If price was the major factor in getting one of the 5 lenses tested by Roger, as is the case much more often than size/weight/resolution and so forth, the Tamron 24-70 VC seems to be the best bang for your buck right now. I was hoping the new 24-70 f/4 IS would be much closer to the 24-70 f/2.8 II but seems to be just marginally better in resolution than the Tammy.

I will still hold on to my 24-105, and see if Sigma have something in this range coming up with OS.

RCicala wrote:
Blog post is up:

http://www.lensrentals.com/blog/2013/01/canon-24-70-f4-is-resolution-tests

I'm pretty ambivalent about it - good (not great) lens, higher price than I think appropriate.

thanks, excellent work!!

any chance you will also do these tests for all of them at f/5.6 and f/8?
a great many shots actually are taken stopped down and lenses do not all perform the same at f/8.

If it acts like a 24-70 II at f/5 and f/8 or if it acts like a 24-105 at f/5 and f/8 makes a big difference. Like the former and it does nothing much at all for someone who'd also want to use it for a lot of landscape shots over which they will be picky about, if like the former than the more purely landscape shooter may find it a savings over the 24-70 II.

vsg28 wrote:
If price was the major factor in getting one of the 5 lenses tested by Roger, as is the case much more often than size/weight/resolution and so forth, the Tamron 24-70 VC seems to be the best bang for your buck right now. I was hoping the new 24-70 f/4 IS would be much closer to the 24-70 f/2.8 II but seems to be just marginally better in resolution than the Tammy.

I will still hold on to my 24-105, and see if Sigma have something in this range coming up with OS.

keep in mind the 24-70 IS was only compared wide open while the 24-70 II and 24-70 VC slightly stopped down. If it pulls ahead of the tamron at f/5-f/8 then that would change things.

True, and I would love to see that data. I also handled the 24-70 f/4 IS in a local store and the macro function is a bad joke in my opinion.

RCicala wrote:
Jim, that's an excellent point about resolution loss with distortion correction. I'm going to see if I can figure out a way to do that.

I'd start with multiple images of the USAF resolution target core in edge and corner positions, and then compare computed resolution for pre- and post- distortion correction, using Canon lens profiles in CS PS6. That part would take time, but it could be done. OTOH, I haven't figured out how to get my hands on multiple copies of the lenses in question, without quickly wearing out my welcome at CPS Canada.

RCicala wrote:
I had a fellow today who'd bought a Nikon 16-35 f/4 VR (AKA the Nikon Fisheye Zoom) from us screaming about the horribly soft corners at 16mm. He sent it back, I tested it twice and it was superb. Turns out he was correcting that 6% barrel distortion and then evaluating the corners, which did, indeed, then suck .

Interesting example.

Thank you Roger!

I will try to do some f/5.6 testing next week. The rest of today, as is traditional, is being spent taking one of the new lenses apart

(Yes, of course, pictures are being taken)

RCicala wrote:
I will try to do some f/5.6 testing next week. The rest of today, as is traditional, is being spent taking one of the new lenses apart

(Yes, of course, pictures are being taken)

hah nice

anyway 5.6 is good but any chance for f/8 too, since landscape people really do care about that, even if just at 24mm (and I know their are fast prime shooters who just go on about how everything is the same at f/8 and who needs a fast prime when any old zoom works for landscape but comparing, say, 28-135 IS to 24-70 I/24 2.8/24-105 to 24 1.4 II/24 T&S/24-70 II that really is not true at all and all too often IMO, wide open or near wide open is the only thing ever discussed). I know it a TON of work though, it's easy to ask and tough to test. Personally I'd rather see f/8 than f/5.6 if it had to be one or the other.

(although f/8 real world usage can become tricky since field curvature can come into play as much as anything else too)

The distortion values at 24 mm surprise me in two regards: first that this new lens should have dramatically lower distortion than the EF 24-105mm f/4L IS USM (nice, but why would Canon aim at this in a lens designed to be compact, in an era when few care about distortion?), and second, that the EF 24-70mm f/2.8L II USM shows lower distortion than its predecessor.

