They render only one image.Just look outside the window with alternatively left and right eye,you might get a clue of what i mean.
The 1,5 percent overhead tells me they probably reconstruct missing data by just copying pixels.
3D Gaming*
- Thread starter Arwin
- Start date
They render only one image.Just look outside the window with alternatively left and right eye,you might get a clue of what i mean.
The 1,5 percent overhead tells me they probably reconstruct missing data by just copying pixels.
Well, there are speculations that say KZ3 can use depth-based approach because of the way Guerilla implement their deferred renderer.
At this point, I would be very happy if developers toy with as many techniques as possible so that we can find a lean and yet effective solution. May be Crysis's copy image approach (if that's what they are doing) is a good approximation except in some edge cases ? (They don't need to lower the resolution as a result. Might be a good trade off for some games).
KZ3 won't ship until 2011 whereas Crysis 2 is shipping this year right ? It may be problematic if KZ3 3D looks better than Crysis 2 today.
At this point, I would be very happy if developers toy with as many techniques as possible so that we can find a lean and yet effective solution. May be Crysis's copy image approach (if that's what they are doing) is a good approximation except in some edge cases ? (They don't need to lower the resolution as a result. Might be a good trade off for some games).
KZ3 won't ship until 2011 whereas Crysis 2 is shipping this year right ? It may be problematic if KZ3 3D looks better than Crysis 2 today.
That means you actually believe that the ps3 doesn't have the hardware to do MSAA, and also that you don't understand that the comparaison with stereo is weak at best because in the case of aliasing ,the whole data is there but in the case of 3d here ,half the data is missing.
Just a little off topic note... I told you all that 3D is going to be big
If Sony can gather enough customer interest, then this feature is going to be a standard on the next gen consoles. I can see even the next Wii introducing full support, especially as it can significantly enhance the motion control based experience...
If Sony can gather enough customer interest, then this feature is going to be a standard on the next gen consoles. I can see even the next Wii introducing full support, especially as it can significantly enhance the motion control based experience...
My point is it does not have to do it exactly as it it is meant to be done in other to be effective or better than existing method.
At the end of the day it is the effect that matters not the method used to achieve it.
Just a little off topic note... I told you all that 3D is going to be big
If Sony can gather enough customer interest, then this feature is going to be a standard on the next gen consoles. I can see even the next Wii introducing full support, especially as it can significantly enhance the motion control based experience...
Yeah , i think too that 3d will be the next big thing.
And it is nice that it will not unlike other games have blurry low SD resolution for 3D that will cause eye strain and probably eye problems in the longrun.
^^^ That's an interesting point of view. I wonder if this has "catches" when you keep on playing for a long time. Not that it happens that much, but I can remember a few times I got dizzy/sick just looking at a 3D game. So what's the result of extending that with actual 3D glasses. Headaches? Insanity? Technique is very cool though!
Repatched missing details and artifacts is not exactly pleasing to the eye (really) , too small paralax is neither and eye strain is caused by focus and convergeance being decoupled (a problem for any method using a flat surface to diplay perceivable depth).
Also Guerilla can still achieve 960 pixels horizontall resolution (640 is too radical).
Then i take the tradeoffs for real 3d and big paralax anyday other depth buffer trick...
Last edited by a moderator:
Other games actually do the work instead of faking it, and they especially don't call this clever or magical. B3D isn't the place to call such methods 'better', please don't start worshipping Crytek here for a PR trick.
So you know exactly their method used, have seen it and played Crysis 2?
Also please dont start doing personal attacks this aint GAF. There is no worshipping going but it is a bit ironic how fast their implementation is dissmised by some. Even if it doesn't deliver as good 3D and/or faults in some types of scenes then it still gives depth, has recieved positive comments by attending people at E3 and is not half-res blur fest.
Also next time read it all in context before going on a "with chunt" for your own personal reasons.
