I suspect Silver Fox is happy to see sketching because it means actual rocks were observed not just data derived from rocks. It is very important that rocks are actually used in the study of geology. (You laugh, but my thesis area was mostly 500+ feet under ground. I started to miss actual rocks and get really sick of well logs.) But I still want discuss the computer stuff. Computer-generated illustrations fall into two categories in my mind - model/reconstruction "photos" and drawings. The sketches I draw would never be published, I'm not that good. But I can, if required, produces drawings in Canvas (or Illustrator or whatever) that are of publication quality. I like Canvas for cross-sections because I can scale them and keep track of line lengths and areas but for general drawing I guess Illustrator is better (partly because Adobe is taking over the world). "Pictures" of computer models can't be substituted for.
Screen shot from a reconstruction of the top Pratt cycle based on well logs in 3DMove (from my thesis work). Surface is colored by elevation (blue is deeper) CI=200 feet IIRC; points are colored by peak gas production (red higher/blue lower); approximately 5x VE, wells average 1/4 mile spacing (400m).
There is no better way to show structure and it's relationship to gas production (or water production, thickness, porosity, mineralogy...) than with pictures like this, IMHO. Structure contour maps, et al. are great. I've done them, lots of them. I've done some by hand. I've had to correct ones done by computers. But for those who don't visualize in 3D well, being able to "see" the rocks 1000feet below their feet is priceless. And the computer models allow for even those of us who can see the structure in our heads to drape other attributes over the structure and see the relationships better. Besides it makes for some really sexy figures.