Hi, Am 15.02.2011 10:07, schrieb Vishnu Viswanath: > Hi, > Myself and some of my friends are working on a project to play a video using > a cluster, in effect parallel > rendering. We would like to add a filter to the video and demonstrate the > rendering process. > We read in your blog about parallel routines implementation. But we are > actually very stuck with this project. > The idea which we got until now is that we need to slice up the video frames > and distribute it into the > cluster nodes to process. Processing includes adding the filter, essentially.
I'm pretty sure that this isn't a good idea. Slicing the video frame makes sense on multi-core machines (which share the same address space), but not for multiple machines. Passing slices to the nodes will probably take more time, than you gain by parallelizing. All approaches I know for render farms are frame-based, that means you pass single frames to the nodes. Now if you have a filter, which uses data from previous frames, you cannot even do that and must work scene based. > Can we get some guidance regarding developing an app built on top of gmerlin > API for this. We are going through > the code of 'cmdlineplayer' to understand the API. Is there some other way > to study the API and are we in the > right direction? > > Thanking you in advance > > -- > Vishnu Viswanath > Computer Science & Engineering 2007-2011 > Model Engineering College, Cochin > India Cool, been in Cochin in 2007 Burkhard ------------------------------------------------------------------------------ The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: Pinpoint memory and threading errors before they happen. Find and fix more than 250 security defects in the development cycle. Locate bottlenecks in serial and parallel code that limit performance. http://p.sf.net/sfu/intel-dev2devfeb _______________________________________________ Gmerlin-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/gmerlin-general
