I did some quick test shots yesterday, indoors as weather sucked. Before I say anything, I have to say, I know I got a good copy of Sigma. Probably the best one there is in stock. I got it straight from the distributer here in Slovenia and he told me himself he checked it.
As for the Tamron, I got the only copy available in the country.
We all know that quality control is not that great with "3rd party lenses" as we like to call them, so margin for error in this kind of comparison test is wide. I'd probably have to compare 3 or more same lenses to get some accurate/average results.
But anyhow, first comparison done, to put it in one sentence: Sigma wins in sharpness hands down.
Sun is out, I'm packing my gear, so check back in few hours for comparison shots and final words.
Update #1:
Ok, I did manage to get some shots, despite the fact the weather is against me and it started to rain again.
All shots done on D300, aperiture mode & matrix metering. Shot in NEF, imported into Lightroom, comparison view for same show of both lenses, screenshot, saved in Photoshop CS3, quality level 10. It's for show, not trying to get the best out of the images. Same process for all of them, so post process in not a factor.
10mm, Tamron f3.5 vs Sigma f4, border
10mm, Tamron f5.6 vs Sigma f5.6, border
10mm, Tamron f8 vs Sigma f8, border
10mm, Tamron f11 vs Sigma f11, border
I guess no comment needed, right?
10mm, Tamron f3.5 vs Sigma f4, center
10mm, Tamron f5.6 vs Sigma f5.6, center
10mm, Tamron f8 vs Sigma f8, center
10mm, Tamron f11 vs Sigma f11, center
20mm, Tamron f5.6 vs Sigma f5.6, border
20mm, Tamron f8 vs Sigma f8, border
20mm, Tamron f5.6 vs Sigma f5.6, center
20mm, Tamron f8 vs Sigma f8, center
I noticed that on almost all the shots Tamron appears to be one stop brighter. Just did a quick controlled test in a closed environment and this is not the case at it appears. Both lenses produced virtually the same exposure data on all apertures and focal lenghts. So it must have been the weather again..
Close up shots..
Tried some shots from the minimum focusing distance.
10mm, Tamron f4 vs Sigma f4, border
10mm, Tamron f5.6 vs Sigma f5.6, border
10mm, Tamron f8 vs Sigma f8, border
10mm, Tamron f4 vs Sigma f4, center
10mm, Tamron f5.6 vs Sigma f5.6, center
10mm, Tamron f8 vs Sigma f8, center
Some CA tests. All 1:1 crops from the edge of the images.
Tamron f5.6 CA sample:
Sigma f5.6 CA sample:
Tamron f8 CA sample:
Sigma f8 CA sample:
Update #2:
Vignette test shots:
Another handling observation: Sigma has about 1.5cm space between the rings (plus some extruded lines on both sides), so it's easy to get a good grip when you're replacing lenses. Tamron has only around 0.5cm and it's a lot harder to hold it and replace it. Especially if you're in a hurry.
Conclusion:
Judging by my two copies, Sigma is a clear winner. It's overall sharpness is better then Tamrons, far better on corners and better in the center. CA appears to be similar, but Sigma is still ahead slightly. Tamron is brighter, has less vignette and has easier distortion to fix. Tamron has a bit longer range, is a bit brighter, costs less, weights less, but I'm not sure those factors overpower the clear image quality advantage Sigma showed? Not for me anyhow.. If you're thinking about buying one of these wide lenses, or any lens for that matter, try to test and compare your copies yourself if you can. One copy can really be totally different then another.
UPDATE (03.12.2008):
I've received some images from a friend overseas, who has tested his copy of the Tamron lens. His results were A LOT better. I talked with our distributer and we have concluded the copy they got is a really bad one. So, I'm awaiting a new copy and as soon as it arrives I'll do another test and comparison with Sigma. Stay tuned.
Saturday, 22 November 2008
Tamron 10-24 compared to Sigma 10-20
Subscribe to:
Post Comments (Atom)
9 comments:
Thank you Mat for posting these photos.
As you say sample variations can make a difference.
What was your focus mode? Could the tree skew the focus distance or did you manually focus on the building?
Yep, until you compare it with something, it's hard to judge the quality.
But I'm still not a fan of this kind of tests. Sure, the findings stand for my two copies of the lenses, but can you really say it's the same for the whole Tamron 10-24 line vs Sigma 10-20 line? I remember my first Sigma 10-20 copy from a year ago, it was terrible, I didn't even remotely consider buying it. But this one, I'd buy it straight away and I probably will. But what about Tamron, is this a good representative of the line? It's the only copy in the country, the Tamron representative here got it as a sample. So it's not like I can take my pick. I think it's still too early to easily say which one is better, from the brand point of view. Some more tests will have to be done, other copies compared, then we'll gave a global idea which line is better.
If you plan to buy one lens it's always best to borrow it and do some tests with it, that's the only way to make sure if the copy is good or not. That's what I always do anyhow.
In those tree shots I focused on a stump in the center. Beside, f8 and about 5m focus distance, dof is from 0.5m to infinity anyhow.
But I see CA is still really unpredictable. On some images Sigma produced much more CA then Tamron did, on others it was the other way around. But overall my Sigma copy produced less CA then my Tamron copy.
"it's always best to borrow it and do some tests with it" - can't do that where I live, it's usually a $100 USD S&H round trip. :-(
I was asking about the focus in the first 12 shots side-by-side series of the building with a tree almost in the center.
CA variations could be affected by dynamic range difference between the sky & the branches that may vary with clouds thickness & angle of the sun.
Meaning if you take a shot & the sky is +4 stops compared to the branch & by the time you swap lenses & shoot again it's now +5.5 for example.
The moisture in the air could also be a factor (fog like effect) & looking at the trunks, there is exposure variations.
Uf, that sucks. I really can't imagine buying a lens without testing it first. Quality changes so much from copy to copy with these "third company lenses", and even main players like Nikon and Canon aren't to be trusted anymore.
Oh, first 12, focus was on a window ornament on the top.. that square thing. Focus point just above center.
You're right about the clouds and sun condition, but looking at the exposure numbers, the difference may have been around 0.5 stop, so not much really. Was quite a uniform weather at that moment, those were some of my last shots, just before the rain started.
P.S. Added another image, vignette test.
Hi,
Nice work you've done. It gives you something to think about. Did you test the lenses before taking the pictures. The sharpness of Tamron was better when the focallength increased. Back or forward fokus could be the explanation to that. I had that problem with my Sigma 70 macro.
Thank's a lot for these photos!
It looks as if the Tamron suffers from field curvature (a characteristic also found on their 17-50). If you look at the close-up shots, the center is very sharp and the extremes border are terrible. What do you guys think?
I think it could even be a slightly misaligned element, the main quality control issue in lenses like this, and the main reason to test the lens before you buy it. All complicated modern lenses have to be extremely precisely constructed. Another copy or three should settle it.
Focus issues *might* apply to the close-ups, as the focus seems to vary across the middle vertical on both lenses as the surface recedes, but I would have expected dof to cover that.
Really, I've seen so much variation between ultra-wides (and even with the same lens) that I would expect to test several samples before coming to any useful conclusion. Apart from the sharpness issue, it's still looking like a pretty good lens.
Keep up the great work!
A reasonable review here:
http://kenrockwell.com/tamron/10-24mm.htm
Another bad copy or... is it just a bad lens?
Pity.
A reasonable review here:
http://kenrockwell.com/tamron/10-24mm.htm
Another bad copy or... is it just a bad lens?
Pity.
Post a Comment