Unlimited Detail = It's A Scam

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
The developer behind Minecraft has a few choice words about the people that put out that "Unlimited Detail" video we posted yesterday (here). Wow, you don't get much more to the point than that...and this guys knows what he is talking about.

"They’re hyping this as something new and revolutionary because they want funding. It’s a scam. Don’t get excited." Or, more correctly, get excited about voxels, but not about the snake oil salesmen.
 
I wasn't very impressed to begin with. However that atomage youtube video was amazing.
 
Watching and listening to that Voxelstein video gave me a warm fuzzy feeling inside. :)
 
OK I was taken in by this video.
I sent it to [H] as news :eek:

Never thought to calculate how much data they claimed to be representing.
 
seems like notch is pointing out what i already stated was the major draw back, the massive amount of ram this would require to make a decent size world.

it is clear the demo is duplicating areas on the map due to the crazy memory consumption. Clearly this engine will not find its way into MMO's anytime soon. Small scale first person shooters, maybe.

as for animation, I have to take developers words on this one, but I suspect better rountines will be made if there is enough desire.

The 2 main reasons to call this a scam are a tad bit weak. We can make computers with more ram ( or find games that need small worlds ), and we have yet to perfect the art of animation using other technologies.
 
The developer behind Minecraft
That's the guy that can't even write a java app? Credible source!

Point clouds are the future once algorithms are developed to calculate them in real time rather than having to store all the points ahead of time. Here's what I wrote in the other thread about this:

I've been working with LIDAR point clouds for years now and the only issue I see is that they are storage intensive. For each point at the very least three values are stored, X, Y, and Z, in my case those three are all doubles, so that's 192 bit just for the position of the point. In gaming you'd need to add color to it which is another 24 bits at least, probably 32, for a total of 224 bits per point.

Now he was talking about billions of points, do the math and you can see how storage requirements are an issue and how there needs to be some way to procedurally create the majority of the content on the fly so that the geometry isn't stored but calculated.

I suspect that will be the true problem in adopting this technology for gaming.
 
seems like notch is pointing out what i already stated was the major draw back, the massive amount of ram this would require to make a decent size world.

it is clear the demo is duplicating areas on the map due to the crazy memory consumption. Clearly this engine will not find its way into MMO's anytime soon. Small scale first person shooters, maybe.

as for animation, I have to take developers words on this one, but I suspect better rountines will be made if there is enough desire.

The 2 main reasons to call this a scam are a tad bit weak. We can make computers with more ram ( or find games that need small worlds ), and we have yet to perfect the art of animation using other technologies.

If we're going to have computers with that much ram, wouldn't it be better spent on that many more polygons? In addition to RAM size, there's the size of it stored on local HDD. For that island in their video, they repeated the same trees, rocks, and square blocks of land to hide the fact that they don't have the space for saving each unique object that it would require to fill that island.
 
If we're going to have computers with that much ram, wouldn't it be better spent on that many more polygons? In addition to RAM size, there's the size of it stored on local HDD. For that island in their video, they repeated the same trees, rocks, and square blocks of land to hide the fact that they don't have the space for saving each unique object that it would require to fill that island.

Specifically the point cloud engine wasn't discredited at all. Notch merely mentions that the storage requirement would be too large to be practical. So what? The challenge now is to calculate the point cloud in real time at a decent frame rate. CPUs and GPUs all add more cores and more cycles, it's a challenge that will in time be overcome and Euclideon may just be the company to do it.

Whenever some human says "this can't be done" another human tends to go out and prove the first one wrong. Happens all the time.
 
I dunno, I read the whole "scam" article earlier 512 Petabytes of info needed? That's all fine and dandy if you feel that there needs to be 1 byte of data per point. But what if I simply give an equation for a sphere x² + y² + z² = 1, now lets make this sphere made out of 10^10 "rocks" with each with 10^10 atoms per rock. Well you don't need to keep track of each atom, you can simply have a limit on how many are actually "kept track of" depending upon your zoom level (some function of screen resolution. Granted you'll get detail that 'pops into view" similar to polygons, however depending upon how fast you zoom, you could have the detail "appear" (in memory) a split second before you physically can see it.

Disclaimer: I am not a computer programmer so I'm going on logic more so than what is can be done. However as a physicist (of sorts) I have experienced more than once something that "could not happen", happen, simply because a new technique made it so.
 
I dunno, I read the whole "scam" article earlier 512 Petabytes of info needed? That's all fine and dandy if you feel that there needs to be 1 byte of data per point. But what if I simply give an equation for a sphere x² + y² + z² = 1, now lets make this sphere made out of 10^10 "rocks" with each with 10^10 atoms per rock. Well you don't need to keep track of each atom, you can simply have a limit on how many are actually "kept track of" depending upon your zoom level (some function of screen resolution. Granted you'll get detail that 'pops into view" similar to polygons, however depending upon how fast you zoom, you could have the detail "appear" (in memory) a split second before you physically can see it.

