Biggest Computer Graphics Advance Since 3D?

[Minoza 11]

Guys you like throwing rocks, aye? :D

Just give them some time and see what they come up with. Forget about voice lol, it's the software what's interesting here.

[Dwarden 1124]

jeez, let me find my post on rage3D

it looks like voxels, it smell like voxels, it sounds like voxels

i bet it's some voxel system and (would not be shocked if they reuse some publicly avilable code

(based on the strange stuff they said in interviews sometimes

it sounds like the don't know what are LODs or instancing

or what all can be done with tesselation etc.)

what's sad that these guys unable to even explain to developers and artists wth theirs engine uses and how does it work

(in terms e.g. how prepare texturing) and yet are getting such huge 'hype' and even funds / investors

while these who actually do interesting voxel engines very close to be used for games are struggling w/o funding

btw. aren't there more threads like this? mergeeee

[metalcraze 290]

The major part of their hype is due to people not knowing that voxels are nearly a decade older than 3D acceleration (at least when it comes to games)

Breakthrough lol

And there are reasons nobody uses them extensively for a decade now.

[dmarkwick 261]

Still the same nay-sayers, same increasingly outdated arguments, same various misconceptions. I wonder if most of you even watched that entire video :) (It is rather long though.)

[maddogx 13]

All other threads were locked, I believe, and rightly so. :)

I'm not so sure whether they're really using voxels, but I'm still unconvinced about the animation and transparency issues. The anims they've shown were just some weird blocky shapes that didn't look like "unlimited detail" at all (I don't care if it's "old material"), not to mention that it didn't address the heart of the problem: how do you animate objects that actually consist of extremely high density point cloud data? (Ooops, I meant POINT. CLOUD. DAAHTAAH!)

So why can they only show us old material? How hard can it be to present something simple, like a rock (of which they definitely have some high quality ones) rolling across the ground or something? It doesn't have to be a fully animated dragon, for fuck's sake, just show something recent to prove that it's doable.

Then there's the issue of transparency of course. I dare say that will definitely require some kind of hybrid solution using polygons, otherwise their algorithm will have to start reading multiple points per rendered pixel... goodbye performance.

@DMarkwick: an argument is not outdated if it hasn't been conclusively refuted.

[dmarkwick 261]

The major part of their hype is due to people not knowing that voxels are nearly a decade older than 3D acceleration (at least when it comes to games)

Breakthrough lol

And there are reasons nobody uses them extensively for a decade now.

I think it's a bit disingenuous to suggest that their "hype" is due to people not knowing the age of older technology. What possible difference could that possibly make?

As the video suggests, voxels mean different things to different people. Minecraft is said to use voxels, but it's still essentially a polygon engine, for example. What does voxels mean to you?

[Dwarden 1124]

Still the same nay-sayers, same increasingly outdated arguments, same various misconceptions. I wonder if most of you even watched that entire video :) (It is rather long though.)

ofcourse i watched it .... i wonder do you even know how long the 'UD' theme is 'repeatedly' going around ?

one of best existing Voxel engines is Atomontage http://atomontage.com/

which i consider really 'game breaking' in several aspects

which btw. is also unlimited detail not like 'UD' tries to claim it is not :D

also animated voxels aren't problem for quite some time

tkn6ubbp1SE

Tl6PE_n6zTk

now just look at the date :D of this one and the name :)

lpfaFrazOn4

"Efficient Sparse Voxel Octrees - Global Illumination & Volume Rendering unlimited detail"

btw. if anyone here remembers Delta Force 1,2 and early Comanche games (1-3) then that engine was Voxel engine :)

Edited August 24, 2011 by Dwarden

[dmarkwick 261]

All other threads were locked, I believe, and rightly so. :)

I'm not so sure whether they're really using voxels, but I'm still unconvinced about the animation and transparency issues. The anims they've shown were just some weird blocky shapes that didn't look like "unlimited detail" at all (I don't care if it's "old material"), not to mention that it didn't address the heart of the problem: how do you animate objects that actually consist of extremely high density point cloud data? (Ooops, I meant POINT. CLOUD. DAAHTAAH!)

