Originally Posted by bfar
good post
Photo realistic gaming won't happen in our lifetime.
I'll raise your bet and say it happens in the next 4 years. Well within any of our lifetimes .....unless you get hit by a photorealistic bus of course
Feel free to quote me when the time comes.
We'll also have robot servants within 10, to the elite at least. Already exist btw.
i'd say 10but easily within out lifetime
"Cast off your fear. Look forward. Never stand still, retreat and you will age. Hesitate and you will die. SHOUT! My name is…"
//James
Yeah, 10 is more likely for 'proper' games.i'd say 10
I was just meaning one of those tech demo minigames to show off hardware.
Like the ATI ball pit or the human head demo
IMO gameplay is far more important than realism. All the effort being put into photorealistic rendering for gaming is entering wasted resources territory. And the heavy price tag are likely starting to show diminishing returns for IHV's and ISV's. Move onto immersion now, then worry about the ultra-fine details. (I bet Avatar isn't anywhere near as impressive on a 20 inch monitor)
A game 5 years old can create a much more immersive experience on a multimonitor display than the best rendered games of today. Makes sense when the ultimate goal of photorealistic gaming is immersion.
Add good gameplay to the immersion of multimonitor and its a winning combo IMO. 3d on a multimonitor setup would probably even better, but it would have to be done in a way that doesn't involve glasses. That just kills it for me, since I where glasses already.
Yet the evolution of graphics stopped in 2007. Crysis's level of graphics has yet to be reached and that's 2.5 years ago. Now that consoles took over there is no incentive for (much) better graphics. The next gen of consoles will only match Crysis level of graphics and that would be by 2012, half a decade after crysis itself.
PC gaming is dead and that means that an industry which was moving in the pace of an F1 car now it is moving no faster than a donkey. It like a switch gone off. I don't think that a lifetime is enough for those who wish to see photorealistic graphics, a millenium would be more accurate...
i wouldnt be so sure... the pc segment is about to collapse, yes, but pc gamers wont just dissapear and go back to inferior interfaces and image quality... the market will continue to evolve, at console pace unfortunately... so it will be very slow steps... but big ones...
Last edited by saaya; 01-27-2010 at 06:12 AM.
No.
For example, Crysis.
So, let's say it represents the current level of our gfx tech advancement. Let's try and get it to the realistic level.
All the modern accelerators even in multi-GPU setups struggle with Crysis with proper realistic resolution (4k x 3k).
So add a year for the tech to catch up at least (++ computing power requirements).
Texture quality is also lacking. Need 2x-4x higher texture quality. 2-3 more years (++ computing power requirements).
Geometry... is WAY behind. If you want photo realistic image quality in a 3D game you need 10-20x more polygons at least. That's 5-6 more years till it's actually possible to render that (++ computing power requirements).
Current lighting is crap. We need raytracing. Realistic raytracing with huge resolutions won't be possible really soon, 5 years at minimum (++ computing power requirements).
Summ up all the hardware performance requirements... this is quite a huge jump needed in order to keep a decent frame rate of such a realistic game. This can not be achieved in 2-3-5 years, for sure.
Once the hardware is out, we can create such a game!
And now imagine creating such a game. The content. Each model would take a TON of time to create. Such a game would take 5 years to develop for a HUGE studio.
So don't expect anything within the next 10 years for sure.
Last edited by zalbard; 01-27-2010 at 06:06 PM.
Donate to XS forums
Sure, I was bit over the top to my speculation. Still my line of thought stands considering that the PC segment was the training ground for next gen graphics. Now that game devs are completely untrained to the new techniques introduced by DX10 and DX11 (since there is no incentive not to be so), they would have to develop those skills from scratch by the time that the next gen of consoles will be released, which won't happen overnight.
X360 looked impressive at first but that was because developers were using techniques which they already knew from the PC arena (DX9, shader 3, HDR), X720 (or however it will be named), it would be far less so; since games are still written in DX9 code there would be no prior experience for anything new by then...
Maybe it won't take a millennium but certainly more than a (normal) lifetime to see photorealistic graphics.
Your wrong about the bolded part.
Right now, a 3d artist, when creating an object/character in a next gen game creates a really high-poly, photorealistic model, using 3dsmax/maya and zbrush or mudbox for finer details, than creates another identical, low poly version of that model (which will be the in-game model) and bakes all details in normal maps, displacement maps and Ambient occlusion maps from the high poly to the low poly model, so in game, with proper shading, the result looks good even if the geometry lacks.
So, in the future, you won't have to create the extra low-poly model, and bake everything from the high to low version. You will just create a high-poly model and use it in-game.
You will actually cut down on production time.
here is an example, made by Vitali Bulgarov.
High poly model, made in XSI and fine tuned in Zbrush. All the small details are actual geometry, everything is real and not a shading trick (bump maps etc..):
Low poly model, with all maps baked in to it from the high poly (normal maps and ambient occlusion maps) + color textures.
Last edited by Florinmocanu; 01-27-2010 at 07:28 AM.
You do realise that high polygon models will take 4-10x more time to create, right?
