Lens Testing

I recently purchased a Sigma 17-35 EX and need to do a better sharpness test than my initial "this is too soft, something must be wrong" reaction.

Any ideas of a good step-by-step method to test a new lens? There are probably several ways but what's your favorite?

When done, what should I expect or accept in terms of sharpness at, say, 17mm f2.8 versus 35mm f8.0? Is a huge difference acceptable?

So many people both love and hate this lens on the D30 that I want to be sure I really have a bad lens, and I look forward to learning something as well.

I went through the same reaction. first the results of my test.....

http://www.txis.com …EST/LensTest/le​nstest.htm

Basically all I did was set up my lights, subject, and camera settings. Took a picture with one lens, then switched lenses and took the other picture. Since both cameras were variable focal lengths I was slightly off on getting that part exact. Can't remember now if focal length is included in the INFO display or not, but I know it is included in the EXIF information.

Hope this helps.

A quick test to check a lens is to photograph a full newspaper page of small classified ads. This makes it easy to see if images are sharp to the edge of the frame.

Darryl

I hadn't considered a side-by-side but that is, after all, a conclusive test unless the test is flawed in favor of one lens over the other. I own the 28-135 IS and felt this would be as good a comparison as any.

So, today I mounted the D30 on a tripod and set AV 5.6 @35mm, ISO 100 with the Sigma 17-35 EX, RAW mode. Subject distance was about 40 feet. More EXIF appears with the images on Pbase if more is important. First, the Sigma 17-35 EX:

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

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

Oops wrote:
FAR SIDE: Great suggestion; I'll try this next for the edge-to-edge test unless everyone agrees "if the center doesn't cut it, the edges won't either".

Chris,

If the centre is soft, who cares if the edges are sharp? (I'm talking about LENSES here... )

From what I've seen so far, that Sigma of yours really, really sucks (sorry)! No other word for it. I cannot see how a chip upgrade by Sigma will improve the optics...

Well both are tests could probably be more "scientific", but I think the end results would be the same...

The Canon lens is better, period.

-darryl

Thank you all for responding and helping out! I have fired off a very nice e-mail to the camera shop and asked for a return authorization. I am still impressed by the praise so many people give this lens and would love to get a good one.

I'll ask them to test it first.

In parting, here's the "Blob" shot just for grins since I included this link for the camera shop:

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

The camera shop called me @ work and said I could return the Sigma for a full refund, apply it toward the 16-35L, or try another Sigma. My choice.

Rudi, I noticed you have a Tokina 19-35 with which you are very happy. If it is the AF193C listed on B&H for $189 with a $20 rebate this would be quite a deal.

Can I talk you into a simple lens test or a close crop example?

The 16-35 is tempting but @ $1600 you wouldn't want to make a mistake.

Chris,

Yes, the Tokina is really cheap! It is a slower lens, though (f/3.5-4.5), compared to the Canon 16-35 L. I am SURE that the L lens will be the better lens, but much pricier!

I will shoot a comparison between the 28-135 IS and the Tokina. Give me about 24 hours to post the photos. (I'm off to work now).

I lucked out with the Tokina. I knew a guy who had one, and I knew his photos were sharp, so I bought one for myself. The lens has a good reputation on the net, too, although I did not find this out until AFTER I bought mine...

Good news about your refund! Now all you have to decide is: Do you want that extra 3mm at the wide end (a BIG difference at that end!) for only $1400 extra?

Chris (and all),

here are my (not so scientific) results, comparing the 19-35mm Tokina cheapie to the Canon 28-135 IS lens. Please keep in mind that I was in a hurry and both shots were handheld. The following are crops from the full frame, at 100% magnification. You are seeing the actual pixels! (No downward resampling here).

My apologies for the large size.

Details: Both shot in RAW (not me! ) at 35mm f/5.6, 1/200 second, flash, handheld. Distance to subject: 1.6 m (about 5.25 feet). Converted to linear using LS342 using normal sharpening. Converted to JPEG at 7/12 (to speed up download). No other manipulation.

The first one was taken with the Tokina cheapie:

Byte size: ZERO | Content warning: NOT AN IMAGE

Byte size: ZERO | Content warning: NOT AN IMAGE

Rudi,

Great job! You are so reliable; wanna job?