So why can they only show us old material? How hard can it be to present something simple, like a rock (of which they definitely have some high quality ones) rolling across the ground or something? It doesn't have to be a fully animated dragon, for fuck's sake, just show something recent to prove that it's doable.

I think one of their points is that they don't need to prove anything to anyone. They'll just work on it, then release something, or not. I'm down with that idea.

Then there's the issue of transparency of course. I dare say that will definitely require some kind of hybrid solution using polygons, otherwise their algorithm will have to start reading multiple points per rendered pixel... goodbye performance.

Given that most of us don't understand the current technology used (I should say all of us) then I think the transparency solution will be something else we won't understand. My guess is that the so-far-unused graphics card abilities will help with that.

@DMarkwick: an argument is not outdated if it hasn't been conclusively refuted.

Well in that case I don't know what conclusive refution means for you specifically, as the guy refutes each voxel example individually.

[Dwarden 1124]

updated my previous post with several videos, so far UD not proved by anything they not using VST or similar voxel technique

and compared to AME theirs engine seems subpar in so many aspects the whole discussion is moot

[metalcraze 290]

I think it's a bit disingenuous to suggest that their "hype" is due to people not knowing the age of older technology. What possible difference could that possibly make?

If people knew this technology was very old they wouldn't call it "groundbreaking since 3D"

As the video suggests, voxels mean different things to different people. Minecraft is said to use voxels, but it's still essentially a polygon engine, for example. What does voxels mean to you?

Voxels can't mean different things to different people.

Voxels = Volumetric pixels

Meaning pixels positioned in 3D space. What other meanings can it possibly have?

"Efficient Sparse Voxel Octrees - Global Illumination & Volume Rendering unlimited detail"

Yeah and "2.7 GBs of data" just to render that room that can look the same with polys at a much lower cost is one of the reasons voxels are not used anymore. Maybe in the future...

Edited August 24, 2011 by metalcraze

[dmarkwick 261]

I think I see one particular main complaint: that it's old technology reinvented. That's a rather simplistic view of it IMO and similar to saying that tesselation is old technology because it still uses polygons.

I'm willing to accept their claims based on what I can see so far. Will they solve animation, transparency, instancing? I don't know, but I certainly will not categorically state that they will not solve it :)

I'll wait & see what comes of it, if anything.

---------- Post added at 11:38 AM ---------- Previous post was at 11:37 AM ----------

If people knew this technology was very old they wouldn't call it "groundbreaking since 3D"

... and similarly if people knew how old polygons are, maybe they wouldn't call tesselation new technology?

Voxels can't mean different things to different people.

Voxels = Volumetric pixels

Meaning pixels positioned in 3D space. What other meanings can it possibly have?

You ever played Minecraft? People call that a voxel game, but it's not voxels as in volumetric pixels.

Maybe in the future...

Isn't that the entire purpose of the idea? WIP.

Edited August 24, 2011 by DMarkwick

[metalcraze 290]

But tessellation is an old technology. Hardware accellerated tessellation was already available on Radeon 8500 which is DX8.1 videocard from 2001.

I remember how I was impressed seeing all those smooth and round shapes in UT2003 in an era where you could cut yourself with models in other games.

You ever played Minecraft? People call that a voxel game, but it's not voxels as in volumetric pixels.

Minecraft isn't a voxel game in any meaning.

People not knowing anything about what voxels are exactly the reason why UD gets the hype as I've said. That quote just proves my point.

Edited August 24, 2011 by metalcraze

[dmarkwick 261]

But tessellation is an old technology. Hardware accellerated tessellation was already available on Radeon 8500 which is DX8.1 videocard from 2001.

I think that, for you, there will never ever be new technology, as it only builds on older technology :)

Minecraft isn't a voxel game in any meaning.

