Sigma AF 10-20mm 4.0-5.6 EX DC HSM: First encounter

The place to discuss third party lenses which aren't covered by the sections above

Moderators: Bob Andersson, Thomas, Bjorn van Sinttruije

Sigma AF 10-20mm 4.0-5.6 EX DC HSM: First encounter

Postby Thomas » Sun Jan 14, 2007 11:58 pm

If you do have, please post your experience here.
The AF 10-20mm 4.0-5.6 EX DC HSM is said to be an excellent lens IF it is well adjusted!
And it's also very interesting as the closest Nikon lens is just the AF-S DX 12-24mm 4G IF-ED.
The Canonites have it just much better with the Canon EF-S 10-22mm 3.5–4.5 USM
Any feedback on this?
Last edited by Thomas on Thu Sep 20, 2007 9:32 pm, edited 3 times in total.
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

Postby Thomas » Tue Feb 20, 2007 7:57 pm

Now I waited for so long, and the trip to NZ is coming near, I had to buy this lens and not curse myself afterwards that I should have...
So expect a user report plus some flickr-pics (tagged as "sigma10-20") perhaps in addition to the test that Gordon promised...
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

test results Sigma 10-20mm

Postby Thomas » Wed Feb 21, 2007 6:09 pm

Oh boy :cry:
My lens arrived, it is made in Japan.
I went out shortly after sunset and shot some bare trees against the darkening sky. I find that the fine twigs make for a fantastic test of sharpness and resolution. I shot the same view twice, but turned the camera 180 degrees.
1. 100% crop from the right side of the first shot:
Image

:idea: then i turned the camera 180 degrees :idea:

2. 100% crop from the left side of the second shot:
Image

These are selections from the middle of the sides (not the corners!). You can easily see the difference between the sharpness of the left and the right side. Both pics were shot at ISO 200, F4, 10mm, 1/30 s.
Unfortunately both pics show a very different sharpness in one side of the pic. So I have to conclude that my brand new Sigma 10-20mm suffers from bad Sigma quality control. I will have to send it in for service! But now I will first take it to NZ, because the perspective and the fov is incredible with this lens.

Here you see a complete pic with the shadow of the built in flash on the floor - I didn't use a lens shade. http://www.flickr.com/photos/38912116@N00/397874069/
Best to avoid by turning the camera upside down, because the shadow would not show up in the sky :wink:
Last edited by Thomas on Sat Sep 08, 2007 2:30 pm, edited 10 times in total.
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

other observations

Postby Thomas » Wed Feb 21, 2007 10:31 pm

The build-quality feels ok.
The focus-ring moves smothly, the zoom-ring is a little thight (I wish the Nikkor 18-200 had such a tight zoom-ring to keep it from creeping...)
Focussing is fast and quiet, under low-light conditions with low contrast objects it hunts (as would almost every lens).

Magnification is 1:11 at 10mm and 1:6.4 at 20mm and no: they don't show any depth of field (dof) markers on the lens.
But don't let you be misguided: if you go close to the subject this lens has a limited dof: Not everything from 24cm to infinity is sharp at f4.
Close-up sharpness (at 24cm) seems to be good at 20mm. At 10mm I'm not sure, maybe it was a focus problem...

The back lens cap is weird: it doesn't fit on my Nikkors, whereas the Nikon back-cap fits on the Sigma. So you have to be careful which lens cap you've got. Plus it can only be attached in one position, not the three that you come to expect from a Nikkor cap!
The front lens cap is also crap: You can hardly attach/detach it, when the lens hood is mounted :? :shock: :(

There is also an odd thing about the focus scale: Object that are 50cm away show a focus distance of 1m. Objects that are merely< 2m away show infinity. But who cares what's on the distance scale?

And as you can see here: http://www.flickr.com/photos/38912116@N00/398100973/ the built-in flash of my Nikon D80 can never illuminate everything when the Sigma is zoomed out to 10mm - which was to be expected.
Last edited by Thomas on Thu Feb 22, 2007 9:54 pm, edited 6 times in total.
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

Postby Gordon Laing » Thu Feb 22, 2007 12:46 am

Hi Thomas, thanks for your detailed report.

The softness on one side of the frame at 10mm is a concern, but sadly it's not that unusual - I've tested many lenses which suffer from something similar. It'll be interesting to see if my forthcoming Sigma 10-20mm sample suffers from the same effect.

