[osg-users] Massive Point Cloud rendering

[Robert Osfield]

Hi Bruno,


On 14 April 2016 at 15:44, Bruno Oliveira <bruno.manata.oliveira at gmail.com>
wrote:

> Hello,
>
> I deal with huge point cloud data and I am rendering it the naive way (via
> VBO or display lists)
>
> However, I've seen software (e.g. Photoscan) do this via some kind of
> progressive loading (based probably on the distance to the points and point
> density). I think this is similar to pagedLOD but I've tried using that
> from OSG and the results are way worse (I created a .ive file from a .ply,
> then opened the .ive with the osgviewer but I didn't see any difference)
>
> Does osg support this in any way? Or do I have to implement on my own?
>

Support exactly what?  Points clouds?  Generation of paged point cloud
databases?

PagedLOD only works as well as the scene graph you set up with it, if it's
poorly balanced there it will perform badly.  If you set it up well then
you'll be able to handle terabyte datasets at 60Hz,

What you are doing wrong is something we can't assess as all you've said is
your tried to use PageLOD in unspecified way and it didn't work out.

The first thing you need to do is tell the community more about the problem
you are trying to solve and how you've gone about trying to solve it.

Details like what specifically is "huge", 10 points? 100 points?  10
million, 100 million? a billiion?

Second what type of hardware are you trying to run your system on?
Graphics hardware, CPU, memory available, OS.

What is the performance you are getting with the naive and PagedLOD
versions of your software?

The number of questions I'm having suggest is sign that you need to spend
far more time crafting your posts to the community so we know where you are
coming from. Remember we don't sit beside you, we don't get see you data,
or code, or know what background you have.

Robert.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openscenegraph.org/pipermail/osg-users-openscenegraph.org/attachments/20160414/5207c1b2/attachment-0003.htm>