TDP retests TS-E 24mm f/3.5L II on 5Ds R

[StudentOfLight]

The TS-E 24L II holds up remarkably well under 50MP.

The following link compare the TS-E 24L and the legendary 300/2.8L II, both lenses attached via 1.4xTC. The 300 is noticeably sharper in the centre of frame but the 24L shows it's strength in the corners of the 1.4xTC image (equivalent to mid-frame of full frame image)
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=486&Camera=979&Sample=0&FLI=1&API=2&LensComp=739&CameraComp=979&SampleComp=0&FLIComp=1&APIComp=3

The above comparison simulates resolving power under 2.9um pixels at least for the central portion of the frame. This equate to about 38MP on APS-C. These new generation Canon L lenses really pack quite an optical punch.

 

[raptor3x]

Seems like the lens may be slightly decentered or maybe the tilt isn't exactly level as the -12/+12 shifts look fairly different from each with regards to corner sharpness.

EDIT: The 17 TS-E shows a similar behavior but a bit more severe.

 

[StudentOfLight]

Seems like the lens may be slightly decentered or maybe the tilt isn't exactly level as the -12/+12 shifts look fairly different from each with regards to corner sharpness.

EDIT: The 17 TS-E shows a similar behavior but a bit more severe.

I think you misunderstand what the shifted crops represent.

The crop is always taken from a specific corner, (e.g. the bottom right corner).

A shift of -12 is shifting the lens 12mm towards the top-left corner of frame.
A shift of +12 is shifting towards the bottom right corner of frame.

The crops will look different because the -12 represents a portion closer to the centre of the image circle while +12 represents the very corner of the extended image circle.

 

[raptor3x]

Seems like the lens may be slightly decentered or maybe the tilt isn't exactly level as the -12/+12 shifts look fairly different from each with regards to corner sharpness.

EDIT: The 17 TS-E shows a similar behavior but a bit more severe.

I think you misunderstand what the shifted crops represent.

The crop is always taken from a specific corner, (e.g. the bottom right corner).

A shift of -12 is shifting the lens 12mm towards the top-left corner of frame.
A shift of +12 is shifting towards the bottom right corner of frame.

The crops will look different because the -12 represents a portion closer to the centre of the image circle while +12 represents the very corner of the extended image circle.

Ah, that makes sense. I wasn't thinking of where the "corner" actually sits within the imaging circle.

 

[Bdube]

Seems like the lens may be slightly decentered or maybe the tilt isn't exactly level as the -12/+12 shifts look fairly different from each with regards to corner sharpness.

EDIT: The 17 TS-E shows a similar behavior but a bit more severe.

This is entirely normal for tilt-shifts. The only lenses to not have much of these issues that I have seen are the schneider PC-TS lenses. All 20 24 TS-Es and 17 TS-EsI looked at had issues like this.

 

[StudentOfLight]

Seems like the lens may be slightly decentered or maybe the tilt isn't exactly level as the -12/+12 shifts look fairly different from each with regards to corner sharpness.

EDIT: The 17 TS-E shows a similar behavior but a bit more severe.

This is entirely normal for tilt-shifts. The only lenses to not have much of these issues that I have seen are the schneider PC-TS lenses. All 20 24 TS-Es and 17 TS-EsI looked at had issues like this.

Brandon, while tilt may be in effect the major difference is the location of the crop being taken in the image circle:

 

[Loswr]

Brandon, while tilt may be in effect the major difference is the location of the crop being taken in the image circle:

I would think the tests are done with vertical or horizontal shift (not diagonal) but the end result is the same, differential positioning relative to the edge of the image circle.