Lenses and Optics

Tamron f/1.8 VC Prime Lenses Sharpness MTF Curves

Published December 7, 2015

OK, for the first time I can remember, I’ve been browbeaten into doing a test I didn’t plan to do. When Tamron released their neat little Tamron 35mm f/1.8 SP Di VC and Tamron 45mm f/1.8 SP Di VC lenses, I thought, well that’s interesting, and the price point is reasonable and didn’t think about them anymore. But I got enough emails, notes on other lens tests, and all around forum nagging to test them on our optical bench that I caved and did them last week. Now I do see the attraction.

We did our usual optical bench MTF testing, using 10 copies of each lens to arrive at the average MTF chart and variation scores. If you don’t know what all that entails, please read our earlier post on testing methods.

You didn’t read it, did you? That’s OK. If you don’t speak MTF, don’t worry. It’s not hard. Higher on the vertical axis is better. Dotted and solid lines of the same color close together are better (far apart is astigmatism). The horizontal axis goes from the center of the lens at “0” to the edges of the lens at “-20” and “+20”. Lower lp/mm have an association with strong contrast while higher lp/mm are associated with an ability to resolve fine detail. So, without further ado, here are the MTF curves.

Tamron 35mm f/1.8 VC

Roger Cicala and Aaron Closz, Olaf Optical Testing, 2015

 

Some of these comparisons aren’t quite fair (the two Canon lenses are either tested at wider or narrower apertures, and, of course, the 35mmf /1.4 II is at a very different price point). But basically the Tamron 35mm f/1.8 is competitive with the other 35mm prime options. It’s probably just a bit better than the Nikon 35mm f/1.8 and competitive with the Canon 35mm f/2 IS (the Canon’s a tiny bit better on the MTF charts, but remember it’s being tested at a slightly smaller aperture, which helps MTF).

Overall, it’s a good showing by the Tamron. On the other hand, the Tamron is about the same price as the Canon and Nikon lenses. I’m not sure the f/1.8 to f/2 difference would make me pick a third-party lens over the Canon. The Vibration Control might be a strong advantage for some people on the Nikon said, but not all that many, probably.

Copy-to-Copy Variation

 

Variation graphs are quite positive for the Tamron. It certainly is more consistent copy-to-copy than the Nikon 35mm f/1.8, which has a lot of variation, and very similar to the Canon 35mm f/2 IS lens, which we consider to be excellent.

Tamron 45mm f/1.8 VC

We don’t have a lot of other 45mm lenses to compare, so I used some similar aperture 50mm lenses for this comparison. This does cause a little role reversal because the Tamron is now by far the most expensive lens in this comparison. On the other hand, it’s the only one that has Vibration Control, too.

 

 

The Tamron is clearly better than either of the other 50mm f/1.8 choices as far as the MTF curves go. It was so good I decided to also compare it to one of the better (and more expensive) 50mm options, the Sigma 50mm f/1.4 Art. The Sigma is at a bit of a disadvantage here, being tested at a wider aperture, but still the Tamron is very competitive with the Sigma, which is a superb 50mm lens. Plus, the Tamron still has vibration control.

Roger Cicala and Aaron Closz, Olaf Optical Testing, 2015

 

Variation

Roger Cicala and Aaron Closz, Olaf Optical Testing, 2015

 

The Tamron 45mm f/1.8 demonstrates pretty remarkable consistency, one of the most outstanding lenses we’ve tested as far as low copy-to-copy variation is concerned.

Conclusion

Well, I have to admit, for some lenses I didn’t want to bother testing, the new Tamron VC primes are eye openers. Let me be clear, I haven’t done anything but test them on the optical bench, I haven’t taken a single picture with either of them. But the bench-test results are pretty impressive. Both lenses show excellent sharpness. Combine that with vibration control at a reasonable price and I suspect that more than a few people are going to be very interested in these two lenses.

There is one caveat that needs to be shouted loud and hard, though. The obvious use for these lenses is shooting in low light situations: night clubs, bands, street shooting at night. For many photographers, low light autofocus performance is going to be of critical importance to that kind of work, and I wouldn’t rush out and buy one until you read some reports about how these lenses perform in that regard. But the MTF curves certainly indicate these are lenses worth investigating.

 

Roger Cicala and Aaron Closz

Lensrentals.com

December, 2015

 

NOTE: For those of you who want to look at other lens MTF comparisons, I just wanted to remind you that all of our MTF and variance graphs are available through The Digital Pictures Comparison Tool.

