Hi-<br>Our group has an application in which we are reading in a series of volumetric medical images in DICOM format. We are displaying orthogonal slices and we wish to update the slices in a rapid manner. In the current implementation, the data for each of the 3D images is read into RAM. Even so, when we update it takes quite a long time for the display to change when switching between the various 3D datasets. <br>
<br>I am curious where the bottleneck in the system is. Does the video card in the machine need more onboard memory? In the final implementation, we would like to have 3-4 large CT datasets stored in memory and switch between the two at a few times per second. Any ideas on what we need to change in order to do so? Thanks in advance-<br>
-Ryan<br>