In the battle for cutting edge technology I think Unity should acquire or merge with Euclideon…Yes I know every bit of promotion for the technology has been as kitsch-y as a “Now’nLater” orange tone polyester suit, but the technology does seem to keep moving forward despite the criticism after several years. My only problem is that every demo I’ve seen has flat lighting, which makes me believe dynamic lighting, dynamic occlusion, self-shadowing, dynamics etc. etc. could just present huge technological obstacles for “Unlimited Detail/Unlimited Engine” which is HUGE. However, even having mentioned all of those potential limitations having an efficient voxel rendering solution built into Unity would be useful to a lot of different developers for a lot of different uses. What do you guys think?
That tech has been around for years, and no one (software developers, hardware developers, publishers, etc) has shown support for it. It wouldn’t make any sense for Unity to pick up tech that has no support and, therefore, no use.
Euclidieon are cranks. There are a few cranks that frequent these forums but thats about as Cranky as Unity gets. Can’t see it happening.
I don’t really see them as being cranks. They – mainly the CEO – seem a bit arrogant, but they seem to have a pretty good attitude in general.
@OP There are reasons this technology has not been adopted by engines yet, even though, as khanstruct said, similar tech has been around for years. It isn’t ready yet. Even Euclideon has stated that it isn’t ready.
No they are cranks. Nothing to “see or not see”. What they claim they can do is cranky.
I don’t find them too irritable. They’ve even let reporters into their offices to take a look at realtime demos and experience the technology first-hand.
I see where you’re coming from now. They are very defensive when it comes to accusations and are quite quick to tell their side of the story.
The only thing I look at with this is the path of features like light and shadow baking, self shadowing and how for a couple of years many people around here were clamoring for robust features in those categories because UDK and Crytek and even Project Offset (when it was still alive) had those things and it was a huge sigh of relief when Unity licensed Beast for a lot of that stuff, but Unity probably paid insane money for that when had they gone to the Global Ilumination much earlier, merging with that company and licensing Beast for UDK, Crytek etc. I don’t know that “Unlimited X” is ready as a general purpose game rendering system, but I do think there are some other areas it could be useful for and it’s far beyond many of the voxel experiments floating around. It always makes sense to evaluate something as potentially licensable earlier and become the licensor vs. paying through the nose to be a licensee.
Euclidion is clearly full of shit when it comes to near everything, there is no… no… noooooo way that if they really had this tech they wouldn’t sell out if it was legit Microsoft Sony and Nintendo would all want it as well as the big developers like Crytech, EA, Activision etc… The only possible answers to this is that they are full of shit or as i have heard they got a large sum of money from the Australian government and maybe this prevents them from selling out, which would include Unity.
Even if the tech turns out to be legit, and ready for sale…
What would Unity do with it?
I doubt it’ll be a plugin replacement for the current renderer - from what I’ve read it’d require a fundamental rewrite of most systems. This means Unity taking on two game engines! TBH looking at the state of Unity, I doubt they’re the best placed to take on even more work - particularly such cutting edge work.
Is it time for another “Let me run Unity!” thread already?
This is just silly because:
- Stop trying to tell a company (that is doing far better than any of its competitors) what to do.
- Anyone that know anything about 3d engines can see they are a bunch of shysters.
- In a far out possibility that they do have an engine that can stream unlimited detail on standard machines there is still no way in hell that Unity would want anything to do with it. They’re goal is to make an engine that works well, is easy to pick up and learn, and make it accessible on as many platforms as possible. They want simple. Not a reinvention of the wheel.
Merging or buying out doesn’t make sense for any of the companies involved. Unity is making a game engine and Euclideon are making their own.
Now having said that I think the people that jumped the gun with Notch’s tweet/blog post should really look into what the engine actually does.
Quite a few videos have been posted, they were at GDC : (http://www.youtube.com/watch?feature=endscreen&NR=1&v=_5hg9VfbyYg)
and the technology is very real. It’s basically a point cloud RENDERER.
Of course jumping from that to a fully functional GAME engine needs work, but as they have already demonstrated mixing it with tech already available is possible and is the obvious path (characters can be classic meshes etc). Of course there are things that won’t be easy to make, but that is true for every tech.
Some links on the use of point clouds in the offline rendering arena:
http://renderman.pixar.com/view/point-clouds
Actually the premise is not far fetched and the idea is pretty compatible with how pixel shaders currently work. The main difference is that instead of doing calculations per pixel based on surfaces,materials, lighting etc, we just have a “which point from the point cloud should we render” model.
As for the arrogance accusations, I think you are confusing his accent with arrogance or maybe some people think that because they play minecraft they are somehow connected to notch and his success.
Also it seems that they might be taking the tech in a different direction, since on it’s own it isn’t easily interactive which is the main focus of gaming
http://www.euclideon.com/ . There are a variety of projects currently in progress that use photos to recreate point base models of real locations
Anyway do some searches for “point clouds”, photofly and such.
It looks like they’ve got a good solution for rendering large pointcloud datasets, but I don’t suppose there’s a lot of games out there that would need to render large pointcloud datasets. Most models aren’t authored as pointclouds - it seems best suited for rendering realworld scanned data - something that currently doesn’t turn up a lot in games.
Well they created a tool to convert from polygon data to point cloud data, which isn’t really difficult actually.
But you’ve got to wonder why you’d want to do that - despite their compression claims, if you’re converting from a polygonal representation it’s hard to imagine a pointcloud coming out smaller.
Whoa, linky?
it’s not a matter of smaller, but consistency I guess.
I’ve already posted it.
That tiny demo? I’ll wait for something a little more substantial.