Euclideon & Unlimited Detail - Bruce Dell Interview @ [H]

So glad to see this video! You have no idea how long I spent in bulletstorm looking at the ground.

This is the problem I see with this technology and the gaming marketplace. Carmacks recent keynote and interviews relating to rage's development point out the differences in technological advances and the actual game delivery. Getting what you can from current hardware has alot more people working on it then voxelwhathaveyous, because thats where the industry (and money) is, and thats what people want. I'm not going to go buy a look down at your feet, check out those rocks! game. I want 120fps, and a clear objective. 15-25fps on an i7 core at 1024x768 is nowhere near the performance needed for a high selling game. And "unlimited" detail is entirely unecessary. It is totally still limited to what the artists can make of the models! Carmack mentioned recently about it 'uninspired content is uninspired content'. Thats the real challenge to overcome. Content creation, both textural and geometrical, and what you can actually deliver on screen. Theres quite a few points in the video that show the limit of the current content set they have. And its the content itself, and the representation at both speed and interactivity that stand before this engine, and any use it may have.

They have a sound set of great ideas though. Using whatever atomy stuff they do in particular circumstances on the cpu, leaving the gpu to make it look good. But the balance there is going to have to be determined, how much gpu is needed to push decent shaders, shadows and lighting with all the detailed geo data? And its that stuff that makes an engine look great. Basically you have people programming for reality, that can ship products that work in actual real world settings and be great entertainment. Then you have this stuff. Which may or may not be relevant by the time they can release it.

I'm calling this the wankel rotary 3d engine. Sure its got fans but give me a 4-stroke anyday!

It also doesn't help that this supposed ceo is a complete amalgamation of every insufferable insecure aussie nerd trait rolled into one. I'd rather see the 2mil go into the arts to be honest. Totally not bias there in any way, either. Honest.
 
bibz said:
So glad to see this video! You have no idea how long I spent in bulletstorm looking at the ground.

This is the problem I see with this technology and the gaming marketplace. Carmacks recent keynote and interviews relating to rage's development point out the differences in technological advances and the actual game delivery. Getting what you can from current hardware has alot more people working on it then voxelwhathaveyous, because thats where the industry (and money) is, and thats what people want. I'm not going to go buy a look down at your feet, check out those rocks! game. I want 120fps, and a clear objective. 15-25fps on an i7 core at 1024x768 is nowhere near the performance needed for a high selling game. And "unlimited" detail is entirely unecessary. It is totally still limited to what the artists can make of the models! Carmack mentioned recently about it 'uninspired content is uninspired content'. Thats the real challenge to overcome. Content creation, both textural and geometrical, and what you can actually deliver on screen. Theres quite a few points in the video that show the limit of the current content set they have. And its the content itself, and the representation at both speed and interactivity that stand before this engine, and any use it may have.

They have a sound set of great ideas though. Using whatever atomy stuff they do in particular circumstances on the cpu, leaving the gpu to make it look good. But the balance there is going to have to be determined, how much gpu is needed to push decent shaders, shadows and lighting with all the detailed geo data? And its that stuff that makes an engine look great. Basically you have people programming for reality, that can ship products that work in actual real world settings and be great entertainment. Then you have this stuff. Which may or may not be relevant by the time they can release it.

I'm calling this the wankel rotary 3d engine. Sure its got fans but give me a 4-stroke anyday!

It also doesn't help that this supposed ceo is a complete amalgamation of every insufferable insecure aussie nerd trait rolled into one. I'd rather see the 2mil go into the arts to be honest. Totally not bias there in any way, either. Honest.
Click to expand...

You make good points, but I think slightly misplaced. This isn't a look down at rocks game, its not a game at all, its a proof of concept. Its running 1024x768 on a laptop i7 in a demo, who knows what it'll run when its actually ready and optimised running on a high end PC.

Bruce clearly isn't a game designer like Notch or Carmack, you can see that from the video, he never pretends to be a games designer. Hell, he doesn't even know of games that use tesselation... most gamers on this forum could spew out a shortlist of games that use tesselation... that's not who he's trying to be. What is going to be interesting is if they can get this technology to a point where it can be used in games and then see what an actual games development studio can do with it. That's when the "content" will appear. The rocks and leaves and dirt on the ground dont make a game, but they're the building blocks to transporting you into the game world.