Author: Roger Cicala

I’m Roger and I am the founder of Lensrentals.com. Hailed as one of the optic nerds here, I enjoy shooting collimated light through 30X microscope objectives in my spare time. When I do take real pictures I like using something different: a Medium format, or Pentax K1, or a Sony RX1R.

Posted in Lenses and Optics
  • Buc Nelson

    FAKE NEWS AND REVIEWS! yet another post about tamron, do you work for them? check his discus post history, unless he’s turned privacy off by now.

  • Buc Nelson

    amazing yet another post about tamron, do you work for them? check his discus post history, unless he’s turned privacy off by now.

  • Buc Nelson

    haha yet another post about tamron, do you work for them? check his discus post history, unless he’s turned privacy off by now.

  • Ah! I understand now. Actually it was just me trying to be funny. There really was quite a bit of enthusiasm about those when they were released and people were excited to see the results, but it took a while because we couldn’t keep them in stock.

  • dyna

    The numbers here make perfect sense. To clarify the reason I’d originally asked the question: The tone in your intro implied a definite bias of some sort, making it seem as though you came to the table with a predetermination that the Tamron lenses would not be worth your time and admitted to surprise that in the end they were. It just seemed that there wasn’t the usual open feel of neutrality you guys are known for and I was surprised by it, especially when initial reviews and reports elsewhere on the new SP series puts them right in with N, L, and ART lenses. That all being said, I appreciate that you did the test, regardless, and I apologize on behalf of the internet heathens who badgered you so :). Thanks, as always, for your (and your team’s) efforts. They do make the photoweb a better place.

  • dyna, I’m not sure I understand the question, but I’ll give it a go. Our primary purpose for testing is to establish norms for the full-frame lenses we carry. We don’t test superzooms (10x zooms) or inexpensive consumer (kit lens) zooms because the results are pretty meaningless. That leaves us with about half a dozen Tamron lenses to test and we’ve tested 4 of them I believe. The other reason is we have to have 10 copies on the shelf available to run our tests. In the case of the other two that hasn’t happened when we had testing time available — we don’t stock all that many copies (maybe 20 of each) because there isn’t as much demand. To give you perspective we rent about 10X as many Sigma lenses as Tamron or Tokina so getting stock to test is pretty easy.

  • dyna

    Is there a particular reason why the Tamron lenses wouldn’t necessarily rate or earn the tests that Sigma or other third party lenses have the “privilege” of going through?

  • This made me realise how nasty the bokeh of the Tamron is – those onion rings are incredibly obvious.

  • Brandon Dube

    The camera is a sony something-or-other scientific detector with a 50x microscope objective. You may consider it a 50,000 x 50,000px 2mm x 2mm camera with several calibrations for the errors of the portions of the optical system other than the lens under test.

    The MTF bench is essentially “cameraless.” At about 1000lp/mm the bench will lose critical accuracy, but no lens you put in front of it will resolve anything of meaning at that frequency anyway.

  • kamran zafar

    comparing mtf from two different lens mounted on two different dslr is
    not really a comparison.please mention the camera used during the test

  • dyna

    Bench testing aside, the Tamron Lenses have other interesting upsides vs their competitors, including ART such as fluorine coatings on the front element, weather sealed, overall construction, the closest minimum object distance for any of their kind… Their price:performance ratio is pretty phenomenal when all the points are considered.

  • Roger Cicala

    Grant, all autofocus is a measure of both lens and camera body, limited by the weaker of the two links. It’s a feedback loop with variables in both.

  • Ron

    Grant – there have been a lot of user reports of unfavorable AF performance with third party lenses (Sigma, Tamron, Tokina, etc.) on Canon and Nikon cameras, vs. lenses from those manufacturer on the same cameras. Either slower focusing or inconsistent focus. Therefore the thinking is that something in the software/hardware of the third party lenses isn’t quite meshing perfectly with the cameras. I believe all of the third party manufacturers have to reverse engineer the Canon and Nikon body-lens communications and that it’s only Sony that has made such info for their FE mount available to other manufacturers.

  • Grant

    Perhaps someone can explain to me why low-light AF performance is a measure of a lens’ performance and not of the camera body?

  • Roger Cicala

    Omesh, I’ll do them, they aren’t that awfully time consuming but I’ve got a couple of teardowns to write up first.

  • Roger,
    You are a legend!
    Are there any plans to do field curvature tests for these Tamron’s as well, or are they just too time-consuming?
    Anyway, thanks again for these sensor-free MTFs it’s much appreciated.
    Omesh

  • Roger Cicala

    Thank you, nepo, that’s most useful.