Are you happy with the lens though? Did you go for it over the Nikkor ultra-wide because of the lower price and wider focal length?

Gordon
User avatar
Gordon Laing
Site Admin
 
Posts: 9978
Joined: Tue Nov 08, 2005 3:32 pm
Location: Queenstown, New Zealand

Postby Thomas » Thu Feb 22, 2007 7:06 pm

Well: Unfortunately, Gordon, you haven't test the Nikkor 12-24mm yet!
I have seen some good test results, but they were not enough motivation to shell out more than double the dough and receive 20% less (in wide-angle that is)! I would never invest in a 12-Xmm when I already have a 18-Xmm - that would be only 33% better!

I knew the optical quality of the Sigma was shifting from sample to sample but also heard that it could be corrected by Nikon-service.
As long as one side is sharp enough I think the design of the lens is good, only the quality control is sloppy.

And, yes, the 10mm is breathtaking!

I will take some more test shots at the weekend, trying to decide whether I would keep the lens even if the service cannot improve the lens adjustment.
Last edited by Thomas on Fri Apr 08, 2011 9:10 pm, edited 3 times in total.
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

contrast, ghosting, flare

Postby Thomas » Thu Feb 22, 2007 9:38 pm

Just compared the Nikkor 18-200 at 18mm with the Sigma 10-20 at 20mm.
Under normal lighting conditions the centre sharpness of both lenses seems comparable.
BUT if you shine some flashlight directly into the lens (same angle for both) the Sigma produces a completely milky pic: http://www.flickr.com/photos/38912116@N00/399103185/
Whereas the Nikon has at least some areas where the visibility is not bad: http://www.flickr.com/photos/38912116@N00/399102059/
Both pics were taken without lens hoods to simulate the worst case...

It looks less and less likely that I keep the Sigma :cry:
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

Postby Thomas » Sat Feb 24, 2007 3:28 pm

This morning I took the lens out with a little more light. You can find more photos under http://www.flickr.com/photos/38912116@N ... sigma1020/
The sharpness on the right side improved with an aperture of f8 and smaller.
With respect to flare, you can see here: http://www.flickr.com/photos/38912116@N00/400719690/ that under less than radical circumstances the contrast remains ok, only one small flare can be seen. The contrast in this picture goes easily from (2,2,2) to (255,254,255) R,G,B
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

Postby Thomas » Sun Feb 25, 2007 8:38 pm

Now I have made a couple of shots under controlled conditions (all pipcs @ 10mm, f4). I took our tiled floor as a means to adjust camera, switched focusing to manual mode and adjusted it in small steps. Than I swivelled the camera to the left, repeated everything, then to the right. Here are the results (distance left=right=75cm, sharpness inspected @ 100% crop, test objects identical on the left and right side):
My Sigma is sharp on the right side an additional 20-30cm further away than on the left. If I focus @ 75cm, the left object is in focus, while on the right side it would be in focus at a distance of 1m, but @ 75cm distance it is clearly not in focus.

One could compensate for that by turning the camera a little bit to the right (I assume around 10°). By that the right focal point (which has been to far away) comes a little closer and the left focal point moves a little to the back so that the objects on both sides come into focus. Only that the left object seems a little larger than the right. Oh well, we don't want to be nitpicking...

That can be taken into account while shooting, but honestly: My head spins, when I have to adapt this calculation to "infinity", because then the error should not be compensated by slightly turning the camera to the right. Infinity is infinity, isn't it?
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

Postby Thomas » Tue Feb 27, 2007 9:52 pm

48 hours later...
I've got my hands on a Nikkor 10.5mm/f2.8 fisheye to compare it to the Sigma and decide which one to take to New Zealand!
In the centre of the pic, both lenses are quite comparable (the Sigma at 10mm, that is). Plus Nikons software CaptureNX has an option to straighten the curved fisheye lines (if you shoot in RAW mode).
I have a lot of observations that I'll write down in the Nikon segment of this forum. But with respect to the Sigma I can say as much:
:!: In addition to the visible softness on one side, my Sigma has a focussing problem :!:
I just made a series of shots under controlled conditions (same subject, Lighting, etc. Both lenses set to 10mm/F4. Focus spot on the same subject (single point focussing). While the Nikkor hit the sharp spot dead on, the Sigma was off to an amount, that resulted in general softness in the complete picture, that could clearly be seen even at only 50% magnification.
When I focussed manually, the picture quality of the Sigma improved (in the center) but still didn't reach the Nikkor. And that after defishing the Nikkor and a little magnification to bring it up to the same magnification.

