justruss Offline Upload & Sell: Off
|
p.1 #7 · p.1 #7 · Does megapixels influence when testing for lens decentering? | |
Short answer: Yes, higher resolution makes it easier to spot (in the way DavidBM describes).
Longer answer/thought: An instructive way (via simplification) to think about it is that decentering results in parts of the image being less sharp than other parts of the image-- whereas in a theoretically centered lens both parts should be just as sharp. How visible such a difference is comes down to a matter of thresholds.
Here's a way to think about it: consider sharpness/resolving power on a scale of 1-100 (again simplification), with 100 being the highest. Let's say a 12mp sensor + specific lens never gets the sharpness above 50 anywhere in the image. That's the best you can get no matter what. You have a moderately decentered lens, so maybe the unsharp side of the image registers a 47, and the sharp side a 50-- not very visible. But consider a 42mp sensor + the same lens, and it scores an 80 on our scale in the sharpest part of the image. The decentered part of the image then scores, say, a 53 (because of the way resolving power works, even the bad side gets somewhat sharper-- but the lens is a greater limiting factor), but the sharpest part of the lens scores an 80. The difference is MUCH more perceptible. The main difference is that the sharp part of the image is a lot sharper (resolves more), whereas the unsharp/decentered part of the image is primarily limited by the lens defect and hasn't changed all that much between systems.
Again, keep in mind this is a major simplification. And keep in mind the limits to where this will show up (re: DavidBM). But, if everything is a little blurry to begin with (low resolution sensor), the difference between the sharp and not-sharp sides of a decentered lens may not be all that noticeable. But if everything is extremely crisp and resolved, the sharp side is much sharper than the side limited by the decentering of the lens-- and the difference between the two is totally visible.
Here's another analogy: If you have a towel that is 50% saturated with water in the driest part, and 55% saturated with water in the wet part-- can you feel the difference by hand? How about if you have a towel that is only 20% saturated with water in the driest part-- and 85% saturated in the wettest part? It's a bit like that. Increasing system resolution (in this case, coming from a higher mp sensor) makes the baseline of the towel drier-- and decentering makes it wetter (or limits how dry the wet part can be).
|