Go Back   Steve's Digicams Forums > Digital SLR and Interchangeable Lens Cameras > Sony Alpha dSLR / Konica Minolta dSLR, Sony SLT

Reply
 
Thread Tools Search this Thread
Old Jul 20, 2010, 8:47 AM   #1
Member
 
Join Date: Apr 2006
Posts: 45
Default Corner softness issue- kit lens better than several others?

Hi.

I have an a100 and did some testing today with the three of my lenses that go wide (18mm or wider) since it seems a bit of a trend that the kit lens should be better in several areas than most lenses I purchase.

Today my beef is corner sharpness, specifically left corner sharpness.

These are three cropped and saved as .png sections of the left side of 100% sized photos at f/8 of the same scene:
1. kit (18-70)
2. Tamron 11-18mm f/4.5 - 5.6
3. Sigma 17-70 f/2.8-4.5 DC macro

Conclusions: "18mm" is rather variable, 'Sigmaish colors' isn't a myth, left corner sharpness is bad at 18mm for all three lenses, at f/8. (It got as sharp as it could at f/16 for all but the kit lens, which was significantly sharper, for this scene)

My query is: is this degree of corner softness normal at this f-stop (with these lenses)?

Edit. It should be noted that corner sharpness for the Tamron doesn't deteriorate when going to 11mm. Similiarly, the Sigma has no corner sharpness issues at 25+ mm.
I'm concentrating on left corner sharpness now as it seems the most pressing issue, though center sharpness for these photos goes like this (and I can post crops of the centers too if wished) : 1. Kit , 2-, 3-, 4-, 5-Tamron, Sigma.
Attached Images
   

Last edited by Lindinblade; Jul 20, 2010 at 9:08 AM.
Lindinblade is offline   Reply With Quote
Sponsored Links
Old Jul 20, 2010, 3:34 PM   #2
Senior Member
 
Join Date: Feb 2006
Location: California
Posts: 2,295
Default

Some of my best shots have been taken with the KM 18-70 Kit lens. It's not the best lens but with good composition and a little luck you can get some great shots.

Most of the time corner sharpness is not the most important thing.

Just takes practice.
__________________
A-mount
E-mount
Canon S95
lomitamike is offline   Reply With Quote
Old Jul 20, 2010, 4:02 PM   #3
Member
 
Join Date: Apr 2006
Posts: 45
Default

Hello, from what I've read the kit lens quality varies a bit. Maybe I got a good copy.
Though this amount of corner softness at apertures larger than f/11 is noticable even if you do not 'pixel peep', and renders the Tamron unusable at apertures larger than f/11, and the Sigma at focal lengths shorter than around 25mm, unless I feel like cropping a lot. What I'm wondering is if I simply got bad copies of these lenses or whether I was expecting too much?
I mean that I'd suspect the camera was dinged if it wasn't for the kit lens.

(At 50mm, the kit is equal to the 17-70 from what I've seen, and at 70mm the 17-70 is sharper. (Though take that observation for what it's worth: at 50mm, I can't tell a difference (in sharpness) between the kit, my Sigma 50mm prime, and the Sigma 17-70mm)).

Last edited by Lindinblade; Jul 20, 2010 at 4:18 PM.
Lindinblade is offline   Reply With Quote
Old Jul 20, 2010, 5:59 PM   #4
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,572
Default

From the test reports on those three lenses at SLRGear.com, I'd say:

The Sony has significant CA.
The Tamron has some CA.
The Sigma has little CA.

The Sony is more than a stop faster than the Tamron.
The Tamron starts off at f/5.6.
The Sigma is two stops faster than the Tamron.

At f/4.0, the Sony and Sigma are about equal.

At f/5.6 through f/11, the Sigma is sharper, with the Sony and Tamron about equal.

At f/16-f/22, the Tamron and Sigma are about equal, and the Sony is less sharp.

I can't tell much from the images you posted. There's no EXIF data, and it looks like ther's some variation in exposure. But it looks like the first and third are about equal, and the second is softer. I think I see some CA in the first, so I'm guesing that's the Sony.
__________________
  • The lens is the thing.
  • 'Full Frame' is the new 'Medium Format'.
  • "One good test is worth a thousand expert opinions." - Tex Johnston, Boeing 707 test pilot.
TCav is offline   Reply With Quote
Old Jul 20, 2010, 6:08 PM   #5
Member
 
Join Date: Apr 2006
Posts: 45
Default

I'm not regarding CA at the moment.
The images are in order, i.e kit, Tamron, Sigma. They're all f/8, and all taken with a tripod. (Perhaps the exposure times differ a tad, but that shouldn't matter much for the sharpness as tripod and 10 sec delay). The kit is sharper at the corners than either of the other two, and significantly sharper in the center. At f/16, the kit has improved some in regards to sharpness and the others quite a bit, at least in the corners.

