Canon EF-S 18-55mm f/3.5-5.6 IS unduly soft at 55mm?

I recently bought a Canon EF-S 18-55mm f/3.5-5.6 IS lens as my new standard walkabout after my Tamron 17-50mm f/2.8 caught fungus and died ( it was a lovely lens once it was calibrated and recentred). Anyway, I'd heard it was optically superior to its non-IS predecessor (which I also have). But at 55mm, I find this not to be the case.

The evidence: the following images were taken at the 55mm end of the lens. The subject was placed at a distance of about 80cm and lit by a fluorescent light, the camera mounted on a tripod, and IS turned off. The camera body is an EOS 400D that's been calibrated twice in its lifetime. The centre AF point was used, trained on the cross. The images are split down the middle: on the left is the old non-IS kit lens, the right is the new IS lens. (I also took test images with the EF 50mm f/1.4, these were all fine.) First, at f/5.6:

What colour is the cross mounted on (if it is white then i would say it is somthing to do with WB)... maybe.

Also was the pic taken with shutter trigger switch or by hand.

Randy McBum. wrote in post #7512068
What colour is the cross mounted on (if it is white then i would say it is somthing to do with WB)... maybe.

The background of the cross is white card (a skin cream box). These images were taken RAW with Standard picture mode and just processed to TIFF with DPP, without adjusting white balance. I could've adjusted it, but the slightly orangey tone is quite normal for the subject, light source, and the other bits of in the image that didn't make the crop. (The 50mm f/1.4's images were like this.) Looking at it now with my eyes, the camera didn't do a bad job in this respect.

Randy McBum. wrote in post #7512071
Also was the pic taken with shutter trigger switch or by hand.

IR remote (with 2 second countdown).

the only pic that is noticably worst for the new kit is the second.

Hmm... I think given that in the wider image (not shown here) no part of the picture within the depth of field is particularly sharp (whereas the older lens has a noticeable sharp in-focus zone). It's gotta be broken.

Lenses vary. Your old kit lens looks like it was a sharp copy, your new IS kit lens might be a soft copy. Worst case send it in to Canon.

I think I would try a differant test method before I condemed the lens. Perhaps some large sharp black text on a white background with good natural light and white balance set.

White balance is totally irrelevant in this test.

Do a formal focus accuracy test with the IS to see if there is a back-focus problem.

Are results reproducible?

MLU used? No? Shaky tripod/head?

xarqi wrote in post #7524303
Do a formal focus accuracy test with the IS to see if there is a back-focus problem.

Are results reproducible?

MLU used? No? Shaky tripod/head?

OK, I'll try tomorrow morning when I've got good daylight. (I have had a past experience where the light temperature affected the body's back-/front-focus tendency, now corrected.) I presume you mean like a 45-degree test on a test-chart? The results shown above are semi-formal test #3 after the first two rounds produced similar results. Playing around with the lens also hints at back-focus. (Also, I thought IS was not recommended with a tripod, or does that not apply to the newer mechanism?)

ettlz wrote in post #7524439
OK, I'll try tomorrow morning when I've got good daylight. (I have had a past experience where the light temperature affected the body's back-/front-focus tendency, now corrected.) I presume you mean like a 45-degree test on a test-chart? Also, I thought IS was not recommended with a tripod, or does that not apply to the newer mechanism?

White balances happens after capture, not before. No setting is going to change AF. That is not to say that the colour of the light cannot affect AF, but unless the light is quite unusual (e.g. illuminated by a blue laser), it shouldn't be an issue. Level of light might be a different matter, and within reason, the brighter the better.

Yes, one of those focus charts, or even the "battery" or "picket fence" tests.

I used "IS" as a means of distinguishing the lenses, not as a suggestion that it need be used, but it should be irrelevant anyway as this lens does have "tripod sensing".

I did a focus test chart test; I'll not post the results because it was glaringly obvious: only the new lens returned a blurry image. It's either soft, or back-focusing like mad at 55mm; I cannot tell. I then took some "ordinary" photos in the bright sunlight under identical shooting conditions with both lenses, hand-held, no unfairness: 55mm, f/5.6, 1/800s. Focus was on the top of the rock. Note that I took multiple shots with each lens, these are the best ones.

This is a 100% centre crop from the old kit lens's image:

xarqi wrote in post #7524473
I used "IS" as a means of distinguishing the lenses, not as a suggestion that it need be used, but it should be irrelevant anyway as this lens does have "tripod sensing".

Hmmmmm.....I *really* don't think the 18-55mm IS tripod sensing works very well at all. I know reviews say that it is tripod sensing but mine I get blurry shots on a tripod with it on...(the 55-250 is fine, tho!)

Try turning the IS off if it's going to be on a tripod.

This is at 55mm from mine...100% crop from center-ish area:

HTTP response: NOT FOUND | MIME changed to 'image/png'

ettlz wrote in post #7526580
I did a focus test chart test; I'll not post the results because it was glaringly obvious: only the new lens returned a blurry image. It's either soft, or back-focusing like mad at 55mm; I cannot tell.

That is what the focus test should have told you.

Ummmm - you don't have filters on these lenses do you?