What area is considered the "edges" of the frame in lense tests?

[orc73]

Hi

Lense test talk about center, midframe and edge performance. 
So where is which area?
Shooting people in portrait mode and having the eyes in one third of them frame.
So what would that be? More far the eyes in full body

Also interesting a full frame tested lense, how would results apply to aps-c sensors?
Like a leica summicron (pre asph)90mm on a fuji x.
Best regards

[dwswager]

Lense test talk about center, midframe and edge performance. 
So where is which area?
Shooting people in portrait mode and having the eyes in one third of them frame.
So what would that be? More far the eyes in full body

Each test setup can have their own definitions based on the software setup used to read the data.  Typically, I think of center as a 1.5x crop out of the middle.  The borders tend to be the band outside that center box and corners all the way toward the corners.  When they say edges, they are typically talking about what happens at the extreme edge of the sensor.

Below is an image from Photozone's Lens Test FAQ that will give you an idea of what they are talking about:

Also interesting a full frame tested lense, how would results apply to aps-c sensors?
Like a leica summicron (pre asph)90mm on a fuji x.
Best regards

While there can still be funny interactions at the edges between different size sensors and lenses, typically you are getting the center sweet spot when you use a FX lens on a DX body.  This is the reason (along with knowing eventually I would get back to full frame)  I only ever purchased 1 DX (APS-C) lens and used FX lenses even when shooting the DX D300 and D7100 bodies.

[orc73]

thx dwswager, great to get that  info

[Slobodan Blagojevic]

Just to be specific:

[Telecaster]

Somewhat OT: One thing I've noticed via using cameras, like Sony's A7r, that accomodate precise focusing in the extreme corners is that corner "softness" is often a matter of field curvature rather than resolving ability. Such lenses, which may require stopping down with manual focus (central split-image or microprism) or D-SLR (no corner focusing points) cameras when the subject is well off-center, usually turn out to perform fine at wider apertures when the focusing point is placed precisely in the frame over the subject.

-Dave-

[kers]

Somewhat OT: One thing I've noticed via using cameras, like Sony's A7r, that accomodate precise focusing in the extreme corners is that corner "softness" is often a matter of field curvature rather than resolving ability. Such lenses, which may require stopping down with manual focus (central split-image or microprism) or D-SLR (no corner focusing points) cameras when the subject is well off-center, usually turn out to perform fine at wider apertures when the focusing point is placed precisely in the frame over the subject.
-Dave-

So the question is
: on these tests- do they refocus for the edges to get optimum sharpness...? i guess not but does anyone  know...?
This field curvature might come handy at some specific instances like interiors...
The Nikkor 24mm PCE is such a lens that does not have a flat field of sharpness... but shines on images taken inside.

[dwswager]

So the question is
: on these tests- do they refocus for the edges to get optimum sharpness...? i guess not but does anyone  know...?
This field curvature might come handy at some specific instances like interiors...
The Nikkor 24mm PCE is such a lens that does not have a flat field of sharpness... but shines on images taken inside.

Field curvature is a feature of most any lens, worse with wide angles.  Macro lenses tend to have this corrected out and are considered flat field lenses. 

I don't know if Photozone does or does not refocus for the corners.  However, you are measuring 2 different things here.  If field curvature is an issue then it does make sense to refocus the corners for measurements, but all that data needs reported.  It becomes a question of whether a lens makes a good normal lens versus a good macro lens.  A lens sharp in the corners when refocused might be a great normal lens even if it exhibits curvature, but might not be so great for closeup work where the curvature reduces sharpness in the corners.

[DeanChriss]

So the question is
: on these tests- do they refocus for the edges to get optimum sharpness...? i guess not but does anyone  know...?
...

I recently read somewhere, perhaps it was at photozone, that each test image (center middle, and corners) are separately focused specifically to prevent field curvature from affecting results of resolution tests. That makes perfect sense since resolving power is different than out of focus blur due to field curvature. They separately test for field curvature at longer distances. Again, I can't recall the specific site, but there's one (perhaps lensrentals?) that sometimes tests field curvature extensively.