Disclaimer: I am not a computer programmer so I'm going on logic more so than what is can be done. However as a physicist (of sorts) I have experienced more than once something that "could not happen", happen, simply because a new technique made it so.

Not sure where you are going with that one. You are just limited to a sphere. Not many games consist of just viewing a single sphere. Not very exiting.

Why even make the sphere out of the 10^10 discrete elements, why not just just represent it directly since you have the equation?

I hope you see what I am getting at...
 
I get that maybe the unlimited detail guys didn't go to great lengths to point out the short comings of the engine they are developing. but I fail to see how that makes them scam artists. giid marketers yes, scam artists not at all.

to give an examplr when nvidia released the gtx 400 series their marketing material didn't say Buy the gtx 480!!! it's loud and Hot woot. why because thats how good marketing works. Now if they had said buy the gtx 480 it's cooler quieter and more efficent than a radeon 5870 that would have been a scam.

To expand more you don't see ads that say "buy a hummer you'll get 5 mpg yay", or "Dominos pizza tastes like shit but hey it's cheap" or "Apple pc's sure they don't do shit but wow your other upity white douchebag friends will be jealous", or "Alienware because hey you've get two grand you just wanted to throw away anways", or "at university of pheonix you can get your BA in business in as little as two years and be managing a taco bell for $15,000 a year faster than you ever thought possible".

tldr: not going to lengths to point out the shortcomings of a product you are selling is not scamming its good marketing.
 
I get that maybe the unlimited detail guys didn't go to great lengths to point out the short comings of the engine they are developing. but I fail to see how that makes them scam artists. giid marketers yes, scam artists not at all.

to give an examplr when nvidia released the gtx 400 series their marketing material didn't say Buy the gtx 480!!! it's loud and Hot woot. why because thats how good marketing works. Now if they had said buy the gtx 480 it's cooler quieter and more efficent than a radeon 5870 that would have been a scam.

To expand more you don't see ads that say "buy a hummer you'll get 5 mpg yay", or "Dominos pizza tastes like shit but hey it's cheap" or "Apple pc's sure they don't do shit but wow your other upity white douchebag friends will be jealous", or "Alienware because hey you've get two grand you just wanted to throw away anways", or "at university of pheonix you can get your BA in business in as little as two years and be managing a taco bell for $15,000 a year faster than you ever thought possible".

tldr: not going to lengths to point out the shortcomings of a product you are selling is not scamming its good marketing.

There is a HUGE difference. These guys will never deliver a product. They just want to raise more "investment capital" No game engine will ever see this.

On the other hand Nvidia actually has a track record for delivering actual products. AND they don't require investements to deliver their products.

These guys are just theives.
 
There is a HUGE difference. These guys will never deliver a product. They just want to raise more "investment capital" No game engine will ever see this.

On the other hand Nvidia actually has a track record for delivering actual products. AND they don't require investements to deliver their products.

These guys are just theives.

Just like every other start up company or technology startup?

It costs money to make money.
 
I dunno, I read the whole "scam" article earlier 512 Petabytes of info needed? That's all fine and dandy if you feel that there needs to be 1 byte of data per point. But what if I simply give an equation for a sphere x² + y² + z² = 1, now lets make this sphere made out of 10^10 "rocks" with each with 10^10 atoms per rock. Well you don't need to keep track of each atom, you can simply have a limit on how many are actually "kept track of" depending upon your zoom level (some function of screen resolution. Granted you'll get detail that 'pops into view" similar to polygons, however depending upon how fast you zoom, you could have the detail "appear" (in memory) a split second before you physically can see it.

Disclaimer: I am not a computer programmer so I'm going on logic more so than what is can be done. However as a physicist (of sorts) I have experienced more than once something that "could not happen", happen, simply because a new technique made it so.

For Halo:Reach, a lot of tesselation was used to negate the popin effect as much as possible (so large maps like Forge World, Spire, and Boneyard (which is two halves, but still huge overall) could exist.
 
There is a HUGE difference. These guys will never deliver a product. They just want to raise more "investment capital" No game engine will ever see this.
And you know this how?
The world is flat.
Humans will never take to the skies.
Humans will never go to space.
Yadda yadda yadda, just because some people can't wrap their mind around possibilities doesn't mean that those possibilities do not exist.

This isn't a conceptual problem, it's a technology hurdle that will in time be overcome simply because there is no logical reason why it wouldn't be overcome.
 
And you know this how?
The world is flat.
Humans will never take to the skies.
Humans will never go to space.
Yadda yadda yadda, just because some people can't wrap their mind around possibilities doesn't mean that those possibilities do not exist.

