been allowed to fuck around so much.

Gonna look into these guys to see if they're closer. I want to believe. Voxels do have enormous potential for VR and real life representation.
 
  • Like
Reactions: blkt
like this
Always excited about something new that might change things - hopefully they'll actually put something out that anybody can play with rather than demos released on youtube?
 
I remember the euclideon video demo. Would be nice to see some actual apps to demo instead of just vids.
 
What makes me excited, is how this applies to VR and lower end PC's
 
Nexus6 said:
First game I ever remember playing based on a voxel raster engine was Comanche: Maximum Overkill.

And judging by those videos, it has the same basic problem that they had in 1992.
Click to expand...
Exactly. There is always a reason why new techniques or technologies don't pick up steam in the mainstream. It would be great to move away from polygons in 3D worlds, but I don't think voxels is going to be the thing that replaces them.
 
Armenius said:
Nexus6 said:
First game I ever remember playing based on a voxel raster engine was Comanche: Maximum Overkill.

And judging by those videos, it has the same basic problem that they had in 1992.
Click to expand...
Exactly. There is always a reason why new techniques or technologies don't pick up steam in the mainstream. It would be great to move away from polygons in 3D worlds, but I don't think voxels is going to be the thing that replaces them.
Click to expand...
Yep. The first thing that ran through my mind was "I recognise that blockiness!!" and then there was that instant sinking feeling of disappointment in my heart. :-(
 
if they function as scalable similar to "Vector images" (look great at 1" pic or 10000' ft stretched) it could be great in the long run in 20 years when we have 32k tvs , 1000 foot 16k billboards etc lol... instead of "Stretching" polygons
 
Seems like we get 1 or 2 of these pie-in-the-sky voxel videos each year and everyone gets fired up for about 15 minutes before they realize we're XX years from it being remotely useful.
 
lol @ mocap guy 2:02 into video looking straight off a sega cd game.

Honestly this looks like the same technology on a slightly better demo reel. There will be applications for it, but it's just not going to replace the current status quo. I could see specific scenarios where this could be a great tool for content, but its no where near what could be considered the next thing. I'm all for better computer graphics and skipping a few generations to get to truly imersive believable worlds, but this isn't the answer, at least not in this decade.
 
Aix. said:
Seems like we get 1 or 2 of these pie-in-the-sky voxel videos each year and everyone gets fired up for about 15 minutes before they realize we're XX years from it being remotely useful.
Click to expand...

I came here to post this. Every few years, they come back out again.

It's not a bad thing at all. I'm all for alternative methods, technologies, etc. It might not fit everything, but it will be useful for something. Similar to ray tracing. Both tend to come up and disappear again in cycles. I've liked a lot of things I've seen with both, but in the in-between time, typical methods keep advancing as well.

Still, cool to see something new with Voxels. I think the first time I saw them was maybe an Amiga demo WAY back, (or maybe a PC demo.) Commanche was close though IIRC.
 
Could be me but i want less realism in games. i live in real life and am tired of it, i play games to escape so let me escape.
 
Devilpup said:
Could be me but i want less realism in games. i live in real life and am tired of it, i play games to escape so let me escape.
Click to expand...

No, not just you. However, in my case, I'll take a mix. Some games I like pretty realistic. Others, not at all. I love things like The Wind Waker for example, play tons of 2D indie games, etc. As long as the artwork is good for the style it's in, I enjoy it. I can get pulled into just about anything if it's good for what it is. There are some things that a bit of realism might go a long way though. Not very important to me, but it's nice to know that the capability is there.
 
they might have an idea but they have no talent nor resources to bring it to life, i think this will die at some point, beside this voxel crap must have some heavy limitation they do not talk about yet, because otherwise ppl would have invested in it.
seriously after 15 years they are still in the proof of concept stage...
 
Well, voxel tech is an established concept. It just has its limitations, and isn't some miracle solution to all graphics applications. It's by no means prevalent, but there are quite a few games (as one example) out there that use this type of tech. Some that even look pretty amazing. Nex Machina for one.
 
  • Like
Reactions: blkt
like this
Didn't nvidia make voxels support in hardware or is that their next card?
 
Galvin said:
Didn't nvidia make voxels support in hardware or is that their next card?
Click to expand...

Good question. I want to say that I remember something like that, but couldn't say at this point. I know that everything that I've seen game-wise used the CPU to do them. (as far as I know/remember anyway) I would think if it was GPU accelerated, it would take a major leap forward.

Once again, this is just from fuzzy memory though. For all I know they already can... :D

Interesting. Just read that Housemarque licensed the Unreal engine. So I guess they're moving away from their voxel engine?
 
Iratus said:
...breadth and depth to do it...
Click to expand...
the timing seems a bit odd, but Forrest fires are circumstantial whereas flint like sparks in entrenched ideology are well timed opportunism ( when they alight ) .

The problem with all this stuff is tool and product ecosystems. If people can't work with it ...
Click to expand...
100% agree...a lot of entrenched business savvy people will notice this like blood in the water , but with the tools and computing power available to us now ( as opposed to when Outcast was released ) the development pipeline should be much shorter , and the benefits to time step integration , collision detection and physics models look very interesting.


It takes a brave company to experiment in new areas as it can be an infinite time and money pit.
Click to expand...
That seems a bit , erm , personal...but it is where the fools venturing into angel free territories that we often find the most legendary stuff ( before it is legendary )

Look at Cloud Imperium and their CryEngine customisation. It's only due to the
Click to expand...
*coughscam*
unique
Click to expand...
*coughscam*
way they were funded that they've been allowed to fuck around so much.
Click to expand...

Gonna look into these guys to see if they're closer. I want to believe. Voxels do have enormous potential for VR and real life representation.
Click to expand...
Consider what kind of engine will be best render(er) for Volumetric displays ( ; http://www.kurzweilai.net/he-princess-leia-project-volumetric-3d-images-that-float-in-thin-air ) . Just a thought.
 
Our 3d printers(Zmorph) came with rendering software that's voxel based. I've used it and a number of other programs for my final files. Not really sure how it can apply to a 'better' experience with gaming. It's been interesting to see in with printing but even then I'm still not completely convinced to it's proclaimed potentials. After awhile it just reminds me of a label and/or video/audio container like matroska. I really got into DivX but eventually just went with the flow of MKV. You're still working with the same details but arranging them in a different way. Different can be better, I'm just not sure if this is.
 
You must log in or register to reply here.
Back
Top