People not knowing anything about what voxels are exactly the reason why UD gets the hype as I've said. That quote just proves my point.

It most certainly does not prove your point, unless your point is one about general misconceptions. Minecraft is not a voxel game, you're correct. And I wrote that Minecraft is often called a voxel game, and so Notch is quoted as being an authority on the subject. Can we agree that Notch has no authority on this subject and thus we can safely disregard his input?

[maddogx 13]

I think one of their points is that they don't need to prove anything to anyone. They'll just work on it, then release something, or not. I'm down with that idea.

Really? Well in that case... did you know I've just completed work on a fully functional game engine that's even moar awesomer than UD? True story. :p

In case it didn't come across, I disagree with the whole "they don't need to prove anything" idea. They are showcasing their technology to the world and touting it as the ultimate killer app for game engines that will revolutionize the games industry, yadda yadda. Obviously they want to be taken seriously, but that means they need to back up their claims with actual facts, otherwise it's all just hot air. Some critics, like Notch (and Dwarden, apparently :D), don't believe them for one second and have outright called it a scam.

Others, like me, are just asking if they have addressed certain issues that we believe will pop up, but so far the answer has been "oh sure! Well, we're not gonna show you anything, but it's all totally sorted, we promise!".

Of course they don't owe us any kind of explanation, but one one owes them the benefit of the doubt either. :)

Given that most of us don't understand the current technology used (I should say all of us) then I think the transparency solution will be something else we won't understand. My guess is that the so-far-unused graphics card abilities will help with that.

I don't think anyone here claims to fully understand their technology, but that's also part of the point. We are mostly left guessing, so the gaps of information they've left open need to be filled in with something. Unfortunately that leads to speculation.

If only they would show some more functional examples, I think many of the critics would be silenced. The problem is that it's a year down the road since their last mega-hype announcement, but there doesn't seem to have been much progress. I think this sums it up pretty well:

---- 2010 ----

Euclideon: Looky here, people! We can make pretty pictures of 3D objects with unlimited detail!

Critics: Aha, cool so far. But how the hell do you animate something with such a huge amount of data?

Euclideon: *goes silent*

---- ONE YEAR LATER ----

Euclideon: Looky here, guys! We can still make pretty pictures of 3D objects with unlimited detail! But now we've copy&pasted a couple of decorated blocks a metric shitload of times to make our logo into a pretty landscape! Awesome, right?

Notch: FAAAKE!

Critics: But...

Euclideon: No, wait! Look, it does reflections too! Soooo awesoooome!

Critics: Yeah, great, but again - what about animations?

Euclideon: Yeah yeah, they work too. Whatever, fuck off...

Critics: Pics or it didn't happen!

Euclideon: Okay, here you go. *shows blocky cartoon animations*

Critics: WTF that looks like shit! Where's the unlimited detail?

Euclideon: *goes silent*

Well in that case I don't know what conclusive refution means for you specifically, as the guy refutes each voxel example individually.

Like I said before, I don't know about the whole voxel thing. The difference between voxels and point cloud data is marginal in any case, so I don't really care. It makes no difference regarding the points I've raised, none of which have been properly addressed.

Edited August 24, 2011 by MadDogX

[metalcraze 290]

I think that, for you, there will never ever be new technology, as it only builds on older technology :)

But that's exactly what everyone does. It's easier to improve what's there instead of reinventing the wheel (which Euclideon tries to claim they do).

And I wrote that Minecraft is often called a voxel game, and so Notch is quoted as being an authority on the subject. Can we agree that Notch has no authority on this subject and thus we can safely disregard his input?

Did Notch ever claim Minecraft was a voxel game?

I don't see what one has to do with the other. Besides it's the first time I hear that someone calls Minecraft a voxel game.