This isn't a conceptual problem, it's a technology hurdle that will in time be overcome simply because there is no logical reason why it wouldn't be overcome.

Tailsman was a technological leap that was DOA because nVidia and ATi killed with simply making GPUs have more brute power.
 
When I first watched the video on this site yesterday I thought to myself, this isn't that impressive and the dude talking sounds like a scam artist.

Then later that day I reflected back on it, and thought to myself, how the hell do you animate anything using that method?

Then I see this today. Still not impressed.
 
That minecraft developer sounds like a bit of an asshat, he goes on to list all possible flaws with the way the tech is used and all it's limitations - goes on to even call it a "scam" and then and the end says get excited about voxels.

It sounds like he's more upset with the way this "unlimited detail" was introduced rather than what it could be capable of, and he makes his mind up on this engine based on a handful of tech demo's.

Well I ask how are we meant to get excited about a tech that he just berated..., you know that tech needs to be embraced before we see great advancement in how it works. If he's so against the tech why would he then refer to other engines using it as "impressive" and to get excited but then pretty much say not to back it? - Personal bias towards the way this engine was advertised it sounds like.

Without any underlying understanding of what this engine is capable of all of the slurs mean nothing as for all we know they have been overcome to some degree.

GO BACK TO MINECRAFT!!!!
 
There is a HUGE difference. These guys will never deliver a product. They just want to raise more "investment capital" No game engine will ever see this.

On the other hand Nvidia actually has a track record for delivering actual products. AND they don't require investements to deliver their products.

These guys are just theives.

don't require investments? you know nvidia is a publicly traded company, right?

anyways, i remember when a 32MB PCI card was just the tits.

i just find it bewildering that here we are, after watching generations before us be blown away by a jump in memory size from 64KB to 640KB, and now we are the very people proclaiming that it can't be done and it will never happen.

I can't wait to see what the landscape looks like in 15 years. you know, like how less than 15 years ago we were all playing duke nukem 3d, enabling free-look and being so impresed with half life. or when geforce introduced hardware T&L and topped out at an earth shattering 64MB on AGP lanes. remember when a 20GB hard drive was enough? or hell, 20 MEGABYTES?

i can't understand how people on a forum dedicated to advancing technology think that we will just hit this impenetrable wall in perception of detail. i feel like i'm in a town hall meeting where people are discussing the idiocy of that newfangled steam powered horseless carriage nonsense
 
I don't think anyone is arguing that detail won't advance; only that this particular product being presented is a scam.
 
Then later that day I reflected back on it, and thought to myself, how the hell do you animate anything using that method?

lots and lots of computations. or some other method that will be employed as technology allows. last time i checked, my arms were not made of polygonal models, and they seem to animate just fine.

technology has always been a hurdle, not a wall. people have a devious way of figuring out ways to use the world around them to get what they want. dude, slingbox or whatever lets some dude broadcast an invisible electronic signal that gets shot into space, then beamed back down exactly to some guys house, then put in a box and sent on cables that go under goddamn oceans for thousands of miles and through hundreds of gateways before winding up at my computer. all because some guy said it would be cool to watch tv shows hundreds of miles away from home. at the time, perceived technology limits prevented that. everything will eventually get broken down.
 
And you know this how?
The world is flat.
Humans will never take to the skies.
Humans will never go to space.
Yadda yadda yadda, just because some people can't wrap their mind around possibilities doesn't mean that those possibilities do not exist.

This isn't a conceptual problem, it's a technology hurdle that will in time be overcome simply because there is no logical reason why it wouldn't be overcome.

Just like every other start up company or technology startup?

It costs money to make money.

don't require investments? you know nvidia is a publicly traded company, right?

anyways, i remember when a 32MB PCI card was just the tits.

i just find it bewildering that here we are, after watching generations before us be blown away by a jump in memory size from 64KB to 640KB, and now we are the very people proclaiming that it can't be done and it will never happen.

I can't wait to see what the landscape looks like in 15 years. you know, like how less than 15 years ago we were all playing duke nukem 3d, enabling free-look and being so impresed with half life. or when geforce introduced hardware T&L and topped out at an earth shattering 64MB on AGP lanes. remember when a 20GB hard drive was enough? or hell, 20 MEGABYTES?

i can't understand how people on a forum dedicated to advancing technology think that we will just hit this impenetrable wall in perception of detail. i feel like i'm in a town hall meeting where people are discussing the idiocy of that newfangled steam powered horseless carriage nonsense

Seriously people, I know it sucks, but it's a fucking scam and the guy is a fucking scammer.

I don't think anyone is arguing that detail won't advance; only that this particular product being presented is a scam.

Thank you
 
I don't think anyone is arguing that detail won't advance; only that this particular product being presented is a scam.

why is it a scam? because it takes more memory and processing power than the typical computer of 2011?
 
