I recently picked up the Tamron 17/3.5 + hood (2nd version 151B) to test against my Vivitar-branded Tokina 17/3.5 on my 5D. Because of the reputation of the Tamron, I was expecting an uneven fight, but was surprised how the Tokina stood up to the challenge. Overall both are very good performers and affordable FF UWA options.
My observations (based on my copies):
1. Sharpness: The Tamron is better in the corners wide open and stopped down, while the Tokina is sharper in the center wide open. Stopped down they're very close in the center.
2. CA: The Tamron tended to be slightly better with CA
3. Flare: The Tamron handles flare considerably better than the Tokina
4. Distortion: I didn't test the Tamron, but from samples I've seen, both have significant mustache distortion
I shot 3 similar scenes, all with similar results and one shown below. My tests were far from scientific, but IMO representative of my real world use (tripod+remote). I need to find a better scene to test, so this for me is considered a preliminary result. I'll try to update this thread again when I get around to more tests.
I would be glad to hear input and/or suggestions for improving the tests. There are some confounding factors which may need to be considered:
-Tamron has the OM-adaptall adapter + OM-EOS adapter (perhaps EF-adaptall would be better?) while the Tokina has the Nikon-EOS adapter for use on my 5D
-Neither reaches hyperfocal distance at the infinity marker, so I used pretested markers
Suggestions to improve test: Probably get a flatter subject, although it seems to have worked for this, as depth of field might throw a closer corner out of focus and make a lens look bad when it isn't.
AmbientMike wrote:
Are you sure you nailed focus? The tamron is blurry at 3.5. If it was focused closer, that might be why it looks sharper in the corner.
I did a pretest to find the hyperfocal distance, so I thought I had achieved it, but I can't rule out the possibility that I did miss. Unfortunately I don't have LiveView to ensure.
My comparison tests of the Tokina AT-X AF Pro 17/3.5 (EOS) and Tamron SP 17/3.5 Model 151B, with no built-in filters, gives the edge to the AT-X. I used the older Tokina SL 17/3.5 RMC manual focus lens with Pentax film bodies, and it was quite nice, but I have no comparisons of it with the SL and the SP. I suspect the SL is the same model as the Vivitar-branded lens you mentioned.
Both the AT-X and SP are better than the Canon EF 17-40L at the edges and corners, but the L-zoom is noticeably sharper in the centre of frame.
I have tried all the lenses mentioned here and have to agree that the Tokina ATX Pro AF version is better than either of the MF options, but it is also more expensive.
It seems to me that the Tamron 17/3.5 is sensitive to register distance, when I first got it and slapped on an aftermarket A-2 to EF adapter it seemed decent enough to warrant the $200 price tag. Then when I picked up an OEM A-2 to EF adapter it improved by a fair margin, especially in the corners. So when doing a test it might be prudent to match the original adapters thickness exactly for optimum performance.
helimat wrote:
...when doing a test it might be prudent to match the original adapters thickness exactly for optimum performance.
Good advice. I'm confident that my SP adapters (Tamron A2 to M42 + generic M42 to EF) are well matched for optimum performance. They turned in great results for my SP 180/2.5, SP 300/5.6, and SP 35-80/2.8-3.8. IOW, I'm confident my conclusion that the AT-X is better than the SP 17/3.5 would not be changed if I had used different adapters.
I can vouch for the mustache distortion on the Tamron. Otherwise, I'm very impressed with it. Might even be a few photos from it showing up in a nice glossy mag sometime later this year...
justruss wrote:
I can vouch for the mustache distortion on the Tamron. Otherwise, I'm very impressed with it. Might even be a few photos from it showing up in a nice glossy mag sometime later this year...
jcolwell wrote:
Good advice. I'm confident that my SP adapters (Tamron A2 to M42 + generic M42 to EF) are well matched for optimum performance. They turned in great results for my SP 180/2.5, SP 300/5.6, and SP 35-80/2.8-3.8. IOW, I'm confident my conclusion that the AT-X is better than the SP 17/3.5 would not be changed if I had used different adapters.
I think it has been shown and discussed around here a few times that adapter thickness is far more critical on ultra-wides than it it is on standards and teles. Remember all the debates/discussions about CZ21 adapter thickness? Even guys like Pham were offering custom adapters for it. I think the same level of accuracy is required with most UWAs. It just so happens that I got incredibly lucky stacking a generic OM adapter on my A-2/OM mount. The results are with my SP 17 are superb.
Regarding your comparison to the Tokina, it seems to be the exception, rather than the rule. Nearly every other comparison I have seen puts the Tamron ahead. Look again at the Modern Photo test results on Adaptall-2.org. The Tamron easily competes with (and convincingly beats some) OEM lenses at similar focal lengths. The description of performance on Adaptall-2.org is exactly what one should expect from a properly functioning Tamron 17/3.5. Sample variation is significant with this lens, so I am not surprised that one sample of the AT-X beat one sample of the SP.
My Nikon 20mm 2.8 AIS was a dog during a shootout with my Oly 21mm 3.5. Weeks later, I remembered the possibility of adapter variation and used a different adapter ... NIGHT & DAY difference. I had been interchanging adapters due to more lenses than adapters. It didn't really show up so much with the long glass, but in the wide ... IT SHOWS UP.
Another +1 here. I recently acquired two Fotodiox Pro adapters to enable comparison of my Voigtlander 20/3.5 and Olympus 21/3.5 - the generic adapters that I was using weren't up to the task. The Fotodiox adapters are much better, and the Voiglander is slightly better, than the Oly. I am familiar with the requirements for good adapters on UWA lenses, and with the results that you get when you don't use a good adapter. My A2/M42/EF adapter is just fine for UWA lenses.
As far as my results with the AT-X and SP, it's certainly not unique. Earlier in this thread, Paul (pdmphoto) notes that he has the same experience. I posted these results early this year on this forum, and others made similar comments. The AT-X is newer, larger, heavier, and more expensive. I think it's also better.
helimat wrote:
It seems to me that the Tamron 17/3.5 is sensitive to register distance, when I first got it and slapped on an aftermarket A-2 to EF adapter it seemed decent enough to warrant the $200 price tag. Then when I picked up an OEM A-2 to EF adapter it improved by a fair margin, especially in the corners. So when doing a test it might be prudent to match the original adapters thickness exactly for optimum performance.
Thanks Mat, this is sort of what I expected might be a factor. Which was the bad adapter (there seems to be several versions on eBay)? And where did you get your OEM adapter?
Hi Kent. I started using the Fotodiox Pros with M645 and then Leica-R, and now it's my go-to solution. I haven't posted the CV vs. Oly pics, maybe I'll work up a summary thread later. Right now, we have a hurricane due to pass by here in about 18 hours, and so I have to go and tie down the dog.
dfresh wrote:
Thanks Mat, this is sort of what I expected might be a factor. Which was the bad adapter (there seems to be several versions on eBay)? And where did you get your OEM adapter?
I got it from 'adplo' on ebay. I should mention that it worked perfectly fine on the 400/4 I had, so I wouldn't call it a 'bad' adapter, it just seemed that the UWA was far more sensitive to the register distance as others have suggested. The difference was visible to the eye in the viewfinder in fact.