Photozone says the EF 24-70mm f/2.8L II USM has slightly more distortion than the EF 24-70mm f/2.8L USM at 24 mm, which matches the visual impression from their grid photos. They also use Imatest.

The Photozone distortion tests are here and here.

Does this lens produce beautiful sunstars like the 24-70 II?

Wonder how the corner sharpness is at f/8-16

More testing on the macro feature would be nice to see it's actualy usefulness.

RCicala wrote:
I will try to do some f/5.6 testing next week. The rest of today, as is traditional, is being spent taking one of the new lenses apart

(Yes, of course, pictures are being taken)

Perfect test of the macro (or near-macro) capability of the lens!

The lens intrigues me, but being the owner of the original 24-70 and a 24-105, I don't think it will be on my list of things to buy. The 24-70/2.8 Mk II might though!

Wow - Canon finally produced a zoom I might be slightly interested in - I'd like to see it compared to the Contax / Zeiss 24-85 "N". The Zeiss has a much higher IQ than the 24-105 & slightly smaller and lighter than the new Canon.

At that size and weight with IS & if it has the punch of the 24II TS-E - might be a good lens.

This appears to be a ho-hum lens. My speculation is that Canon can produce this lens in quantity cheaper than the 24-105 and this seems to be the reason why Canon is slotting it in as a kit lens. I wouldn't be surprised if the return on kit sales with the 24-105 is not the best.

Canon won't be getting a sale from me and especially at the price they are asking.

One of the great things is that those who want a 24mm-whatever lens from Canon have three great choices that can all produce excellent image quality. With this being the case, one can make a choice based on other important functional factors and be assured of getting a fine lens no matter what.

Dan

Thanks for the info, Roger! I'm very thankful that we have you as a resource for this stuff.

Gunzorro wrote:
Kevindar -- I'm more optimistic. Canon has recently been putting out some great lenses, really improving their line-up (apparently preparing for higher MP bodies). I don't see how this new one can be compared to the old 24-105/4 IS or 70-200/4 IS. First of all, it has a much better IS -- personally, I love my 100L and the same sorts of arguments were made against it because the non-IS Macro was so well established and widely available used. Similarly the 24-105 -- it used to sell around $1200, but price has come down and the lens has sold so well that new versions battle against used and previous kit lenses, driving prices down further. The 70-200/4 L is no different -- its price is down from its premiere.

We've seen the new 24-70/2.8 II come down and some specials making it more practical for consumers. I expect the new 24-70/4 to follow the same trajectory -- high price at first, but probably around $1100 or less in a year.

The real story is that Canon has likely hit another optical and mechanical home run and that will trump the dissatisfaction over the price. . . eventually.

I'm anxious to read Roger's views and experience tomorrow. Fingers crossed!
I do not disagree with you. I am sure it is optically better. I said if its as good as the 70-200 f4 is (which I thing is an excellent lens) it will be worth about a grand, even though the 24-105 goes about 800 from the kit on the forum all the time. just not 1500 regardless of how good it is.

Roger for president! Thanks for all the work Roger. You're awesome!

Thanks for sharing these results.

Seems like the wide-open performance of 24-70 f/4 IS is similar if not better than wide-open performance of 24-70 f/2.8 MkII. I like the f/4 size and weight too.

The only worrying thing is decentered elements. Looks like I'll have some quality check to do when I purchase the kit.

Thanks for the tests Roger!
From the numbers, the Canon 24-70mm f/4L IS performed very well in regards to resolution, distortion and CA control. At 24mm, it performed way better than the 24-105L, and a hair better wide-open than the Tamron stepped down to f/4. It also held its own when compared to the superb 24-70 II, especially at the long end.
Where it really sets itself apart from the 24-105L is 3x less distortion at 24mm! Fixing barrel distortion in post, reduces resolution even further and that is reason enough for some to upgrade.

Fred Miranda wrote:
Fixing barrel distortion in post, reduces resolution even further and that is reason enough for some to upgrade.

It turns out that this concern is vastly over-rated, and that barrel distortion and other similar sorts of lens distortions can be corrected in post with typically no visible degradation of the image.
Dan