"Unlimited Detail"

Started by
30 comments, last by Lode 14 years, 3 months ago
Quote:Original post by maspeir
I'll believe this when I have a working demo sitting on my computer.

This is essentially my attitude as well. I am extremely interested in both these technologies as I consider them the closest competitors to my own engine, but until I actually play an interactive demo on my own machine it is hard to know if they work as claimed. Though I have to admit the videos are pretty impressive (Atomontage at least, I didn't watch the Unlimited Detail ones yet).
Quote:Original post by Ysaneya
...games are trying to move away from pre-processing as much as they can, and I think the future will be interactive environments...

While I generally agree, the Atomontage videos do show some rather cool examples of truck tyres leaving tires marks in the ground, and of dynamic smoke being animated in real time. It would be interesting to know how much preprocessing is actually involved.

[Edited by - PolyVox on December 30, 2009 7:55:13 AM]
Advertisement
Without looking at anything technical at all, I find the website not really convincing at this time. They're quite vague, and quotes like "Unlimited Detail is believed by many to be the one of, and possibly the most significant piece of technology of the decade." don't really convince me unless there's some proof to back it up. They show lots of pretty pictures (strange enough, in 256-color palette though), and spend a lot of paragraphs explaining the problems of conventional engines, but again I miss the proof to show their engine does it better.

But nonetheless, would be cool if it's true, so hopefully we really get to see this some day.

There is something that disappoints me in their pictures, and that is that in some of those there is grass but you can clearly see a repeating tile pattern. I really hope repeating tile patterns (especially in nature-scenes) will be a thing of the past.

[Edited by - Lode on December 30, 2009 9:13:54 AM]
Interesting, but there are some new DirectX 11 demos that show off the tesselator unit and they almost get pixel sized triangles and do real displacement mapping (not parallax occlusion mapping).


-----Quat
Quote:Original post by Lode
Without looking at anything technical at all, I find the website not really convincing at this time.


Same here.

Quote:
It really is Unlimited, Infinite, endless power, for 3D graphics.


I might be wrong, but this statement seem quite strange. Everything involving infinite (expecially when computers are used) sounds not very professional.


I'm a little skeptical as well. They claim unlimited by we all know computers are quite finite. They always mention a 1024x768 resolution but that's, IMO, a thing of the past. If we're really talking about revolutionizing graphics, the target is at least 1920x1080 which is 2.6x as many points. I'm curious to not only see this in action on my computer, but see whether their algorithm actually scales to HD given that it's quite a bit more points to be put on screen and they never mention anything about any resolution besides 1024x768.

In addition there's all the other issues such as authoring the content, memory usage, animations, etc. Even if you could use this for non-animated objects, would you want to? Mixing something like this with traditional polygonal animated meshes would likely look worse due to the contrast in geometry.

I'm putting this in my bucket along with OnLive and all the other technologies that sound great but need to be proved to me still.
Quote:Original post by NickGravelyn
I'm putting this in my bucket along with OnLive and all the other technologies that sound great but need to be proved to me still.


See here for an update. Taking into consideration the numbers, custom video encoding hardware, custom routing and custom controller, crunching the numbers given by them, it is actually possible for online to have less latency for some than running it via stock x-box controller locally.
Quote:Original post by NickGravelyn
I'm putting this in my bucket along with OnLive and all the other technologies that sound great but need to be proved to me still.


See here for an update. Taking into consideration the numbers, custom video encoding hardware, custom routing and custom controller, crunching the numbers given by them, it is actually possible for online to have less latency for some than running it via stock x-box controller (when multiple controllers are connected) locally.

Guess we'll see soon enough.
Quote:Original post by Antheus
Quote:Original post by NickGravelyn
I'm putting this in my bucket along with OnLive and all the other technologies that sound great but need to be proved to me still.


See here for an update. Taking into consideration the numbers, custom video encoding hardware, custom routing and custom controller, crunching the numbers given by them, it is actually possible for online to have less latency for some than running it via stock x-box controller locally.
That's not possible. If I use a wired 360 controller, OnLive can't possibly beat that. And even when he talks about the wireless controllers, he says there's about 20ms of latency which is still 1/4 of OnLive's target 80ms.

At the end of the day, they've demoed it with one guy playing 250 miles from a server that is likely only being used by, what, a few hundred beta testers? I live in Seattle where my nearest server would be, according to their map, southern California (i.e. well over 1000 miles away and surrounded by some massively populated areas). I simply won't believe that experience is going to be better than my Xbox 360 sitting in my living room until it's out and proven.

(Not to mention I like my 360 game pad and their controller looks ugly. :P)

EDIT: Whoops, guess this is a tad off topic. Back to Unlimited Detail: I'm still doubtful. :D
My problem with this, other than the obvious doubts, is that assuming it's true, what does this mean for games? In the past few years, I have seen an apparent rush towards better graphics at the expense of content. I'm an Elder Scrolls fan. Take Morrowind for example. The graphics sucked, but the story was very in depth and it took at least 100 hours to get through the main quest alone. Then oblivion comes along and the graphics are amazing and I can beat the main quest and most of the side quests in the same amount of time.

I'd much rather have lower quality polygonal graphics with a good, in depth, story and awesome game play than near realistic graphics that take so much time and resources to develop that the developer cannot spend time on what is important.

No, I am not a professional programmer. I'm just a hobbyist having fun...

I have to heavily agree with that. A few weeks ago I was so bored by recent games, that I grabbed SCUMM VM and replayed Monkey Island 1+2 and DOTT with amazing joy, which defenitely didn't all came from nostalgia (which certainly was present too). Good and witty story telling is without doubt falling short when looking at a percentage of total games released per year.
The thing is, that this isn't the fault of any coder or art department. It quite often is big publishers that often are somewhat ignorant and deadline pusing, which so often cripples titles that sounded great, that it hurts. Thus it isn't the mere availability of better GFX, but the marketing oriented companies, and in the end the customer, which often seems to be lured more easily with stunning graphics, than good games. Since companies strive for cash, and only with few exceptions for ideals, they of course cater to that. Unfortunately on the expense of those, that enjoy and create quality games, which seems to be a miority.
Stopping progress in graphics isn't going to change one bit of this machinery I fear.

This topic is closed to new replies.

Advertisement