Same order here, center parts:
Attached Images
   

Last edited by Lindinblade; Jul 20, 2010 at 6:13 PM.
Lindinblade is offline   Reply With Quote
Old Jul 20, 2010, 8:08 PM   #6
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,572
Default

From the images you posted, I'd call the Sony slightly sharper than the Sigma, with the Tamron clearly coming in third. But I still see some CA from the Sony, so I'd say that the image quality from the Sony and the Sigma were equal.
__________________
  • The lens is the thing.
  • 'Full Frame' is the new 'Medium Format'.
  • "One good test is worth a thousand expert opinions." - Tex Johnston, Boeing 707 test pilot.
TCav is offline   Reply With Quote
Old Jul 21, 2010, 2:11 AM   #7
Member
 
Join Date: Apr 2006
Posts: 45
Default

Something like that.

Still, it is a bit mysterious that two different lenses should display such heavy corner softness. (Though it isn't in exactly the same areas, so presumably it has nothing to do with the camera).

http://www.bobatkins.com/photography..._review_0.html and http://www.photozone.de/canon-eos/28...review?start=1 do not indicate that their copies of the Tamron 11-18 give images that are terribly soft in a major portion of the left side, regardless of focal length, at f/8 (not even at the largest aperture, really).

Nikonians on the Tamron 11-18:
"Wide open, the lens was very good in the center, but quite soft in the corners. Stopping down to f/8 improved matters a lot, and it then turned in excellent performance in both the center and corner"
Not even close in my case.

Photozone about the Sigma 17-70: "At f/8 even the extreme corners are very good though" (this at 17mm), which clearly doesn't rhyme with my copy.
They do go on to say that corner softness issues go away at around 24mm, which matches what I've found.

So some things indicate a small corner-sharpness issue for the Sigma at shorter focal lengths, same as for Tamron, but it should be a non-issue at f/8.

The camera has had a few knocks and bumps throughout the years, but that seems an unlikely culprit. The Tamron lens was dropped once some time ago, but shows no external marks, makes no rattling sounds when you gently shake it, and has no issue focusing or anything else. Don't know if that fall could have generated severe border softness, but I doubt it.

It might have been beneficial to try them with a different camera ,but lacking that option I can only conclude that the copies of these lenses are worse-than-average in this department. Maybe some un-cropped images could give an idea of how noticable this is?

Last edited by Lindinblade; Jul 21, 2010 at 2:36 AM.
Lindinblade is offline   Reply With Quote
Old Jul 21, 2010, 7:39 AM   #8
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,572
Default

I don't think changing cameras would affect the results. I'm surprised with the results of the Tamron, though.You might try something like the Focus Test Chart to see if you 've got a front-focus or backfocus problem, but I don't know how that would work with an ultra-wide lens.
__________________
  • The lens is the thing.
  • 'Full Frame' is the new 'Medium Format'.
  • "One good test is worth a thousand expert opinions." - Tex Johnston, Boeing 707 test pilot.
TCav is offline   Reply With Quote
Old Jul 21, 2010, 8:42 AM   #9
Member
 
Join Date: Apr 2006
Posts: 45
Default

I agree with you there, the issues are likely to be lens-bound.

Were it a matter of front- or backfocus though, it seems both edges of the photos would be affected, as we would have the plane of focus pushed back/forward a bit along the z-axis. It's just the left side with the Tamron, and the left and a bit of the bottom side with the Sigma.

I uploaded full pictures from a test - (flickr downsizes them automagically and after many complaints that they stripped embedded exif-data they seem to have discarded the exif-view page too. And Photobucket looks like myspace. Have to fix a more private hosting solution one of these days ) -together with crops of the lefthand, bottom side of the photos.
-> http://www.flickr.com/photos/3967767...7624425641185/ If you click one of the full-scene images and hit the 'all sizes' button, you'll notice the left side is always the bad guy.

Clearly the Sigma shouldn't be this bad? Would this be covered by the warranty? As it stands the largest aperture for anything beyond 1000*700px is f/16 (at shorter focal lengths than around 24mm.) (This goes for both of them, though the Tamron has no more warranty).
Additionally, as you can see the left 'corner' means approx. 1/3 of the image.

Edit (And yes, it's a fairly dull scene- I think most people in Sweden would call this 'forest' :/ )

Last edited by Lindinblade; Jul 21, 2010 at 8:51 AM.
Lindinblade is offline   Reply With Quote
Old Jul 21, 2010, 9:48 AM   #10
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

It's kind of hard to see what's going on with those downsized samples (and I think flickr does have the option of letting you keep the original size available to viewers, as I often see the "Original Size" available under "All Sizes" when looking at albums that some of members have posted links to. I don't use Flickr, so I don't know how you'd need to set it up for that feature.

But, if you're focusing on something 20 or 30 feet away, you'd expect anything much less than about 7 or 8 feet from the camera to start getting softer at f/5.6, even on the wide end of those lenses (just because of Depth of Field limitations). So, you may need to stop the aperture a lot before closer areas of the scene start becoming sharper, depending on what you were focusing on.

To see if you've got a centering defect of some type, set the camera up on a tripod and take a photo of something flat (like a brick wall), making sure the camera is perfectly square to the target (not pointed up, down, right or left). Then, see if all 4 corners are equally soft or not.
JimC is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 2:05 AM.