Thank you for your reply. For some inexplicable reason, I discovered that if
I re-compiled the opendx code (before I was using pre-compiled rpms) then
this error message no longer appeared and hardware rendering functioned. I am
unsure what exactly was the problem but I do not believe it is because I was
missing a component because even the sample programs did not function when I
tried to turn on hardware rendering.
In reference to your question about my data set, I simply imported an ascii
file where the data was in (x,y,z,data) column format. This was inefficient
so I was looking for documentation on the DX format. Would you happen to
know where I could find some?
Finally, (my apologies for bombarding you with a third question), but are
there any good tutorials for dealing with scattered 3D data. That is what I
am working with and I am having difficulty learning how I can do different
things on the data. For example, how do I 'slice' the data or ask it to plot
a (x,data) subsection? The 'slice' module seems to work only on gridded data.
Thank you for your help,
Elvis Dieguez
On Monday 06 August 2001 10:17 am, you wrote:
> Elvis Dieguez:
> |I am new to OpenDX and I have a question. I installed the latest version
> | on my Linux system (Mandrake 8.0 with kernel v2.4.6-ac2) that has a
> | GeForce2 NVIDIA card using the latest NVIDIA drivers (and their files
> | for OpenGL). When I try to use hardware rendering, I get the following
> | error message:
> |
> |ERROR: Image: Display: Missing data: #10240
> |
> |I have no idea what this means and I cannot find any reference to it in
> | the documentation I have seen. Can anyone help? Software rending seems
> | extremely slow on my system so I'd like to try hardware rendering.
>
> When rendering, DX expected to find a component attached to one of the data
> fields fed into Display, but it was missing (I agree it would be nice if
> the error message was a little more verbose about it, and told you which
> one).
>
> If you haven't tried it, the output of the Print("r") module should be
> useful in diagnosing the problem. Let's check that out.
>
> Also, are you building your own datasets in DX format by chance, or did
> standard DX modules generate this?
>
> Randy