Seriously people, I know it sucks, but it's a fucking scam and the guy is a fucking scammer.



Thank you

hey guys, stop wondering, because we now have the Okashira Official Decree that this is, indeed a scam. Thank you, you've answered all my questions.
 
I'm no expert, but I can see both sides. On the one hand, I think voxel rendering is an exciting technology that will probably have its day. On the other hand, I think the overall tone of the video, which to me was "support our work and in no time games will all switch from polygons to voxels" is suspect at best.
 
well, I don't think the concept is necessarily a scam but I don't think it's the holy grail it's made out to be, either. At any rate, let's just wait and see instead of getting the pitchforks out.
 
I'm no expert, but I can see both sides. On the one hand, I think voxel rendering is an exciting technology that will probably have its day. On the other hand, I think the overall tone of the video, which to me was "support our work and in no time games will all switch from polygons to voxels" is suspect at best.

What makes it more suspect is that they don't once use the word voxel, not even to compare their technology to voxels (if it is indeed different).
 
hey guys, stop wondering, because we now have the Okashira Official Decree that this is, indeed a scam. Thank you, you've answered all my questions.

Aw crap... there goes my hopes and dreams... :p

Who cares if it's real or not. It's not available to the average consumer yet (i.e. You and I in case you forgot). If it turns out to be vapor ware, then it's vaporware. I'm really not getting the reason to be so heated over something that we cannot use (real or not) yet...

If you're seriously popping a vein in the forehead over this... I think you may need to take a long trip away from the basement...
 
lol technology that is used in weather mapping and physics simulations is ar scam!!!11

Shut up and come out with the adventure update, you Swede.
 
What makes it more suspect is that they don't once use the word voxel, not even to compare their technology to voxels (if it is indeed different).
True enough. I guess I got caught up in all the voxel talk here. ;)
 
If you're seriously popping a vein in the forehead over this... I think you may need to take a long trip away from the basement...

well i certainly didn't mean for anyone to mistake my sarcasm for rage. thought the fact that i've been working no-shit 18-22 hour days (sleeping at work) for the past 6 days and the fact that this is the first time i've been home earlier than 2130 may have colored my posts. oops!

i just think it's kind of silly to immediately dismiss new technology and ideas because it "cant be done today." i prefer to see my posts more as the musings of some well-respected life philosopher speaking out the side of my mouth as i puff on the pipe in my chair by the fireplace, instead of some fat greasy nerd breaking my keyboard in mom's sewing room :p
 
For Halo:Reach, a lot of tesselation was used to negate the popin effect as much as possible (so large maps like Forge World, Spire, and Boneyard (which is two halves, but still huge overall) could exist.

What? Isn't tesselation DX11? How can the 360 run that?
 
yep, the gaming jesus said it is a scam, so it must be.

You didn't have to read Notch's statement to figure that out, just listen to the video and count the number of times the douche talking said "unlimited". He's been trying to hype this up since 2008 by the way. He's using well known tech and trying to push it as new and exciting so he can skip off with investor money. He even stated that when he presented it to the techs of a developer company, they wouldn't let them even get to the board of directors to pitch it.

Simply, the guy is full of shit exploiting old well known tech that's currently impractical for gaming and marking it off as new and exciting, while purposefully leaving out its limitations.
 
What? Isn't tesselation DX11? How can the 360 run that?

The 360 uses a sort of hybrid DX10/DX9 GPU from ATi (now AMD of course) and I believe it has a tessellation engine in it (AMD's DX10/10.1 GPUs have had tessellation engines in them) if I recall correctly.

I think this guy is totally full of shit and his specific "OMG IT"S GONNA BE THE NEXT BIGGEST THING EVERRRSSS LOLOLOLOOLOL" strain of marketing annoys me to death because he doesn't even bother to state any of the downsides. I do think there is a future for this technology, just not from this guy because he seem out to just get investor money.
 
I watched the video and thought it sounds like something cool in the future. But I never heard him say coming soon I took it as a look at what we are doing and where we will hopefully go one day.
I don't know if this is a hd-dvd or a bluray we will all find out sooner or later. But a scam? come on he didn't ask me to buy anything, but every development has to have some money to make it happen, every development, microsoft, the atom bomb, the electric car, a nuclear reactor, whatever is developed has to have someone pour some money in it sooner or later.
 
Man, there's a lot of people that must lead real boring lives in here. Almost pity them for looking at everything so negatively.
 
What? Isn't tesselation DX11? How can the 360 run that?

Tesselation has been in ATi GPUs, in some form or another, since ~2002. It was not accessible via the DX API until DX11. OpenGL had an extension for it, along with a near machine instruction set open for it, too.
 
Back
Top