Polygons Might Be Gone Forever!

[psi29a]

The terrain in that video is still very blocky, meaning you can tell the surface is composed of flat triangles. But, I think that may be a product of the procedural generation/sculpting system they are using. In an environment like a single player RPG where the terrain is static and artist-designed and hand built, I think the ability for terrain to be smooth (well, or realistically jagged for rocky terrain, you get my meaning) would be a huge step for the quality of landscape. I mean, could you imagine how much better Morrowind would look if its terrain was like that instead of a collection of large flat triangles?

The reason for that is because they are copying the minecraft approach which lends quite well with how voxels work. The reason they can get away (In terms of so much on screen at a time) is because voxels are super cheap to duplicate which is why they are so often repeated. That was the dead give-away with Euclideon's stuff:

^-- one of their photos... sooo many of the same object

So while it looks detailed and 'amazing', it is heavily duplicated to save on memory. That is it's downside, if you start making unique objects like you have in modern 3d polygone engins, then Euclideon's and other Voxel engines would slow down to a crawl. It is a trade-off in terms of performance / uniqueness (more models).

If you want terrain to be smooth and flowing, you'll not get it from Euclideon's software. You'll still be using polygons, but that is a good thing. The more polygons you have the more natural things look. A ball made of a 100 triangles looks great compared to a 'ball' made of only 20 or 30. Up the number and you get even smoother/round/fluid results which is where everyone is going anyway.

Look for yourself:
http://unlimiteddetailtechnology.com/pictures.html

There is a reason why there are so many repeated objects, and it is not because they are 'lazy'. Quite the reverse, they need to show off in order to get funding so if this is the best they got... sorry, even the amateurs (ones doing it for fun) have this kind of tech for years.

[Rhys]

No-one really knows anything about this thing, just some speculation and lots of people thinking they know exactly the methods used.

There is a 40min interview type thing here, with some much more interesting (live) footage of their demo. Still doesn't really say anything whatsoever.
http://www.youtube.com/watch?v=hxtuZE5p ... re=related

-All the employees and the main guy seem a bit... WTF ...but then often people breaking new ground are WTF (I'm not saying they are or not)
-At one point the guy goes on about contemporary engines using LOD which he says multiple times as "Level of distance"
-You could say they give crap all information about the system so that another team will not be able to create a competing product until Euclideon is at prime time and able to protect itself. Who knows.
-Australian government granted them approx 2 million dollars at one point.

-You could think, If they are a scam they are doing a terrible job at it!

Only Euclideon know's what's going on around this project.

[psi29a]

-Australian government granted them approx 2 million dollars at one point.

-You could think, If they are a scam they are doing a terrible job at it!

Only Euclideon know's what's going on around this project.

Yes, because that worked so well for the tax payers of Rhode Island and 38 Studios...

The guy who said government-funded programs are nothing but trouble, then built a company with $75 million in Rhode Island taxpayer money. The guy who blamed the state that gave him all that cash when he couldn’t pay his bills and laid off his entire staff of 379 employees.

http://money.cnn.com/2012/05/28/technol ... /index.htm
http://www.emergingmarketsoutlook.com/?p=1854

It doesn't have to be a scam... it can also just be a failed business plan. Many startups fail, my own startup company died a horrible death in 2007. If you cannot fund your R&D and/or market what you currently have, "you are going to have a bad time."


To steer this sucker back on course, polygons might not be gone forever but it doesn't mean that it can't be replaced with something else. The reality is that voxels (pount cloud, 3d atoms or whatever) have their place are actually pretty damn cool.

Atomonedge Engine 2010:
http://www.youtube.com/watch?v=Gshc8GMTa1Y
http://www.youtube.com/watch?v=_CCZIBDt1uM
^-- every particle (voxel) of sand... can be pushed aside by a moving vehicle and there is also destructible surfaces and buildings.

[Necrod]

With all said, they could at least hire WeirdSexy to do the narrating on videos. Because this guy is so annoying.

[Rhys]

Yes, because that worked so well for the tax payers of Rhode Island and 38 Studios...

The guy who said government-funded programs are nothing but trouble, then built a company with $75 million in Rhode Island taxpayer money. The guy who blamed the state that gave him all that cash when he couldn’t pay his bills and laid off his entire staff of 379 employees.

http://money.cnn.com/2012/05/28/technol ... /index.htm
http://www.emergingmarketsoutlook.com/?p=1854

It doesn't have to be a scam... it can also just be a failed business plan. Many startups fail, my own startup company died a horrible death in 2007. If you cannot fund your R&D and/or market what you currently have, "you are going to have a bad time."

I'm not saying anything's good or bad about the project, just that no-one really knows anything about it, whilst stating some random points of info we do know about it.

With all said, they could at least hire WeirdSexy to do the narrating on videos. Because this guy is so annoying.

Yeah the guy sucks at communicating.

[psi29a]

With all said, they could at least hire WeirdSexy to do the narrating on videos. Because this guy is so annoying.

Yeah the guy sucks at communicating.

Agreed... it was almost as if he doesn't want to market his company's software.

[Chris]

I'm not saying anything's good or bad about the project, just that no-one really knows anything about it, whilst stating some random points of info we do know about it.

I don't think it's unreasonable to listen to people familiar with the tech, though. Voxels (and it is voxels; he admits it even though he does his best to dance around the term) aren't new, and there are people familiar enough with them to make make educated guesses about what it's doing.

What's most telling though is his defensiveness and inability to actually provide a rebuttal or any sort of clarification for the criticisms laid against him. He also doesn't seem to really know what he's talking about, which isn't good for someone who claims to have been working on it by himself for the first 5 years and only gotten some help in the last 2 to 3.

EDIT:
And by rebuttal, I mean an actual answer that discusses the claims, why people could think it's similar to other stuff and how it's not similar. Instead, his answers were basically "that doesn't look like our stuff" and "we're not using the GPU." The problems with animation are never really addressed, and given the contradictory statements "we've been able to do animations since the beginning" and "it's not ready yet" (while showing old videos of voxel animations, which exhibit the issues surrounding them). The problems with storage space are similarly glossed over, saying "we're not taking up a lot of space, but we're not discussing how we do it yet".

EDIT 2:
He claims it's not ray-tracing, but what he describes practically is just that... for each output pixel, it quickly searches for the pixel/voxel to render. He claims his software is special because adding more voxels to a scene won't slow down rendering, but that's what you'd expect of a proper voxel ray-tracing engine. Removing object LOD (level of detail, dude) and having no object "popping" is also something you'd expect.

People are familiar with doing similar things to what he showed, They're able to discuss how it works, and the drawbacks (which you can even see in the video; such as the repetition that such an engine needs for memory efficiency). He completely fails to discuss the big problems that plague the tech, which are also the biggest criticisms against his stuff.

[pvdk]

Interesting stuff, a bit off-topic but it kinda reminds me of the fuzz around the Sloot Digital Coding System.

[natirips]

How can a missing compiler stop you when it comes to something of such greatness?

Couldn't they just re-write it in another language if they had the source? I mean, a compiler is not what should stop you for something THAT epic (a movie in 8KB).

Thus, it's obviously a hoax or a misunderstanding where a "movie" was an in-game cinematic-like thingy.

[xXShaddowTXx]

Very interesting discussion.

I honestly am very suspicious now after reading all these comments but the $2 million Australian gran is very strange.

Sadly I don't know enough to contribute to this discussion.