Testing lenses: The "bag of primes" vs 35/2 and XC50-230

Started 2 months ago | Photos thread
ForumParentFirstPrevious
Flat view
norjens
norjens Regular Member • Posts: 311
Testing lenses: The "bag of primes" vs 35/2 and XC50-230
4

For the sake of science, and after perceiving occasional smearing in winter conditions with the XF16-55, I printed an ISO 12233 chart to compare it with two other lenses.

Samples and link to full set below, here are my conclusions:

1) The 16-55 is good. Really good. My concerns about having a mediocre copy are dispelled, although it's not perfect at all settings.

2) Over view of performance through its focal range:
Short end: Barrel distortion and significant chromatic aberration (CA) at 16mm.
Mid point: Perfection at 27mm.
Long end: Pincushion distortion and CA at 55mm. Vignetting and corner softness, which gets better when stopped down.

3) The 16-55mm beats the 35mm prime at similar focal length and aperture. At 27mm f/2.8, resolution is indistinguishable from the 35mm prime. The prime suffers from CA, so overall the zoom looks better. Same story at f/7.1. Capture One can correct some of the CA, but not all of it.

4) While 55mm corners is the softest part for the 16-55mm, the XC50-230 is still a bit softer at f/4.5 than the 16-55 is at f/2.8. Both benefit from stopping down, with the 16-55 gaining the most as it reaches really good performance at f/7.1.

5) In the center, resolution seemed to be limited more by pixel density than lens sharpness with all of these three even at maximum aperture, although some lens differences were visible. Towards the edges, resolution appears on average to be about equally limited by pixel density and lens sharpness.

Samples:

Note that the loupe and '100%' viewing on the website here doesn't zoom in quite far enough to really see the differences. Downloadable files, full set of 14 with RAWs are here:
https://www.dropbox.com/sh/nmdzvk56y858ihn/AABx8VZ6pVRcE9JMtw8foke3a?dl=0

Tests were made on a solid tripod at minimum height with 2 second timer, so motion blur should be zero even for exposures near 1s. JPEG files were developed in Capture One 12, all corrections turned off and default sharpening of 140 set down to 100 (effective zero, if I understand it correctly). The chart print was from pdf on A3 from a large office printer, but not photo printer. It is sharper to the eye than in the captured files, so it was not a bottleneck.

ForumParentFirstPrevious
Flat view
ForumParentFirstPrevious
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow