View Single Post
Old 01-10-2020, 12:39 PM   #7
Senior Member
Decapod401's Avatar
Join Date: Nov 2014
Location: Lancaster, PA
Posts: 491

Originally Posted by TedG View Post
I'm not pursuing this any further--I can't--I've got no more chrome to work with. But, just out of curiosity, Doug, did you drop a guideline, or just eyeball it as far as level?

I think I leveled it based on the right side frame of the large bay window. Should it have instead been leveled based on the background building windows? (In which case even I can eyeball that it needs some CCW rotation.)

Leveling is a continuing source of frustration for me. So I appreciate your comments.
My assessment is very approximate, as I didn't put a grid on this - I just narrowed the window and compared verticals to the nearby edge.

Leveling is based on a reliable reference vertical in the center of the frame. In this case, that would be one of the stacks on the building. Often, there isn't a reliable vertical near the center, so an off-centered reference must be used.

In a perfect world, all vertical references would align to be perfectly vertical, but this almost never happens out of the camera, because of lens distortion. This can be addressed by adjusting the vertical distortion setting in your photo editing program. Between this setting and the rotation setting, you can get all references to be level. Initially, it is an iterative process, but I can now usually predict the distortion setting because most of my photos were taken through the same 50mm prime lens.

Although, rolling stock is not generally considered to be a reliable reference, a roster shot on level track can be used. In your photo, the caboose ribs can definitely be used for distortion correction, and also should be reliable vertical references, but always compare to other references (building elements in this case) before calling it good.
Doug Lilly

My RP Pics are HERE.

I've now got a Flickr. account, too.
Decapod401 is offline   Reply With Quote