So in addition to the on-sided effect of the bad adjustment, there are problems with bad focussing and that kicks the Sigma out of my lens collection. That may come from the smaller aperture of the Sigma (4.0 vs 2.8 ). Whatever, I don't believe that the sloppy focussing can be corrected by the Sigma service.
That's very unfortunate for the Sigma, because under ideal conditions it's pretty sharp and well CA-corrected. But without reliable focussing that's all for naught...
Last edited by Thomas on Sat Mar 03, 2007 10:00 am, edited 3 times in total.
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

Postby Gordon Laing » Tue Feb 27, 2007 10:38 pm

Hi Thomas, thanks for your detailed reports on your Sigma lens.

It'll be interesting to see how your Nikkor fisheye compares, and I think many of us would like to see a before and after shot with the CaptureNX correction. If you do a posting in the Nikon forum about it, be sure to publish the link in this thread.

As you know, a fixed focal length lens should outperform almost any zoom, but the big question is whether you'd miss the zooming facility. Personally speaking, I always seem to use my zoom lenses either zoomed all the way out or all the way in, so perhaps fixed focal length models would have been better!

Gordon
User avatar
Gordon Laing
Site Admin
 
Posts: 9978
Joined: Tue Nov 08, 2005 3:32 pm
Location: Queenstown, New Zealand

Postby Thomas » Tue Feb 27, 2007 11:08 pm

If you're interested in the Nikkor fisheye, continue here: http://www.cameralabs.com/forum/viewtopic.php?t=248
Well, Gordon, with respect to using a zoom: I found out that 80% of the Sigma 10-20mm shots I made were in the 10mm position...
So, you're right! With these zooms the questions is, what you really need.
If you have a standard lens like a 18-xxx, the need for extreme wide-angle zoom is not so urgent.
The major issue that I found is the following: If your zoom is not sharp dead on, you're better off to take the fixed lenght lens that is sharp and do some postprocessing to get to the right crop/magnification of the pic.

Addendum: I sent the Sigma lens back today and will keep the Nikkor 10.5mm fisheye :)
See both contenders here:
Image
http://www.flickr.com/photos/38912116@N00/407189084/
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

Postby AndrewC » Wed Aug 01, 2007 9:09 pm

Its a pity you scored a bad sample of the lens, my local retailer lent me his own to try before buying, I was very impressed with it. The lens that I purchased actually gained a little edge sharpness over the version that I was loaned and to date I have only found 1 very small amount of CA in one shot from the lens and that was at the extreme edge in a very contrasty situation that would have produced CA in any other lens I own. Stopping down to f8 at 10mm is about the optimum for sharpness and seems to suit the wide angle aspect by adding more DOF. At under 1/2 of the price of the Nikon 12-24 here in Australia and the 12-24 having a bad reputation for CA , it was a logical choice.
AndrewC
 

Postby Thomas » Wed Aug 01, 2007 9:16 pm

Nice to hear that you have a good sample!
The CAs measured by Klaus Schroiff for the Nikon 12-24mm are

Image

and for the Sigma 10-20mm are
Image

(c) www.Photozone.de
So the Nikkor is indeed worse than the Sigma at open aperture but a bit better at smaller apertures.

If you read through the great comparison at http://www.kenrockwell.com/tech/digital ... arison.htm you'll see, that the Nikon is better in contra-light with regard to contrast, flaring and ghosting.
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D810+assorted lenses
User avatar
Thomas
Moderator
 
Posts: 8003
Joined: Sun Nov 05, 2006 11:08 pm
Location: Germany

Postby AndrewC » Wed Aug 01, 2007 9:29 pm

Sorry tombomba, I should have been clearer, I was relying on real life shots, not lab test charts, I deliberately tried to induce CA in the lens, both as NEF shots and JPG, out of the 200 or so that I attempted to produce bad results with, I managed one. The Nikon on the other hand CONSISTENTLY produces shots with noticeable CA, that comes from a person who has had a 12-24 Nikon sent back to Nikon 3 times and replaced TWICE due to both CA and focus issues.
AndrewC
 

Next

Return to Third party lenses: Sigma, Tamron, Tokina, Zeiss etc

Who is online

Users browsing this forum: No registered users and 0 guests