[dmarkwick 261]

Really? Well in that case... did you know I've just completed work on a fully functional game engine that's even moar awesomer than UD? True story. :p

In case it didn't come across, I disagree with the whole "they don't need to prove anything" idea. They are showcasing their technology to the world and touting it as the ultimate killer app for game engines that will revolutionize the games industry, yadda yadda. Obviously they want to be taken seriously, but that means they need to back up their claims with actual facts, otherwise it's all just hot air. Some critics, like Notch (and Dwarden, apparently :D), don't believe them for one second and have outright called it a scam.

I'm not saying they've solved everything they claim to have solved to any satisfaction, and it may be that their current animation progress is pretty ugly right now and would be detrimental to their agenda, I don't know. I think they're in a situation where they will simply not impress everyone and in any case aren't particularly interested in impressing everyone. They have sufficient funding, so why would they need to prove anything to anyone? Especially as there are always nay-sayers no matter what is shown. As a lot of people say, only a demo can alleviate doubt, and if the tech is not demo ready, what's the point in making the effort?

If I were him, I probably wouldn't have bothered to make that 41 minute video, but I guess he's human and would like to address some possibly unfair accusations by respected developers. (I'm not suggesting I'm not human BTW ;) :D)

Others, like me, are just asking if they have addressed certain issues that we believe will pop up, but so far the answer has been "oh sure! Well, we're not gonna show you anything, but it's all totally sorted, we promise!".

Of course they don't owe us any kind of explanation, but one one owes them the benefit of the doubt either. :)

Well, whatever :) I don't think it's going to make any difference to anybody in the long run. We can waft and flail at mist all we like, but either something will come of it, or nothing will come of it :)

I don't think anyone here claims to fully understand their technology, but that's also part of the point. We are mostly left guessing, so the gaps of information they've left open need to be filled in with something. Unfortunately that leads to speculation.

Well that last video went a long way to alleviating a lot of previous speculations. It would be an unusual video that alleviates ALL speculations :) there's always something. Always.

[sabre4809 0]

But that's exactly what everyone does. It's easier to improve what's there instead of reinventing the wheel (which Euclideon tries to claim they do).

Bullshit. Almost any major development in any area is down to innovation and not constant linear development. It is easier to do but nothing amazing ever happens from it.

Really? Well in that case... did you know I've just completed work on a fully functional game engine that's even moar awesomer than UD? True story. :p

A response: Cool man, hope to see it in a few years, any teasers

Yours and many others response: Nah rubbish show proof or not hapning, I seen dis movie before, why no videos, scam you try take my moneys.

It seems these days a teaser and early unveiling isn't just that anymore, its like it needs to be proof to people of its existence, and is subject to out of context examination.

Real or not, and I really hope its real just to shove it up the self appointed experts' arses(notch or whatever blockmans name is, included). If its real think what it would do with things like grass view distance, no more grass layer, and view distance in general and, and, and.The benefits are worth banking on and being hopeful for but as ever without getting over-hyped for disappointment.

Edited August 24, 2011 by SAbre4809

[dmarkwick 261]

The benefits are worth banking on and being hopeful for but as ever without getting over-hyped for disappointment.

This is exactly it.

[Dwarden 1124]

words just words ... show me some usable work for game development like AME has ...

in meanwhile i coded hyperrealistic 1kB engine it blast You away

because You don't need measure detail, it's just so real :)

[maddogx 13]

I think they're in a situation where they will simply not impress everyone and in any case aren't particularly interested in impressing everyone.

Bwuh? Have you actually listened to that guy talk? (I really wouldn't blame you for muting the sound and just reading the captions. :D) Why else are they making such a hype about their tech every once in a while if not to impress people? That's what generating hype is all about.

They have sufficient funding, so why would they need to prove anything to anyone? Especially as there are always nay-sayers no matter what is shown.

The term "nay-sayer" is a little strong, I think, as it implies that the critics are just being negative for the sake of it. However, you yourself admit that valid points of criticism have been raised, so it's not really nay-saying is it? ;)

Btw. if you look up the old thread on unlimited detail that was posted on this forum last year (after their last hype-video), you'll see that it was much more positive on the whole. I myself was much more positive about it then, and raised the animation and transparency issues mainly out of curiosity, because I wanted to know how they were addressed.

