23 September 2009
Okay, let's start with the first example of the 2-pt interior, Assignment#3. This one has the advantage of having "rotated objects," i.e., using more than one pair of VPs. It has the disadvantages of some sloppy linework and flat drawing and being, I think, overly photo-dependent and lacking much visual interest. The most serious issue is that it is not a 2 pt setup. Look at that back wall facing directly at us. It's a one point scene with rotated objects. Those being the farther counter, part of the ceiling structure. I think I gave this one a C.
The next is a redo. It has neater drawing, a more interesting--though far from lively-- scene. On the downside, the artist hasn't figured out what happens at the tops of the columns. Where do they contact and support the canopy? Where that happens, we should see the front line of an ellipse. The nearest column to the viewer magically shares floor space with the structure behind it (To check this, look at the base of the column and mentally complete the ellipse.) The stairway is poorly drafted, ignoring the right VP. The canopy over the far structure seems mysteriously to overlap the near one. This at least indicates that this is not a trace job (or perhaps that the original photo had a foreground column covering this area). On a good, compassionate day, I might just have given this one a B-, but it really should be C+.
For a better example refer to the last example below, which I don't have a copy of in its pre-toned state.
Next is the toned version of the redone interior. Whatever its other virtues, this one would rate a C-, as there is no real shadow plotting (and no overlay). The issue has been ducked by making every shadow soft-edged, and very likely just placing them wherever they were in the original photo. The original photo seems to have contributed positively to the rendering of the columns, at any rate.
Finally we have an A-plus example, which is not only ambitious (huge complexity) and attractive but largely successful. There is a second lightsource, the lamp at the right, whose influence has been drastically overplayed on the rounded chair and the woman in the center. The shadow of the wall of windows seems to have not taken into account the 3D depth of that wall; the cast shadows of the horizontal members seem too narrow (See carpet, tops of bookshelves). The artist should have plotted the shadows of the upper inner edge and the lower outer edge of the horizontal members. It appears he only plotted from the inner face of the wall.
(To determine which edges matter, put yourself, imaginatively, in the sun's place, looking down from on high. For you, Sunny, the contour, or silhouette, of a horizontal structure is defined by the far upper edge and the near lower edge. That's what makes the outline of the shadow. You never once get to see a shadow you've made, by the way. That seems unfair. Plus you feel hot all the time. Better lie down.)
Still, the diligence and detail shown, plus the nice balance of tones and the pleasing, convincing overall effect far outweigh the negatives.
Posted by Bink MMX at 11:10 AM
21 September 2009
This one was quite impressive. A lot of complexity, 95% of which was dealt with correctly. I think I gave it an "A."
But I have to point out that there is a mistake in plotting the shadow of the notched awning. That the shadow of that awning keeps angling all the way down the front of the building is incorrect. This would only happen if the sun were directly above the VP or nearly so. (It's not really a good idea to place the sun even this near the VP--we'll see why in a minute).
One way to deal with the shadow of the awning is to find the plan of the forward corners of the awning on the sidewalk and make sure that, where the d-lines hit the wall, they go straight up. This is the method I've emphasized in class, because it always works.
I attach a second piece of art--the grayer, close-up one--that shows another way. This time the d-lines are on the vertical surfaces. To make this work we need to establish an elevation of sunlight, at the height of the sun, directly above the VP. This will be the source of the d-lines. The thinner of the two blue d-lines is for the surface of the double doors only. Because the sun is so close to the VP, the light makes a very shallow angle across the door. Look, for instance, at the very darkest cast shadow on the door. That is the shadow caused just by the building’s face being a few inches forward of the doors! Yet it falls more than halfway to the left side of the door, meaning most of the entire doorway is in cast shadow. If this sort of surprisingly long shadow is not for you, then place your sun (or SVP) farther from the VP than in this example. (Notice the artist failed to reckon this accurately in the next doorway, which should have been about the same.)
The a-line and the bolder d-line show how I found the shadow of the lower front right corner of the awning on the building's face (yellow dot). To the left of that point, I made a Law-of-Parallels line coming from the VP. This is what was missing from the student's original version.
(The Plan of Sunlight being so close to the VP makes the shadow of the awning fall much farther down the door than the face of the building. Give a shadow farther to fall and it will, especially when it comes coasting in at a shallow angle like this)
Note: If you were to turn the drawing ninety degrees counter-clockwise, and essentially make the faces of the left buildings horizontal, the green line with the three green dots on it (which shows the clearance that the front of the awning has from the surface of the double door) could be viewed as the upright-stick-in-the-ground that was the first archetype I showed you re shadow plotting. With the drawing thus reoriented, the d-lines are running on "horizontal" surfaces as in that archetypal image.
Posted by Bink MMX at 12:47 PM