unlimited detail update
polycounter lvl 18
Hey
Didn't find the old thread on this, except for a closed one.
Unlimited detail/euclideon has updated their site.
I don't know if it's worthy it's own thread but here goes..
http://www.euclideon.com/home.html
Didn't find the old thread on this, except for a closed one.
Unlimited detail/euclideon has updated their site.
I don't know if it's worthy it's own thread but here goes..
http://www.euclideon.com/home.html
Replies
that mega mesh thing looks more realistic and promising. Maybe these guys need to attract some game developer to actually make something cool with the tech instead of the crappy programmer art vids they have
Well, they're not making their own shaders or anything like that... their main focus is the geometry using point clouds at the moment.
Once it is perfected it'll make that mega mesh technology look obsolete.
Probably, I think myself and 99.9% of all other industry artists will continue using polygons for the forseeable future. Afterall we are approaching real time tesselation, and polygons have never been that big of a system hogger... it's always shaders.
Animating and Colliding with point cloud data seem like major issues that they will need to solve for games and that they haven't addressed yet.
Collision could most likely be solved fairly easily. Animating point clouds is possible, but even if its too expensive or whatever, most of the game environments are static, or rigid bodies...so a hybrid engine would be a good start.
Personally I can't wait for this stuff to mature...
Throwing in a zbrush model straight into the game engine, skipping low poly meshes, lods, baking normalmaps etc is pretty much the best thing that could happen to prop development heh. Better results in a shorter amount of time.
a lot of times low poly meshes are used for collision, because the denser the collision mesh the slower the collision check. They'll have to figure out a fast way to do it using millions of points. I'm sure they'll crack it, but until they show it I don't assume its in the bag.
that thing you just described about the zbrush model straight into game though.. that's what the mega mesh thing is.. and it works now
Unlimited Detail is taking too long, and mega mesh works now.... which is also a reason why I don't see unlimited detail taking off anytime soon
Have you even read up on unlimited detail technology? You can make high poly meshes that are instantly converted to point clouds, you don't model directly with the point clouds in the first place.
Additionally, their entire team is compromised of programmers. They have no artists at this point.
I've actually missed the mega mesh thing...gotta look into that now, cheers!
Still hoping for a point/voxel based system but it'll take a while before we can fully use the advantages of it so something like mega mesh looks like a nice alternative for now.
Am I missing something?
me too?
Do you work for these guys or something? What basis do you have to make such claims?
http://www.youtube.com/user/Quipster99#p/u/5/l3Sw3dnu8q8
I have read up on it. I know exactly what it is. My original statement still stands, you only misunderstood me. The point of my statement was to assert the opinion that I think the game industry will continue to develop engines around the use of polygons NOT point-cloud systems.
If they want to be taken seriously by any facet of the game-dev community they will have to show that the tech they are developing works with current systems, and provides better and more efficient results.
If they don't have animation, lighting, materials, and the pletora of other issues in game development figured out their entire system is a moot point.
I honestly don't give two shits that their entire team is made up of programmers either... contract a 3-D artist to make a decent piece of art using their system. If it's really as simple as porting over a polygon model from z-brush and converting it I see no reason why you can't have someone create a nice model for $1,000 or less to use as your show piece.
I'd have to imagine it would be helpful in finding funding.
Right now, they haven't come close to proving that this technology will be useful in creating games and I for one will remain skeptical until they do.
Other industries? Sure I can see some application, but not here... not yet.
Absolutely correct.
EDIT:
That cars in the videos are polygonal models, maybe the fusion of pointclouds and polygons is the future...
seems that nobody was noticing this animation test that was uploaded nearly a year ago..?
so animations are possible with this technology but without a beta or something else it will still stuck in the "well ok this could be a great thing but hey there ar other things I am able to look at it right now" section.
Yeah, saw that one before, it didn't sell itself in any way.
It makes it highly suspicious of this being a technology with several disadvantages, but the creator wants to sell it as a highly competitive one.
Megamesh is a proper implementation, does it without breaking any of the good game-art conventions set down, and integrates normally with every other way of doing it.
This is how I visualize the unlimited detail guy:
[ame]
Isn't point cloud data a royal pain in the hole to shade?
if it didn't look like an render/animation from the 90's I'm sure we wouldn't be doubting it this much, but so far we haven't seen anything amazing.
[ame]
Eld I can´t argue with you about the "Unlimited detail" thing I agree.
I might change my mind about that unlimited detail technology if there would be a public version to test.
thats all. if there is nothing to test or a gasp cool graphics (with some ai etc) movie I will and have to agree that it is useless.
That video made me a bit sad inside :P
http://www.youtube.com/watch?v=r7HmAPsCj-g
We have polygons that work perfectly fine... but unlimited detail can potentially be ANYTHING! It can even potentially work perfectly fine!
and haha eld... that's hilarious. He invented a game we played one time.
When my friends schufflepuck table broke down, (and he lost one paddle). So we played with our forarms, and used a tennis ball.
We would've much preferred to have played Schufflepuck.