 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
http://www.youtube.com/watch?v=-_IKqjb9TRY&feature=colike
This is the Titanic recreated in Cryengine 3. Holy crap! And it's better than
what I can do in Povray...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 25.11.2011 05:18, schrieb jhu:
> http://www.youtube.com/watch?v=-_IKqjb9TRY&feature=colike
>
> This is the Titanic recreated in Cryengine 3. Holy crap! And it's better than
> what I can do in Povray...
Pretty impressive, yes, but did you notice how the mirrors (and the
reflections on the polished wood, for that matter) are all fake? (They
apparently actively try to avoid looking at them in the animation until
close to the end.)
Also note that this is, as far as understand, the collective work of
multiple persons.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
jhu <nomail@nomail> wrote:
> http://www.youtube.com/watch?v=-_IKqjb9TRY&feature=colike
> This is the Titanic recreated in Cryengine 3. Holy crap! And it's better than
> what I can do in Povray...
Actually it doesn't look like it would be too hard to render any of those
frames with POV-Ray if you had the models, textures and light settings.
It's just that POV-Ray wouldn't be able to render it nearly as fast (we
are probably talking about several orders of magnitude).
One thing where the real-time engine has an "unfair" advantage is that
the radiosity is pre-calculated, while POV-Ray has no direct support for
precalculating polygon-based radiosity. OTOH, if you also had the lightmaps,
I think it would be possible to kludge them to work with POV-Ray as well,
greatly speeding up the rendering (because then you wouldn't have to use
any kind of radiosity calculations at all).
Where POV-Ray could beat the engine in terms of accuracy is in refractions
and reflections from non-flat surfaces (and maybe even from flat ones).
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Fri, 25 Nov 2011 06:18:29 +0200, jhu <nomail@nomail> wrote:
> http://www.youtube.com/watch?v=-_IKqjb9TRY&feature=colike
>
> This is the Titanic recreated in Cryengine 3. Holy crap! And it's better
> than
> what I can do in Povray...
>
I'm sure that if you can get the same models you can do a render that
looks better with pov-ray.
It may take a couple of weeks to render though...
--
-Nekar Xenos-
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nekar Xenos <nek### [at] gmail com> wrote:
> I'm sure that if you can get the same models you can do a render that
> looks better with pov-ray.
> It may take a couple of weeks to render though...
Actually if you had the same models, textures and lightmaps as the
game engine has, it wouldn't take all that long to render. After all,
the scene consists of a bunch of meshes and lightmaps. That's very light
to raytrace.
If you wanted soft shadows, that would take a little longer, but not
a whole lot.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp <war### [at] tagpovrayorg> wrote:
> One thing where the real-time engine has an "unfair" advantage is that
> the radiosity is pre-calculated [...]
No it's not. Cryengine3 renders global illumination in real-time!
http://www.youtube.com/watch?v=FqeXuO2AlEE
http://www6.incrysis.com/Light_Propagation_Volumes.pdf
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Roman Reiner" <lim### [at] gmxde> wrote:
> Warp <war### [at] tagpovrayorg> wrote:
> > One thing where the real-time engine has an "unfair" advantage is that
> > the radiosity is pre-calculated [...]
>
> No it's not. Cryengine3 renders global illumination in real-time!
>
> http://www.youtube.com/watch?v=FqeXuO2AlEE
> http://www6.incrysis.com/Light_Propagation_Volumes.pdf
Oh, thanks. I was going to clear that up.
In any case, here's what I found:
http://freesdk.crydev.net/display/SDKDOC4/Static+vs.+Dynamic+Lighting
Scanline tech has come a long way thanks to massive investments from such a
thriving industry.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Roman Reiner <lim### [at] gmxde> wrote:
> Warp <war### [at] tagpovrayorg> wrote:
> > One thing where the real-time engine has an "unfair" advantage is that
> > the radiosity is pre-calculated [...]
> No it's not. Cryengine3 renders global illumination in real-time!
Since all the lighting in the titanic scene is static, calculating
lighting dynamically is an enormous waste of resources. That exact same
scene could be rendered just with static lightmaps. (Turning lights on
and off does not require dynamic lightmaps, it just requires switching
between alternate sets of precalculated lightmaps.)
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp <war### [at] tagpovrayorg> wrote:
> Roman Reiner <lim### [at] gmxde> wrote:
> > Warp <war### [at] tagpovrayorg> wrote:
> > > One thing where the real-time engine has an "unfair" advantage is that
> > > the radiosity is pre-calculated [...]
>
> > No it's not. Cryengine3 renders global illumination in real-time!
>
> Since all the lighting in the titanic scene is static, calculating
> lighting dynamically is an enormous waste of resources. That exact same
> scene could be rendered just with static lightmaps. (Turning lights on
> and off does not require dynamic lightmaps, it just requires switching
> between alternate sets of precalculated lightmaps.)
>
> --
> - Warp
I know. Still, the engine is capable of rendering global illumination in
real-time and that's what it does. The fact that it is able to do that makes me
doubt that there are any resources being wasted ;)
Though, I assume the calculation heavily relies on reusing samples from scene
parts that have not changed, thus saving calculation time.
Either that or deciding which parts to reuse is actually more time consuming
than just re-rendering everything from scratch each frame.
Anyway, it works, and it's real-time.
The point being: Whether the scene is static or not, there were no
pre-computations!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 26/11/2011 10:55 AM, Roman Reiner wrote:
> Warp<war### [at] tagpovrayorg> wrote:
>> One thing where the real-time engine has an "unfair" advantage is that
>> the radiosity is pre-calculated [...]
>
> No it's not. Cryengine3 renders global illumination in real-time!
I was under the impression that it only approximates GI using
screen-space methods.
> http://www6.incrysis.com/Light_Propagation_Volumes.pdf
...or not...
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
