Voxel rendering (formerly Death of the GPU as we Know It?)

He doesn't address any of the complaints.

I like the guy, and how he made his way out of a dead-end existence. His alternative view of graphics has merit and he deserved a break. He doesn't answer any of the vital questions though, and he shows a fair bit of ignorance that's gone from quaint naivety to being somewhat embarassing. LOD = level of distance? And the examples Notch gave weren't explicit but about voxel data, yet Bruce just says, "we're not doing that (raytracing etc.)," without recognising the complaint. The subject of storage is raised and Bruce answers saying they have no problem with storage, but then they are using only a handful of objects, all regularly placed. There's nothing to indicate his sorting and search method will scale up suitably quickly. There's nothing mentioned about decent animation when a few basic wobbly objects are shown. There's nothing said about shaders and lighting except "that's coming". I have serious doubts he'll be able to pull it off. Basically the tech is a nice, rudimentary graphics demo, like the SAARCor detail demos, not bothering with any of the essentials required for a game. We need facial animation, dynamic lighting, physics interactions, and huge variety of content, none of which are a good match for compressed databases of points. Without showcasing any of that, or saying anything more than, "yeah, it's coming, trust us," I can't see any reason to believe in this tech. It's a spacial sorting algorithm on compresssed data, he's confirmed, fetching one datapoint per pixel (how's he going to deal with AA other than post or supersampling?), which isn't new, yet with his head in the sands Bruce believes it is.

 

It's clear that this will not be suitable for game engines any time soon. But application fields, like medical and seismic visualisation are good target group -- rendering huge amounts of raw information, similar to data mining.

 

Is this going to be anything new to what's already out there or in the pipeline from elsewhere? At its core is an efficient spacial database, which can be generate in different ways to the same 'lots of detail' effect. I'm not seeing anything new that others can't achieve with the same advancements in hardware tech that Dell has had since he started. And we'll surely have CUDA based SVO rendering on GPU that'll blitz what Euclideon can do.

 

I viewed this video about a year ago, with fractal datasets including the funny cube with cubic holes in it, and data based datasets (duh), the last one being big (uses over 32GB of memory)

 

There's a video interview on PC Perspective talking with id Software technical director John Carmack about the GPU race, Intel graphics, ray tracing, voxels, and other suitably techie topics. Due to popular demand (and threats), they have also subsequently posted a transcript of the interview, since, after all, reading is fundamental. The recently released Unlimited Detail Real-Time Rendering Technology Preview Video has attracted so much interest and criticism, his expansive answer on the topic is of particular interest at the moment, and interestingly he seems more skeptical about the actual value of "infinitive detail" than of the Voxel technology underlying the controversial video, saying: "in may ways it sounds awesomely cool: 'infinite detail,' but if we look at all of the trends that we’ve been doing and Rage epitomizes in many ways, procedurally generated detail is usually not what you want. This has been an argument going back decades: 'now is the year of procedurally generated textures and geometry.' We’ve heard that for a decade and it never has come true. What has won is being able to manage the real data that we want." He also says he now believes ray tracing is more likely to eventually win out over voxels anyway, and here's the portion with his take on this:


I’ve revisited voxels at least a half dozen times in my career, and they’ve never quite won. I am confident in saying now that ray tracing of some form will eventually win because there are too many things that we’ve suffered with rasterization for, especially for shadows and environment mapping. We live with hacks that ray tracing can let us do much better. For years I was thinking that traditional analytical ray tracing intersecting with an analytic primitive couldn’t possibly be the right solution, and it would have to be something like voxels or metaballs or something. I’m less certain of that now because the analytic tracing is closer than I thought it would be. I think it’s an interesting battle between potentially ray tracing into dense polygonal geometry versus ray tracing into voxels and things like that. The appeal of voxels, like bitmaps, [is that] a lot of things can be done with filtering operations. You can stream more things in and there is still very definitely appeals about that. You start to look at them as little light field transformers rather than hard surfaces that you bounce things off of. I still wouldn’t say that the smart money is on voxels because lots of smart people have been trying it for a long time. It’s possible now with our current, modern generation graphics cards to do incredible full screen voxel rendering into hyper-detailed environments, and especially as we look towards the next generation I’m sure some people would take a stab at it. I think it’s less likely to be something that is a corner stone of a top-of-the-line triple A title. It’s in the mix but not a forgone conclusion right now.

linky :
http://www.pcper.com/reviews/Editor...ce-Intel-Graphics-Ray-Tracing-Voxels-and-more

 

By any chance did anyone save off any or all of the larger versions of the images from the tkarena article that was the first post of this thread? Strange thing is I was looking at them and reading the article last night and now it is all gone and archive.org does not have all the pages or pictures. There was a good amount of information in the pictures alone in the high resolution ones and I just want to be able to study them more.

 

After a short panic attack I was able to find a repost of the article here http://www.gamingface.com/2011/08/death-of-gpu-as-we-know-it-re-post-from.html with the images.

 

Thanks for posting the link TKArena is gone and I backed it up on www.gamingface.com for you all.

I also posted some extra information a on what I got to see here :
http://www.gamingface.com/2011/08/unlimited-detail-in-swtor-or-other-mmo.html

 

People are still using depth buffers.

It has been clearly formulated by Turing himself (in his famous paper):

Such a machine would be circular and not mark its own S.D with the symbol "u" (unproductive loop on D's S.D): contrary to the hypothesis.
Note that infinite/unlimited/unbounded is weaker than transfinite.

The original UD had no LOD. Still, it remained very fast.

History abounds with confirmations of the above; cf. http://www.pnas.org/content/108/22/9020.long

 

Discussion on use of UD in games moved here.

 

Why the console forum of all places ?
this tech while hardware agnostic is curently running on the pc and will most likely debut there if it ever does.
Plus religious grounds prevent me from spending too much time there

 

If the suspected high RAM needs are true, then it's definitely not going to debut on consoles. And I subscribe to Davros's orthodoxy.

 

Because it was a stupid discussion headed by a stupid poster who got banned for being stupid. And we all know the place for the dregs of forum noobs is among the console users.

Also, it was a discussion on the relevance of UD being the new black for gaming, which by nature of the market means it needs to be a console tech which is why it was accepted as a console tech thread. The opening post was, "is this good for consoles".

 

Can I nominate repi for the B3D "Best One Word Post Of The Week" award.

 

I don't mean the streaming bandwidth, but the processing data bandwidth. The main problem with casting rays into hierarchical homogeneous grid geometry is incoherent data access.

 

What happens some times, is that voxel based forum threads have high latency replies.

 

I just started a new thread, not much interest in voxels it seems.
With all the memory the new consoles have, I'm anticipating some pure voxel games.