p.2 #1 · 24-70/2.8L II field curvature (is this normal?)
Jeff, you looked at this what do you think? As far as I can see, it all looks soft but then this is what I see at 100%. All look about the same so no field curvature here.
I am putting up some dropbox samples of this image and the brighter images of the bracket set. They can be seen here. Feel free to take a look. This image will be super sharp printed at 20-30 after processing.
p.2 #2 · 24-70/2.8L II field curvature (is this normal?)
I'll check out he lighter full-res images, thanks for doing that, and for posting the initial crops. I agree with your observations from what little I can see, and certainly see nothing obvious like I've seen in my image, but looking for said effects in my landscape images was much harder than the closer-up image, in which it was blatantly obvious. The difference between f/5.6 and f/11 at those distances could also play into it, with the onset of diffraction making it even harder to see.
Roger Cicala thinks it could simply be that this lens is so much sharper than the Mark I that the field curvature wasn't as easy to see with the older lens. Replacement lens shows up today, I'll A:B them both, and if they perform similarly I guess I'm stuck with an $1800 tripod weight. I just can't believe all the rave reviews if this performance is normal.
If this field curvature is characteristic of this lens, maybe I'll have to start shooting at f/22 to soften things up a bit. lol
p.2 #3 · 24-70/2.8L II field curvature (is this normal?)
Here's the field curvature plot for the 24-70mm f/2.8 Mk II at 70mm at infinity. This is at f/8 but field curvature doesn't change shape much stopping down although the in-focus red area is much narrower, of course. Doing it at f/8 eliminates some of the other aberrations that look like field curvature at wider apertures. Field curvature can change a bit at closer focusing distances, though, and I can't measure that.
Please, before you go nuts, realize this is really pretty good. Almost every zoom has at least this amount of curvature until you get telephoto. The Mk I is actually a little worse. BUT, the Mk II is very sharp compared to the Mk I and off focus falls off more quickly, so it's probably more noticeable.
p.2 #4 · 24-70/2.8L II field curvature (is this normal?)
This lens got its stellar reputation shooting test charts at close distance...at further out and infinity, it obviously not a good lens. Many modern lenses from all manufacturers have to be judged away from the test charts.
p.2 #5 · 24-70/2.8L II field curvature (is this normal?)
RCicala wrote:
Here's the field curvature plot for the 24-70mm f/2.8 Mk II at 70mm at infinity. This is at f/8 but field curvature doesn't change shape much stopping down although the in-focus red area is much narrower, of course. Doing it at f/8 eliminates some of the other aberrations that look like field curvature at wider apertures. Field curvature can change a bit at closer focusing distances, though, and I can't measure that.
Please, before you go nuts, realize this is really pretty good. Almost every zoom has at least this amount of curvature until you get telephoto. The Mk I is actually a little worse. BUT, the Mk II is very sharp compared to the Mk I and off focus falls off more quickly, so it's probably more noticeable.
Thanks so much for that Roger, I appreciate you taking the time to throw that lens on the bench.
So, I *can* find a straight line to arrange a group with your lens that will produce acceptable results, And I can fathom that being typical for such a well-regarded lens. I guess my problem with *my* lens, then, is that my zone of acceptable focus at f/5.6 was probably 10 to 15% of what is shown by your red zone. Honestly, it's the kind of DOF I'd expect from f/2.0, not f/5.6, and I can't imagine the one-stop difference between your test and my image could account for that kind of razor-thin DOF (let alone my wider angle of view should be expected to produce a slightly wider DOF?).
I'm no lens expert, but I can't think of any optical reason that a lens could have a narrower DOF than it 'should' have...
Thanks again, and if you get a chance to test at wider angles next week, I'd be curious to see the results.
p.2 #6 · 24-70/2.8L II field curvature (is this normal?)
Jeff,
Spherical aberration will alter the depth of field. Undercorrected extends the rear DoF and shortens the front DoF, overcorrected is the opposite. Almost all lenses are undercorrected for spherical aberration, and at small aperture it is very small.
If you feel that the depth of field for your lens is not "working properly" and that is truthfully the case, there is either an error in technique or your copy of the lens has some issues. At 24mm the 24-70L II is decently corrected for field curvature, at 50mm the situation is better than at 24mm, and at 70mm the situation is a bit worse than at 24mm.
p.2 #7 · 24-70/2.8L II field curvature (is this normal?)
Thanks for that insight Brandon, I'm not an optics expert, but I've been shooting seriously since the early 80's, and I've never encountered a lens where, at f/5.6, you couldn't get a group of people standing in a line in at least reasonable focus.
My initial results with this lens found it to be front-focusing, and no amount of fiddling with correction in-camera would produce a *consistent* result; I think I settled on a correction factor of 1 unit. It now sounds like it very well could be over-corrected for spherical aberration, as you suggest, that could make some sense with at least some of the results I'm seeing.
p.2 #8 · 24-70/2.8L II field curvature (is this normal?)