I'm just imagining a huge open world game, walking through the forest and it actually looks like a real forest. You can play Crysis and walk through a forest and think "wow this looks awesome", but that's because you're used to seeing other shitty graphics in other games. If you walk through an actual forest you realise there's a long way to go yet. No, I dont play Crysis to look at the ground... but I do notice the ground, I do notice the LOD pop in, I do notice the textures getting progressively blurrier, I do notice the low poly models. I'm not playing it for those things, but that doesn't mean I'm blind to them.

I find it amazing the double standards people seem to have, people were whinging non stop about Crysis 2 graphics, then play another game and say the graphics aren't important.

Personally I'm excited about the concept of walking through a game world which is detailed down to the bumps on an elephant's backside and the individual fibres of bark on a tree are modelled.

I think its an interesting concept for games developers instead of sitting infront of a PC all day modelling a tree or a car or a rock to go outside and look for a tree that already exists to scan it in then modify it to their own desires.

Of course its still just a concept, and that's how people should be looking at it.
 
Is this engine using voxels? Who knows...

Bruce basically suggests that if people want to call them voxels, then they can. Most often he refers to point clouds of atoms versus voxels. Is there a difference? Are those just plain ol' voxels used in Unlimited Detail? We don't know yet. Assuming it is, is just that-an assumption.

A totally new way of doing things is very possible. All the mathematics presented here and strewn about the net concerning Unlimited Detail is merely speculation using what has come before. Methods that have been shown to have major pitfalls and trappings that make them unsuitable for this or that reason. However, if this is something new, all that math shows its basis in guesswork.

Concerning SVOs... Sparse Voxel Octree's are a partitioning structure for geometric data sometimes used with ray-casting, traversing the tree is essentially a search algorithm to figure out what pixels will be blasted to the screen. Bruce suggests that his search algorithm is something completely new... so SVOs may or may not have anything to do with Unlimited Detail's workings. Then again, they are just a partitioning structure for data, so they could be used, but just because they might be doesn't mean they are being used. The detail is in the search algorithm...

And another thing... this is all done in software. And although it is using only diffuse lighting and fairly weak shadowing method. It still looks pretty darn good. If this can be ported to a GPGPU or just taking advantage of a GPU to do pixel shader operations, adding per-pixel lighting, Ambient Occlusion, High Dynamic Range lighting... This engine would be absolutely stunning. Even if it if turns out that current hardware acceleration can’t do much for UD, if it really can reduce importance of geometry constraints, future hardware should be built to support it.

Maybe the whole thing is bunk, but maybe it’s truly something new, and may potentially change the way we do graphics in the future.

All in all, there is a lot of assuming going on. If you want to stand on your assumptions so steadfastly, wow, sure, go right ahead. I feel for you. :(
 
BlueStorm said:
The first "technology" is called "occlusion culling", and the second "technology" is called "depth test". Both are part of modern every-day graphics pipelines *today*.

Sorry guys but I think the reason this is seen as "revolutionary" is that the people who say that have no clue how graphics works today.
Click to expand...

You do realize that "occlusion culling" and "depth testing" have undergone and include many innovative approaches. I mean, BSP and Z-Buffering were amazing new ways of doing occlusion culling in their own times, instead of the ol' painter's algorithm, which were both able to vastly improve performance.

Just saying that it is doing occlusion culling and depth testing, so it must be the same ol' thing.... well, that's pretty naive. Then again, maybe they are using a traditional system for occlusiong culling and depth testing, and maybe they have some other innovation that lets them achieve what they have. We'll find out eventually, but it isn't worth our time nor does it offer a functional use to make empty claims about what they are doing.

They showed us what they got so far, if they can improve upon it... great. If they can't make good on their claims... so what? They aren't asking you to put your money where your mouth is.
 
Last edited:
There's a short new third party-made Euclideon video showing some imported laser-scanned assets within Euclideon's engine editor, and a basic shadowing demonstration. These were captured from the GDC event held in Europe this past week which Bruce Dell (CEO of Euclideon) attended.

http://www.youtube.com/watch?v=_5hg9VfbyYg
 
HardLuck said:
There's a short new third party-made Euclideon video showing some imported laser-scanned assets within Euclideon's engine editor, and a basic shadowing demonstration. These were captured from the GDC event held in Europe this past week which Bruce Dell (CEO of Euclideon) attended.

http://www.youtube.com/watch?v=_5hg9VfbyYg
Click to expand...

Was the tech demo mentioned on this video being showcased on that same laptop from the [H] article?
 
HardLuck said:
There's a short new third party-made Euclideon video showing some imported laser-scanned assets within Euclideon's engine editor, and a basic shadowing demonstration. These were captured from the GDC event held in Europe this past week which Bruce Dell (CEO of Euclideon) attended.

http://www.youtube.com/watch?v=_5hg9VfbyYg
Click to expand...

ffs just show the damn animation, lol... looks pretty rough IMO, maybe in another year :D
 
Youn said:
ffs just show the damn animation, lol... looks pretty rough IMO, maybe in another year :D
Click to expand...

Yeah, its clearly just a demo to show the naysayers that lighting is possible. I dont blame them for not showing more given they aren't a games development studio in any way shape or form, they're developing a technology, not the content to go into that technology.
 
I can't help but think of that saying about something being too good to be true.....
 
darkstarcow said:
I can't help but think of that saying about something being too good to be true.....
Click to expand...

We still don't really know how the technology would work in games but if they can nail animation and make sure it's not too resource demanding I can see this working quite well for many types of games.
 
Great video, much more than I was expecting.
Really opened my eyes to the tech being presented and I really hope this will one day come to fruitation.
 
Llian said:
So, what I am getting from people like yourself, and there are alot, is that what was shown in the video was not real, the lap top was a super pooper computer and that they are lying in general. Because there is no way most of you watched it as it is the same arguments again and again and again and again, even AFTER they say how they deal with certain issues. Seriously, some people are such pessimists.... no hope for a better future in graphics industry because they know more than a guy who is refusing money and saying we are doing this.
Click to expand...

yeah, and that this site would need to be either in on the scam or completely inept since they've gone on record saying it's legit
 
astroboy said:
Well he claims he went to the GPU guys with his tech and got denied.
If his intention was to not reveal anything, why would he even go?
Click to expand...

actually what he said was that he went to a couple companies to seek investors but none of them were GPU manufacturers. He went on to say that he wsn't denied based on the tech. quit trying to make straw men
 
I don't know why you guys are bashing this guy and his company. As a gamer and an IT soon-to-be graduate, I am literally shocked at how good it looks. If nobody told me, I would not have guessed that the demo was in 1024x768. I would have guessed a much, much higher 4:3 resolution, because it looked very crisp. I can't wait til they show us the GPU version, or just another video of their progress.

I see this tech, I see potential. Lots and lots of it. Especially when they tap into the GPU's vastly superior processing power. However I am not sure if the GPU is actually any good at doing searching algorithms. Hell, I don't care. Just make it multi-threaded. I think that will give a major speed boost, as searching algorithms IMO are good threading candidates.

I am very shocked at how he can manage such a feat all by himself, before founding the company. Just wow.
 
trandoanhung1991 said:
I don't know why you guys are bashing this guy and his company. As a gamer and an IT soon-to-be graduate, I am literally shocked at how good it looks. If nobody told me, I would not have guessed that the demo was in 1024x768. I would have guessed a much, much higher 4:3 resolution, because it looked very crisp. I can't wait til they show us the GPU version, or just another video of their progress.

I see this tech, I see potential. Lots and lots of it. Especially when they tap into the GPU's vastly superior processing power. However I am not sure if the GPU is actually any good at doing searching algorithms. Hell, I don't care. Just make it multi-threaded. I think that will give a major speed boost, as searching algorithms IMO are good threading candidates.

I am very shocked at how he can manage such a feat all by himself, before founding the company. Just wow.
Click to expand...
Hi Bruce, nice new account
 
Very Impressive !!!

All this kind of reminds me an oil business where the giant companies where slowing down the development of new engine types based on other fuel sources to maximize their profits from the oil production.

I think that polygon based graphics has much more limitations than this one. Everybody should should stop mourning and be happy that there is someone out there with a different approach to the subject maybe it will work out for the best of all of us.

There is nothing worse than monotheistic way of thinking. Each year there are produced thousands of programmers around the world and all of them are trained in the same stuff, so simply speaking all of them know the same and they will be thinking in a very similar way, so this is actually bad thing as it will be very hard for them to come up with something unique and maybe better than what is already on the market.

So stop bashing someone for non linear thinking!

Good luck with the project Euclidean!
 
tx00824 said:
So stop bashing someone for non linear thinking!
Click to expand...

