Hi everyone,

I have a txt file containing a list of camera positions at which I'd like to
render my scene at. The scene is an obj file converted to POVRay format (.pov,
..inc, .mat, .ini etc. using poseray) and there are around 60K tokens so it
spends a lot of time parsing this scene file when I send a render command for
instance

"
Render Stream to console........On
Statistics Stream to console....On
Warning Stream to console.......On
==== [Parsing...] ==========================================================
Parsing 18284K tokens

"

The image renders quickly after it is done parsing. I'm issuing a fresh
rendering request every time I render the scene at a different camera position.
Is there any possibility of rendering all the images within one rendering
instance without having to re-parse it every time? This would save me a huge
amount of time. I know animation can be helpful here but I don't know how to use
it together with the camera positions (numbered from 1 to N). If you have any
suggestions, it would be of immense help.

Thank you for your time.

Kind Regards,

Post a reply to this message

On 03.11.2014 22:37, handos wrote:

> Is there any possibility of rendering all the images within one rendering
> instance without having to re-parse it every time? This would save me a huge
> amount of time. I know animation can be helpful here but I don't know how to use
> it together with the camera positions

Even for animations, objects do not persist across frames. However,
UberPOV now features #persistent variables for this purpose, see

https://github.com/UberPOV/UberPOV

Post a reply to this message

> Hi everyone,
>
> I have a txt file containing a list of camera positions at which I'd like to
> render my scene at. The scene is an obj file converted to POVRay format (.pov,
> ..inc, .mat, .ini etc. using poseray) and there are around 60K tokens so it
> spends a lot of time parsing this scene file when I send a render command for
> instance
>
> "
> Render Stream to console........On
> Statistics Stream to console....On
> Warning Stream to console.......On
> ==== [Parsing...] ==========================================================
> Parsing 18284K tokens
>
> "
>
> The image renders quickly after it is done parsing. I'm issuing a fresh
> rendering request every time I render the scene at a different camera position.
> Is there any possibility of rendering all the images within one rendering
> instance without having to re-parse it every time? This would save me a huge
> amount of time. I know animation can be helpful here but I don't know how to use
> it together with the camera positions (numbered from 1 to N). If you have any
> suggestions, it would be of immense help.
>
> Thank you for your time.
>
> Kind Regards,
>
>
>

For an animation using the defined camera positions, you have a few options:

1) In a loop using frame_number as a controll value, read your 
positions, stopping at the desired one.

2) Read all the positions at once and place them in an array, say, 
Cam_Pos[xx]. Next, use a specific element:
camera{location Cam_Pos[frame_number] look_at 0}

By the way, 60K tokens is not even large. I have scenes with a tokens 
count 2 order of magnitude larger... And the source is less that 4Kb. 
The joy of loops.



Alain

Post a reply to this message

Le 05/11/2014 02:58, Alain a écrit :

> By the way, 60K tokens is not even large. I have scenes with a tokens
> count 2 order of magnitude larger... And the source is less that 4Kb.
> The joy of loops.
> 

60k token is not large, but if it is a single mesh, it can be slow near
the end. (whereas a mesh2 of the same object would not slow as parsing
progresses).


-- 
Just because nobody complains does not mean all parachutes are perfect.

Post a reply to this message