This is an interesting comparison and the Tokina looks very good at this range. I don't think the tripod would make a large difference here with 1/200 sec unless other readers would like to see it.

I will bang out a text @ 5 to 6 feet tomorrow and try to get it posted tomorrow night.

Those of you who are happy with the Sigma 17-35 EX please don't be shy on this. I would love to see how this lens SHOULD perform as there are several happy owners on this forum.

For the record, I paid $616 for the Sigma. It was high compared to B&H but small businesses need to survive and most offer excellent service after the sale. I figure I paid almost 40% more than I could have so, if added to the $616, now I'm in the "L" range. (At least for my dream lens, the 28-70L.)

Stand by for the next shots and hopefully some other folks will jump in as well.

Chris

Chris,

I don't think a tripod would make much difference re: camera movement, but I still have a sneaking suspicion that *I* might have moved slightly after focusing with the Tokina (after the focus locked, I might have moved slightly).

All-in-all, I think the Tokina is bound to be softer than the Canon (because of the different focal range and the difficulty in making a wide-angle zoom such as that in the 35mm format), so the results above are probably very close to the truth. Let's not forget that there are people on this and other forums that think that the 16-35 L (and some with the 28-70 L) is soft compared to their other lenses!

As to the job offer, only if I can work my own hours, have a company house, a car, the just-about-to-be-released 1D replacement, and a willing assistant (Fern??).

As my bonus, after my first year, I'd like all my lenses to be upgraded to the "L" versions...

OK folks,

Just to further clear up (confuse?) the issue, I have shot some newsprint to test all my lenses. (Something I could never see myself ever doing, but there you go... ).

You can view the results of my test here:

http://www.users.bigpo​nd.com …emo/lenstest/le​nstest.htm

As I determined before, my Tokina 19-35 is slightly softer and less contrasty than my 28-135 IS at 35mm. Not bad though, considering the difference in price.

I tested all of my lenses. Not sure about how scientific the test is, or whether it is of any use to anyone here, but please have a look and let me know your thoughts...

I have tested 70-200L2.8, 35-350L and 100-400L IS

I have done a low light test on my L lenses indoors.


The test can be reviewed at my web, http://www.roine.nu and the raw files can be downloaded at http://www.roine.nu/le​nstest.zip or at a link near bottom of the page (under headline Bildspel)

The camera is a Canon D30.



Due two the debate on all sites I have compared them by using only my model light on my two Elinchrom 500 studio flashes against a newspaper. That to give some low light conditions.

I am aware that this will give some long expose time, eg up to several seconds. Se EXINFO at my web under each picture.

The first I saw was the wb issue, I used auto WB.

70-200L/2.8

---------------

zoomed too about 150 mm
CRW_9816 70-200L/2.8 at 2.8
CRW_9817 8
CRW_9818 16
CRW_9819 32

CRW_9820 100-400L IS at 5
CRW_9821 8
CRW_9822 32

CRW_9823 35-135L at 5
CRW_9824 8
CRW_9825 32

---------------

Zoomed at about 350mm
CRW_9826 70-200 with Solignor 1.7x teleconverter (showed 2.8)
CRW_9827 8
CRW_9828 32

CRW_9829 100-400 at 5.6
CRW_9830 at 8
CRW_9831 at 32

CRW_9832 35-350 at 5.6
CRW_9833 8
CRW_9834 32


The jpg files are processed in PS with auto contrast and unsharpen (150, 1, 0)


There are a summary jpg, raw from the camera, side by side, for the two zoom range at the bottom of the page.


--
roine@roine.nu / http://www.roine.nu

Rudi wrote:

As to the job offer, only if I can work my own hours, have a company house, a car, the just-about-to-be-released 1D replacement, and a willing assistant (Fern??).

As my bonus, after my first year, I'd like all my lenses to be upgraded to the "L" versions...

Sooo, what kinda *car* we talk'n 'bout? I drive a Dodge Cummins but I suspect this ain't yo-cup-o-tea.

My oldest son had his 29th birthday last night and I had to join the activities. I still burned the steaks while preparing these and got "the look" from the missus. I did the right thing: Log off, Smile, don't go back on-line for 24 hours.8) Works for me.

First, the 28-135 IS- same specs as yours:

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

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

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

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