Jeff wrote:
Thanks for that insight Brandon, I'm not an optics expert, but I've been shooting seriously since the early 80's, and I've never encountered a lens where, at f/5.6, you couldn't get a group of people standing in a line in at least reasonable focus.
My initial results with this lens found it to be front-focusing, and no amount of fiddling with correction in-camera would produce a *consistent* result; I think I settled on a correction factor of 1 unit. It now sounds like it very well could be over-corrected for spherical aberration, as you suggest, that could make some sense with at least some of the results I'm seeing.
p.2 #9 · 24-70/2.8L II field curvature (is this normal?)
dhphoto wrote:
The bottom-line is always the same, if you don't and won't ever be able to really trust it, get rid of it.
Just my 2c
Luckily I'm within 30 days, though the UPC from the box is gone, so I think I'm mostly stuck with this model of lens. Hopefully the second copy arriving today will be different…
p.2 #10 · 24-70/2.8L II field curvature (is this normal?)
Jeff wrote:
Luckily I'm within 30 days, though the UPC from the box is gone, so I think I'm mostly stuck with this model of lens. Hopefully the second copy arriving today will be different…
I haven't used the lens but had the V1 and the V2 is supposedly about as sharp as zooms get, so I hope your first one wasn't right and the second will be better.
My first 16-35IS was completely wrong so Canon QC isn't as good as it should be
p.2 #11 · 24-70/2.8L II field curvature (is this normal?)
Man, my 16-35IS is sweet, every bit as good as reviews say, and am thrilled with the lack of coma in corners for astro. Was expecting similar of the 24-70II.
p.2 #12 · 24-70/2.8L II field curvature (is this normal?)
dhphoto,
Canon's QC / manufacturing process is industry leading at this point in time with their newer models. Occasionally a dud will slip through, but this is true of all industries and all manufactures.
p.2 #13 · 24-70/2.8L II field curvature (is this normal?)
ben egbert wrote:
Jeff, you looked at this what do you think? As far as I can see, it all looks soft but then this is what I see at 100%. All look about the same so no field curvature here.
I am putting up some dropbox samples of this image and the brighter images of the bracket set. They can be seen here. Feel free to take a look. This image will be super sharp printed at 20-30 after processing.
p.2 #14 · 24-70/2.8L II field curvature (is this normal?)
Brandon Dube wrote:
dhphoto,
Canon's QC / manufacturing process is industry leading at this point in time with their newer models. Occasionally a dud will slip through, but this is true of all industries and all manufactures.
I'm just reporting what happened to me and I was disappointed
p.2 #15 · 24-70/2.8L II field curvature (is this normal?)
Here is an example I just grabbed from a recent set of landscape photographs where I happened to use this lens.
The basics: Canon EOS 5DsR, Canon EF 24-70mm f/2.8L IS at 50mm, ISO 100, 1/80 second, f/11. All images are 100% resolution crops except the first one showing the entire scene.
Manually focused on subject in roughly the center of the frame. F/11 for sufficient DOF, though with a conscious decision to allow the distant mountains to end up slightly soft. (Crops of that portion of the frame not included.)
p.2 #17 · 24-70/2.8L II field curvature (is this normal?)
Any 24/70 is really designed for close in focusing and that is where they will shine (of course they can be used for landscape). They are snapshot lenses. The 16/35s on the other hand are usually designed for infinity focus and usually excel at landscapes. Try comparing where a 24 mm lens excels compared to a 24/70 zoom, different lenses even though they overlap.
p.2 #18 · 24-70/2.8L II field curvature (is this normal?)
mogul wrote:
Any 24/70 is really designed for close in focusing and that is where they will shine (of course they can be used for landscape). They are snapshot lenses!
"Snapshot lenses?" Seriously? ;-)
And, of course the 16-35mm lenses (especially the f/4L IS, which I also use) is going going to be the best choice in the center of its focal length range at 24mm compared to a 24-70mm lens at the far edge of its range. Meanwhile, let me see if I can hunt up a 24mm example...
p.2 #19 · 24-70/2.8L II field curvature (is this normal?)
mogul wrote:
Any 24/70 is really designed for close in focusing and that is where they will shine (of course they can be used for landscape). They are snapshot lenses. The 16/35s on the other hand are usually designed for infinity focus and usually excel at landscapes. Try comparing where a 24 mm lens excels compared to a 24/70 zoom, different lenses even though they overlap.
When I bought my 24-70 f2.8 mk2, I owned a Ziess 50 f1.4 and 35 f1.4 and a 24TSE. I sold the three primes after testing the 24-70. And I use it 99% landscape.
p.2 #20 · 24-70/2.8L II field curvature (is this normal?)
ben egbert wrote:
When I bought my 24-70 f2.8 mk2, I owned a Ziess 50 f1.4 and 35 f1.4 and a 24TSE. I sold the three primes after testing the 24-70. And I use it 99% landscape.
.
Field curvature is not a problem on your lens and Canon let a bad lens get through?