You don't have to see it to know it's inferior. Stereoscopic vision is all about two separate veiwpoints. Using one viewpoint and guessing the other viewpoint is inherently inferior, as it's missing data. It's in essence the difference between MLSS and 16x supersampling - one fakes the subpixel resolution and one actually determines the data at the subpixel level. Or that upscaling a 640p image to 1080p can create a better image than rendering 1080p natively. Any system that creates 3D from a single camera is not going to be as capable as a system that generates 3D from two cameras, and the only way the latter can appear worse is if the method has been poorly constructed. That or Joe Public thinks exaggerated depth of 2D cardboard cutouts looks better than proper 3D images, in much the same way blurry sub-HD content is more cinematic than HD!
grandmaster
Veteran
Surely history has told us there's no such thing as a free lunch in rendering... especially in a CryTek engine
What I find interesting is that they are referring to the effect as "concave 3D", ie. it doesn't pop out of the screen, you only get depth. This really does sound like exactly the effect Quaz51 is describing.
And without adding reprojection to render the missing bits, it's not true stereoscopy. If you walk up to a wall, which obscures the left eye but not the right, you're not going see what you should be seeing from the right eye.
I'm a bit annoyed I didn't get to see this at E3 as they were demoing on Xbox 360 - as they have for all the PR so far. Xbox 360 can't output 1280x1470. Certainly not yet, maybe not ever (if HANA is limited to 1080p max as the lack of 1920x1200 scaling suggests). So either CryTek were rendering both images from a 1080p framebuffer, or else they're doing exactly what Guerrilla are doing - using half-res 720p.
What I find interesting is that they are referring to the effect as "concave 3D", ie. it doesn't pop out of the screen, you only get depth. This really does sound like exactly the effect Quaz51 is describing.
And without adding reprojection to render the missing bits, it's not true stereoscopy. If you walk up to a wall, which obscures the left eye but not the right, you're not going see what you should be seeing from the right eye.
I'm a bit annoyed I didn't get to see this at E3 as they were demoing on Xbox 360 - as they have for all the PR so far. Xbox 360 can't output 1280x1470. Certainly not yet, maybe not ever (if HANA is limited to 1080p max as the lack of 1920x1200 scaling suggests). So either CryTek were rendering both images from a 1080p framebuffer, or else they're doing exactly what Guerrilla are doing - using half-res 720p.
Silent_Buddha
Legend
Just a little off topic note... I told you all that 3D is going to be big
If Sony can gather enough customer interest, then this feature is going to be a standard on the next gen consoles. I can see even the next Wii introducing full support, especially as it can significantly enhance the motion control based experience...
I still think we're at least 5 years away from 3D getting mainstream consumer adoption. And even then I'm not at all convinced it'll take off until display technology matures much more.
At best, I feel it's going to be a gimmick for late 2010 and 2011 and will start to peter out and die after that.
It's telling that the most compelling 3D that most people can agree is good is the 3DS. There's a lot of people that are unconvinced of the TV experience thus far. And the compromises in order to do it with current consoles on current and near future TV's is extreme, IMO.
Regards,
SB
grandmaster
Veteran
I wouldn't be too surprised to see 120MHz support standard in HDTVs in 18 months to two years, and from there surely you're just one HDMI controller and glasses interface away from a full 3D screen?
3D isn't going away. More movies are being made in 3D, more TV stations are adopting it and more games are being made. I find it hard to believe it's going to peter out. Indeed, I expect 3DS to fuel more of a thirst for 3D "content".
3DS really does live up to all the hype by the way. And in a sense it's a mini-preview of how 3D should be: no degradation of brightness, no glasses required, etc etc. The only thing that isn't so good is that the screen itself is very small. Would've liked it to have been PSP-sized at least.
3D isn't going away. More movies are being made in 3D, more TV stations are adopting it and more games are being made. I find it hard to believe it's going to peter out. Indeed, I expect 3DS to fuel more of a thirst for 3D "content".
3DS really does live up to all the hype by the way. And in a sense it's a mini-preview of how 3D should be: no degradation of brightness, no glasses required, etc etc. The only thing that isn't so good is that the screen itself is very small. Would've liked it to have been PSP-sized at least.
