[vtkusers] Is VTK suitable for Real-time Simulation?

[Sean McBride]

On 1/24/08 11:20 PM, Pengfei Huang said:

>I think if it is in the large computation case, our multi-core CPU is
>running full performance and memory also starves.
>
>What's more, I think VTK's point is its visualization (kind of
>static); on the other hand, we find its data structure update
>mechanism for animation rendering is slow. Sometimes we only modify a
>minor set of points, but we have to update the whole data.
>
>And if there are too many points/cells (e.g. 100,000 points), it is
>even very lag to do a small special rendering on some special points.
>
>Maybe our algorithm or system design has problem, but we believe that
>we haven't found the suitable method to accelerate VTK's speed on
>dynamic rendering.
>
>But we still look forward to improve the system by using some of VTK's
>convenient algorithms and some of other rendering engines for speed.

Well, I'll just repeat my former suggestion: use a profiler to _measure_
where you performance problems are, instead of guessing/assuming.  If
you're not familiar with profilers, see:
<http://en.wikipedia.org/wiki/Profiler_%28computer_science%29>

-- 
____________________________________________________________
Sean McBride, B. Eng                 sean at rogue-research.com
Rogue Research                        www.rogue-research.com 
Mac Software Developer              Montréal, Québec, Canada