Look at the movies, that's not actually realistic gfx yet, cause you can't really come closer and look at everything carefully.
They take 5+ years to draw and render (Avatar for example), and this is very little content comparing to any interactive game, cause they only create the stuff that makes it into movie's frames... And for 2-3h of entertainment only...
Last edited by zalbard; 01-27-2010 at 07:34 AM.
Donate to XS forums
zalbard, have you read what i just said in my previous post? read again.
Right now, every object created in a game is created in 2 version, a high poly one, with million of poly's and a low poly one, between 200-300 polys or 5-10k for characters.
So in future, with better geometry performance, artists won't have to go through the extra step of creating the low poly version, they will just do the high-poly version.
And avatar started to be worked on around 2005, so around 4 years in production, but a production pipeline in a movie is way more complex, with a lot of research, inventing new ways of doing things.
You cannot compare a movie production pipeline with a game pipeline.
But better geometry will mean faster workflow, the low poly models are the hardest to create, since you have a small poly budget and you have to convey that into the lightest shape, with the best detail reproduction possible, while also being carefull how you do the model so you don't get weird deformations while animating, since the model is low poly and they will appear.
Also, in movies if you make a forest, you have to make hundres of different trees, you cannot make 2-3 models and duplicate them. In games you create 2-3 models of trees, 10 max and then you multiply them and that's it, the forest is done. And so is the situation for any model/prop in-game. It takes a lot more time to create those 2-3hours of entertaiment for a movie than to create a game.
PS: i work as a 3d artist, trust me, better geometry performance in the future will mean faster workflow.
Last edited by Florinmocanu; 01-27-2010 at 07:56 AM.
How long do you think before we see the geometry implemented in pc gaming? Just picking your brain a bit here since you do it for a living.
well, for the moment this workflow with low and high poly model will continue. But, it will take advantage of tesselation to get rid of the low poly looks.
You have the high poly version and you extract a displacement map from it. You use it on the low poly version, the in-game one and with the aid of tesselation and displacement you create geometry on the fly and enhance the low poly model to mimic the geometric look of the high poly.
And since tesselation is distance dependent, you get better performance since you only need dense geometry in the near-by area.
Even in movies they use medium detail models (so you never see weird geometry on them) and use displacement to create small stuff like skin pores etc..
Frankly, i don't believe we will see actual full blown models in games or movies, it wastes to many resources. We will see that games in 10-15 years will start to look closer to what movies look, since DX 11 introduces tesselation and displacement. These 2 features will make possible for games to move towards movie visuals in 10-15 years, depending on how hardware and software API's evolve.
Photo-realistic.
Not physically perfect and exact to real life environments. Just has to look like it.
Slap some better lighting and massive AA onto gran turismo 5. That's photorealistic.
http://generationdreamteam.free.fr/a...alLifeJuly.jpg
Lol, come on... That's in motion... Just stop driving for a sec and look around, should be enough to ruin the impression unless you're blind.
Donate to XS forums
You're thinking of hyper-realism.
Not photo-realism. We're practically there for photorealism right now at this very point in time.
If the brain can mistake something for being real, that's enough to qualify.
We're not going to be taking electron microscopes to game footage to verify their physical correctness. There comes a point when you cross the barrier of what realism is in terms of game graphics.
I think it was the 4k x 3k resolution comment. You can't view a photograph or video at 1920?
no... it's not photorealistic, if it were, you wouldn't be able to see difference between a photo of the real things vs a screenshot in-game
getting close != there yet
Belgium's #1 Hardware Review Site and OC-Team!
photo realism in anything other than people/animals is 1000x easier. racing games will be the first to reach photo realism, everything else will be a decade behind. and due to just raw power needed, photo realism in a small group of people will be done quicker than something like an MMO.
color is more important.....
Dear audience, while we're excitingly waiting for the mystical Fermi, the topic-du-jour is photo realism gaming
It a very subjective question anyway (how do we measure it?). I'll stick to my guns tho! I'd say they haven't even nailed it in pre-rendered techniques yet (Avatar anyone?), but others would disagree.
Neither Fermi nor Cypress will do it, that's for sure!
The Fermi Paradox
source: http://en.wikipedia.org/wiki/Fermi_paradoxThe apparent size and age of the universe suggests that many technologically advanced Graphics Cards ought to exist.
However, this hypothesis seems inconsistent with the lack of observational evidence to support it.
If we see FERMI enter the consumer market before 2011, I will autophilate.
Haha
That was a good one. Thanks
i7 950@4.05Ghz HeatKiller 3.0
EVGA E762 EK WB | 12Gb OCZ3X1600LV6GK
Razer Tarantula |Razer Imperator | SB X-Fi PCIe
480GTX Tri SLi EK WBs | HAF X | Corsair AX1200
____________________________________________
Loop1: Double_MCP655(EK Dual Top) - MoRa3Pro_4x180 - HK3.0 - EKFB_E762
Loop2: Koolance_MCP655(EK Top) - HWLabsSR1_360 - EK_FC480GTX(3x)
Posting Permissions
Bookmarks