Fellow DRT members, please help..
I'm pasteing here what I posted over on OM in a question to Jeddyih.
I'm sorry it is a long rambling post asking the same question tediously..
Here it is:
Where you see a shadow inconsistency on the pole, I see many as per my example of the pole vrs his illustration.
How can his illustration be correct if the shadows on the pole have so many inconsistencies to his illustration. I think his work and time is commendable, however, dragon says and I agree with him, to determine the correct angle of the sun, the shadow on the side of the pole would help determine angle.
Where there are shadows or shadows missing between his illustration of the pole and the original picture, is an indication that there is something wrong with the angle established by Torvald. Unless the angle of light is as exactly as it is in the pole, how can you begin to determine what is missing off the drone in the picture.
The only other possibility I see to resolve this shadow discrepency is that the pole is fake, it's shadows are fake and then the drone was dropped in as well. Now that solution would in my mind be ridiculous. Why create a cgi composition of a telephone pole when a photograph would work just as well, per this original picture. But to get the exact conditions to determine what is happening on the shadows of the drone, you have to first get the shadows exactly replicated on the pole.
Reflective light and the elevation of the pole has left certain shadows out of areas on the original pole. These shadows are in Torvalds illustration, and he says certain shadows are missing from the drone.
Sorry, but it doesn't make sense to me how you can attach importance to missing shadows on the drone when you can't provide a replication of the shadows on the pole. If you can get the shadows on the pole with its ambient lighting replicated to the point where it matches the shadows on the photographed pole spot on, then say the drone is missing shadows, I would say that is a cause for determining something is wrong with the shadows of the drone.. but until the shadows can be matched, it says to me it is not an accuate means of identifying missing shadows.
Please help me see where I may be wrong here in my reasoning.
Here is my example I made yesterday to show these inconsistent shadows:
http://img504.imageshack.us/img504/8172/torvaldexp1ts1.jpgThanks for anyones help here.