No one is bashing them for "non linear thinking". Everyone is bashing them because they didn't explain how they solved the problems of memory usage (no, "compression" is not an acceptable answer), they din't show the more problematic and complex issues with the engine they are developing (animation, physics - and again, "it's not done yet" is not an answer, because it is going to be a obvious limitation for their engine).
 
No one apart from them really knows how their engine works, there are some hints from them it is similar to voxels (obviously heavily modified that it can handle so tiny particles), so let's wait to see what it really can do.

In my experience I learned that there are no impossibilities in computers world. They will improve it in time so it meets peoples expectation. It will be hard task as they are only a few people against huge industry. (I remember years ago reading that internet speed through phone line can't be faster that 128bits/s [ISDN] as it is physically "impossible") As we can see now it was not true).

They are very secretive as Bruce Dell said himself and this is what people can't stand it makes everybody to think it is a scam OR he doesn't want to share any details of his work with anyone until he has is sure it wont be stolen from hm, so maybe he will sell it to the bigger player when it is ready.
 
Dikk R. Pimmel

By saying no impossibilities I mean that even though at some point in time some problems let's say with technology limitations seem impossible to overcome, are at most of the time solved in the future, either near or many ears ahead in time.

I agree that there are frustrating situations in computer world, but this is to be found in any industry.
 
For lack of a better word: Scam. Way too many things about the video that reek of dishonesty or simply withholding of all details. Compare what this guy says to a John Carmack keynote speech and you'll see this guy sounds like a walking infomercial. Probably fishing for investor money to do who knows what with.
 
Last edited:
PornoSatan said:
For lack of a better word: Scam. Way too many things about the video that reek of dishonesty or simply withholding of all details. Compare what this guy says to a John Carmack keynote speech and you'll see this guy sounds like a walking infomercial. Probably fishing for investor money to do who knows what with.
Click to expand...

He only sounds like that because of the way he talks and honestly I dont even think he means to talk the way he does, as he talks the same way regardless of whether he's being asked questions or presenting something. His accent isn't even super Australian, he just has a voice like a permenant advertising :p So its kinda like saying "I dont trust what that guy is saying because he sounds far too Scottish and I dont trust Scottish people" ;)
 
Tudz said:
He only sounds like that because of the way he talks and honestly I dont even think he means to talk the way he does, as he talks the same way regardless of whether he's being asked questions or presenting something. His accent isn't even super Australian, he just has a voice like a permenant advertising :p So its kinda like saying "I dont trust what that guy is saying because he sounds far too Scottish and I dont trust Scottish people" ;)
Click to expand...

LMAFO!!!!! But who knows...hopefully this works but really only time will tell..but games look amazing as it is right now very happy with progress being made
 
Their website still shows the same crappy ugly demo and hasn't had any posts since august last year... Same as the facebook and youtube page. Still no SDK or anything.

Guessing they are still working on adding all the stuff they were "working on" (you know the flaws in voxel type systems everyone has known about since forever which they didn't have any answer for like reflections and animations). So it looks at this point like this was a big scam?

That guy is a huge slice of annoying. I'm no expert at pr, but surely having someone less...irritating would be better.
 
Youn said:
Anyone know anything new about this? lol
Click to expand...

I think the PR man behind this is waiting on quantum computing power for Vortexls can be a reality. He did show a really nice demo of 5 models repeated over and over and over again though. ... ... oh and a model of deformation of Vortexls from years ago that seemed to be working at 1980s resolutions. So if you take that technology, and increase your computing power say... 100 fold, you could make a really high quality game out of it. So I'd say 8nm technology or quantum computing based of electron spin should make this a reality. Whichever comes first.

Just... don't mention to bruce dell what Ray Tracing might look like by then. He makes him turn into this guy here:

http://www.youtube.com/watch?v=QeHVYuRnIjY
 
Last edited:
yes plz, news update. This is one topic to keep on the bulletin board for making sure their publicity stunt backfires on them (if not succeed) :D
 
Thanks folks, for keeping up on this. I was expecting to have to post a snarky necro-accusation.
 
Neat. I was just thinking about these guys a few days ago, wondering if they were still around. I remember the last time this came up, the guy said something like, "and now we are going to disappear again, until a later time..."

There's still a part of me inside that says "Where is Phantom?" on this..
 
You must log in or register to reply here.
Back
Top