Now it's a year later, and still nothing much seems to have happened in this regard. Hence the increasing negativity. :)

Believe me, if they made a YouTube video showing nothing more than a couple of (high quality) rocks rolling down a (high quality) hill with a little bit of semi-transparent smoke along the way, I would retract all of my criticism and be nothing but deeply impressed.

As a lot of people say, only a demo can alleviate doubt, and if the tech is not demo ready, what's the point in making the effort?

Indeed. What is the point of showing up a year later without much in the way of progress? They must have some reason for making the video (especially in that attention grabbing tone), and I don't really blame people like Notch for assuming it's all a cash grab, even if I'm not quite there myself.

If I were him, I probably wouldn't have bothered to make that 41 minute video, but I guess he's human and would like to address some possibly unfair accusations by respected developers. (I'm not suggesting I'm not human BTW ;) :D)

That just goes to show that they are trying to impress people, especially the ones who are not already impressed. ;)

Well, whatever :) I don't think it's going to make any difference to anybody in the long run. We can waft and flail at mist all we like, but either something will come of it, or nothing will come of it :)

The same can be said for almost any subject that doesn't immediately affect our lives. Does that mean we should stop wafting and flailing? I think not. Personally, I enjoy it too much. :D

Well that last video went a long way to alleviating a lot of previous speculations. It would be an unusual video that alleviates ALL speculations :) there's always something. Always.

All it did for me was prove beyond a doubt that the pretty graphics are rendered in real time. In my mind, there wasn't much doubt about that in the first place. My main concerns remain.

Edited August 24, 2011 by MadDogX

[dmarkwick 261]

The same can be said for almost any subject that doesn't immediately affect our lives. Does that mean we should stop wafting and flailing? I think not. Personally, I enjoy it too much. :D

Yeah, me too :D

I might fess up to doing the opposite too (as I'm mostly doing now) ;)

---------- Post added at 12:38 PM ---------- Previous post was at 12:35 PM ----------

words just words ... show me some usable work for game development like AME has ...

Well, yeah, that's what a presentation is :) I'd guess it is a rare WIP engine that appears fully formed with all elements intact enough for immediate 3rd party development. As such, I guess the presentations of the last year or so have done their job, which was apparently to garner investment. Does this represent scamming? Only if there is NO development, which quite clearly is not the case.

[Dwarden 1124]

Phantom was fully working console ... remember that ... and whole portfolio of games

we all know how it ended

[dmarkwick 261]

Phantom was fully working console ... remember that ... and whole portfolio of games

we all know how it ended

I don't, never heard of it. *runs to Google*

[sabre4809 0]

words just words ... show me some usable work for game development like AME has ...

in meanwhile i coded hyperrealistic 1kB engine it blast You away

because You don't need measure detail, it's just so real :)

Why do they have to show anything to anyone before they actually unleash it properly into the world when its done.

The refreshing thing is that they don't actually give a fuck that so many people are rapidly bashing their keyboards to crush it in their own minds. Its normally a good sign.

[dmarkwick 261]

Minecraft isn't a voxel game in any meaning.

I'm compelled to pursue this a little :) on the subject of a voxel, what exactly is one single voxel, and what does it look like? It must be a cube right? And as monitors cannot render in cubes (unlike pixels which are naturally 2D and square) then each voxel needs rendering. If you get close enough to a voxel, it becomes larger than one pixel, and so needs to be rendered as a cube. Which I guess is what Minecraft does. A Minecraft world is made of cubes, very large voxels. As Notch has decided to speak authoritatively on the subject, I might make the not unreasonable link to his own engine, and the Euclideon engine.

If I were to make a gamble on this, I would gamble that Minecraft uses a rendering process not a million miles away from more "voxelly" looking voxel engines.