Hello Richard.

BOINC is not very suitable for POV for 2 good reasons.
1) Some POV tasks can't be run on separate clients if you use radiosity or a
macro that output datas.
Each node would do the rad (or macro) and sometimes it's the big part of an
image.

2) Privacy. Each node would get the source to be rendered.
Would you let your sources available to everyone ?
SETI for example is a public project, your images may not be public.

I think POV should evolve and have its core functions redefined to implement
a master/slaves architecture.
This way, a "master" console would dispatch its tasks to connected nodes,
including radiosity and macros.

Regards,
 Luc H.


"richard dot quadling at bandvulc.co.uk" <richard dot quadling at
bandvulc.co.uk> wrote:
> Hi.
>
> Is there anyone here working on getting POVRay available as a BOINC project?
>
> BOINC : Berkeley Open Infrastructure for Network Computing is available at
> http://boinc.berkeley.edu.
>
> BOINC basically allows multiple projects to run across thousands of
> computers and all the results correlated at a single point. Distributed
> processing basically.
>
> There is a set of projects already available for BOINC. Seti, Climate
> Prediction, etc.
>
> The old SETI@Home program has become a BOINC project.
>
> But, other projects can be developed.
>
> With the thousands of unused CPU cycles available, some of the MASSIVE
> rendering projects could be completed a LOT quicker by sharing the load
> around.
>
> I am a programmer, but have not worked in C for a LONG time, but I am sure
> that converting POVRay to a BOINC project would be an very interesting
> project.
>
> Especially as some of the animation rendering can take months to achieve at
> high resolutions.
>
> Regards,
>
> Richard Quadling.

Post a reply to this message