Quote:Among the uncropped sample images are several taken at 600mm (taken with the 5D II). They are quite sharp but not pixel-level sharp (which is what I define as tack-sharp). Thus if they aren't pixel level sharp on the 5D II, they can hardly be great on the 5Ds R.
Just to mention - if we take the rating at 600mm f/6.3 we get an effective local resolution of ...
- 20.8mp in the center
- 5.7mp at the borders
- 4.8mp at the corners
That is, of course, on the 5Ds R. This figures will be smaller on the 5D II but at least the center will be more than decent here.
FWIW, Tamron's own MTFs do not suggest a great border/corner performance at 600mm.
It is also worth to mention that air diffusion has an impact due to the chart-to-camera distance - no, I do not test in vacuum conditions ... ;-)
As mentioned I will provide equivalent 21mp charts once I have a sufficient number of sample data for the conversion.
As a matter of interest what distances were used in the tests?
Yes lens testing without the use of air for breathing can make your head spin.
This sort of testing is know as the "suck it up method"
I was out shooting today using the D7100 with Tamron, the APSc sensor testing it's resolution to the limit. Coming home and looking at the images, I just had to scratch my head!......Have I got a really good copy or what?
Following on from those comments, I use almost always 1/1600 sec. at F11 with manual exposure at around 250-320 ISO, this generally underexposes by around a stop. The weather here...the sun is bright bright bright, contrast is ferocious and there are a lot of white birds.The sun today was fairly near to overhead, shadows being 1/2 meter. VC is always off!
Nikon's sensors are almost "ISO invariant" and it changes little raising the ISO to expose correctly...in fact the opposite , underexposing preserves the highlights and boosting exposure in post effectively produces the same final result...without any blowouts. It is rare that I can just balance exposure just with sliders, such is the ferocity of contrast, a typical image has about four or more adjustment brush pins with at least one exclusively eliminating sharpness in those frigging specular highlight onion balls.