  • Roger Cicala

    grubernd,

    Your point is valid and I’ll start (sigh) putting the full names in. It’s been a really passive aggressive thing on my part. I despise that the manufacturers have to name their new lenses a “12-250mm f/2.9-6.3 Dg Xy XFD HSM VC 123 double secret v3.2” so I purposely cut them off. But you’re right, in a few years it will lead to confusion if these things are still being read.

    Roger

  • Here is a field comparison of Tamron 35mm f1.8 VC with Canon EF 35mm f1.4 L II: http://lenswork.tistory.com/entry/Canon-EF-35mm-f14-L-II-vs-Tamron-35mm-f18-VC .
    Please note that my comparison is only one copy comparison.

  • Here is field comparison of Canon EF 35mm f1.4 L II with Tamron SP 35mm f1.8 VC.

  • Tim D

    Andre Y: Improper field tests are why I get so much acceptable lenses pre-owned or refurb I think. What you consider ‘decentered’ looks nothing more like an unrealistic expectation of a lens getting things in focus that are clearly not going to be equally in DOF. It looks like you are shooting on the same level of elevation as the bench and slightly tilting the camera upwards. The bricks at the top being significantly smaller than the bricks at the bottom seem to indicate the relative distance between the bricks on top vs bottom. Thus the bricks on bottom are starting to be close enough to be ever so slightly in front of DOF compared to the bricks at top. I think these tests show why they are so beneficial…cause the chances of folks doing proper field tests are few and far between.

  • Shakey

    VR not interesting for Nikon photogs?? I just spent a whole day shooting for a client in a dark dark dark convention center lit only by light from the stage. Speed lights were verboten and I would have KILLED for a stabilized prime. There are more use cases between heaven and earth than are dreamt of in your philosophy.

  • Abi

    Optical bench test in one thing. How does it compare during low light shooting such as indoors and the synergy with Canon bodies? My friend’s Sigma barely focuses in low light indoors situation.

  • Brandon

    Andre,

    I do not see any issues in the image you posted.

    The consistency information is also heavily averaged. So, if all of the sample are decentered/tilted the same way, there is almost no variance. It is not really mathematically possible for the average to be different. This is why we are looking into doing “intra-lens” stats in addition to “inter-lens” stats. That type of analysis will be coming after I finish finals and can complete the software for it.

    grubernd,

    The next revision of the software will most likely include a “model name catcher” routine to improve the consistency of plot titles. It is in the pipeline.

    Regards,
    Brandon

  • uhm. which Nikkor 35mm f/1.8 and Sigma 50mm f/1.4 would those be?

    btw, this is a recurring theme in your lovely lenstests: shortened names of lenses. especially some of the classic focal lengths have so many variants that we can often only guess you mean the current popular variant .. i hope these tests are still available when newer versions come along. please make your invaluable material future-proof.

  • James Scholz

    Thank you Roger, once again your provide really useful information. A few weeks ago I traded the Sigma Art 35 for the Tamron 35 and could not be happier. I am doing a project on museum photos where I can’t use a tripod and the addition of vr in these low light venues is huge. I suspect a lot of wedding photographers will really love this lens as well.

  • Albert Silver

    Many thanks for the tests. While CameraLabs agreed with the sharpness results you found, making it a very attractive lens, the most common complaint I have seen is that it also presents some nasty CA wide open. That said, the examples I saw, full res, were easily corrected in Lightroom, so probably not a deal breaker unless you refuse to PP.

  • Roger, I think you’re to be applauded for doing and publishing all of these optical bench tests since there is really nothing else out there like it … but I have to say that based on my experience with one copy of the 45/1.8 I rented from you guys a couple of months ago, I can’t agree with these results. The copy I had was slightly decentered across its wide axis, and had relatively low microcontrast. So I don’t know if that copy’s faults (you can probably look up which one I rented) were lost in the averaging, or I’m going blind (either one is a distinct possibility), but I can’t emphasize enough how important it is for people to evaluate lenses in the field rather that rely on test results alone.

    In case anyone else wants to see an example, here’s a picture taken of the Frank Gehry building at MIT with the rental 45: https://www.dropbox.com/s/wootbq9je0jyf7s/20151015-0151-Edit-2.jpg?dl=0 It’s a full-resolution JPEG saved at maximum quality from LR. I have done my usual post-processing workflow on it, but they are relatively mild. The full EXIF information should be intact.

Follow on Feedly