John,
> Looks like you're saying this is a feature. My concerns with resampling
> the grid to make things cubic are that I'm going to pay a large memory
> penalty (maore than an order of magnitude greater in the case of the
> 61x61x5 example) and I risk aliasing the original grid. In addition,
> users of ncBrowse will be stuck with how I choose to resample the grid.
I wouldn't worry about the aliasing, since volume visualization
isn't very quantitative. Its just a colored fog.
You could use a test to only resample beyond a certain degree
of out-of-square. Also, give users a choice of other techniques,
such as interactive iso-surfaces, that they can use when volume
visualization isn't effective.
Also, I'd still be interested to see a JPEG or GIF of one of
your problem images.
Bill
