Newly tested: AF-S VR Micro-Nikkor 105mm f/2.8G

Discussion in 'Nikon' started by douglas anthony cooper (, Oct 22, 2009.

  1. I added these tests and remarks to the bottom of an ancient thread, and I now realize that it's not likely to get any traffic whatsoever. Here's the link:
    Essentially, after reading a number of wildly conflicting reviews on the web, I decided to do my own tests -- which, not surprisingly, conflict wildly with a number of reviews.
    This optic seems to get slandered a fair bit, and praised to the skies by others -- I think these images speak loads. Which probably means that there is (or was) serious sample variation with this particular model.
    I'll add a couple of images to this thread, as soon as the uploader decides to play nice. (It's currently on strike.)
    For those who've worked with the 105VR, I welcome your response -- do you see anything like this?
  2. Ah! The upload worked. Excellent. That was shot with the lens wide open, which at this distance is f3.2. The shutter speed was 1/2000 sec, so it's not much of a test of the VR -- but it does seem to have nailed the focus.
    Here's the poster it was cropped from -- I chose the sharpest bit in the center:
  3. All this to say: despite some very prestigious reviews to the contrary, this lens is *not* a dog, by any stretch of the imagination. In fact, it can hold its own with just about every lens currently or previously in my arsenal.
    My images from the previous thread -- at 1:1 and portrait distances -- back this up. Thoughts?
    Oh, and here's what it can do out of focus:
  4. Some folks will pooh-pooh it no matter what. I'd probably get some use from it if the VR is effective at around 1:2 to 1:4 magnification. Reports seem contradictory on that issue - what's your impression of the usefulness of VR at close range?
  5. I like my copy. I haven't used it enough to have an opinion on the VR.
  6. Here's a recent sample.
  7. Tom, that's lovely. (Although it might be time to consider fumigation.) Looks as if you stopped down a fair bit?
    And Lex, I definitely get more than a stop, predictably, at 1:4 -- generally at least two. At 1:2, I can get a couple of decently sharp images at 1/30 sec, if I shoot a blitz of four or five.
    As a test, I just tried shooting at 1:1.5, at 1/30 sec, and the results are more than respectably sharp for an 8x10. Attached below:
  8. Not a masterpiece, but it gives some idea of what's possible.
    The point of having VR is not simply for quasi-macro shots, however. This really is one of the best portrait lenses in the Nikon line, and VR is unquestionably useful there. I posted a portrait in the other thread; I'll move it over here. (I assume I can delete the version over there? This is a bit messy...)
  9. My experience has been the same. I know from sample variations, but related to VR issues. Looks like my sample is simply perfect.
    Last weekend I decided to shoot with the VR "on". Although I don`t like VR results at all (of course it is not substitute of a good tripod when looking for the sharpest pics) I must admit that helped me to have way more acceptable images than without it. In fact, I had some "by commitment" portrait shots that were saved thanks to the VR... only a slightly bit of blur (I was shooting near the limits).
    About other`s reviews, some are really unfortunate. Other are more serious. I think this lens although a Micro-Nikkor, -nowadays- is not a macro specialist like I think it is the newest T&S macro line. I can understand that some macro specialists are not very fond of this lens (nor the previous ones, Rorslett` best rated 105/2.8 Micro-Nikkor version is the AiS... at infinity! Otherwise he top scores the 85PC).
    I did a comparison time ago with my 105 favourites: 105/4AiS and 105/2.5AiS, vs 105VR. Shooting wide open I try to remember that probably the f4 version could be very slightly better in some aspects at the macro range (there isn`t any trace of CA), and the f2.5 a bit better in the distant range. I think the 105VR works pretty good at all ranges. Add fast focus, VR, non moving external elements, good bokeh, etc.
    BTW, I love your dog pic!
  10. Thanks! Funny, I did the same thing recently: compared the 105VR to the 105/2.5 AI-s. I noted this in the previous thread: my AI-s is in fact marginally sharper at infinity... maybe. They're incredibly close. At its closest focus, the AI-s holds up in the center, but the corners go to pieces. In terms of portrait lenses, I'd say it's a wash: they're both about as good as it gets. I have a slightly easier time coaxing the AI-s into producing weird bokeh -- ringed specular highlights, etc. -- but for the most part they're unimpeachable. Both have mild issues with lateral chromatic aberration; the Micro has bokeh CA as well, which I haven't really noticed yet in the older Nikkor. But nothing extreme, and neither shows much purple fringing.
    The few (but prominent) negative reviews of the 105VR are a puzzle: I really don't know what else to ask from a lens, frankly. I suppose if you're a macro specialist, the focus breathing is going to cramp your style, but surely you can focus with a rail to get around it?
  11. I like the 105VR. It is sharp, has a nice bokeh (esp. with VR off). Handles nicely. It has a specific rendering (that it is not to everybody likes) and stepped down hold well even against the ZF100/2.
    Yes the focus can hunt a bit. It is prone to CS. It vignettes wide open of FX if there is a filter on. But this problems exists with any lens and they are rarely even worth looking for. A winner honestly.
  12. Douglas, good point regarding emphasizing this as a close up lens for portraits. I've blown quite a few candid photos of people the past year or so because I can't reliably hand hold steadily enough with my 105/2.5 AI any more. Manual focusing isn't a problem, but it's tougher for me to handhold steadily enough for decent sharpness below 1/250th sec now. I don't do much macro photography but do enjoy hand held candid and casual snaps.
  13. This was an informative thread so far.
    But how about some of you post some of the "good" reviews for the rest of us
  14. Well, this one, at photozone, has the 105VR breaking records when it comes to corner resolution:
    The problem is that these MTF tests are restricted to performance at infinity. There are lots of anecdotal raves by professional macro photographers out there, however, and here's a Siemans-star test at closer focus:
    Also, take a look at the images I posted at the end of the previous thread:
    (Starting a new thread may have been an error, in fact: now *both* are active...)
  15. OK, my experience of the 105 VR "only" comprised 4 different samples independently obtained over a period of 2.5 years from different dealers. All behaved in a similar fashion. Bad luck?
    The 125 Voigtländer has no CA issues on the D3X and is my standard lens for landscapes as well as for close-ups. It is as sharp as the D3X permits, which suffices for me.
  16. No... I'm not doubting your judgment. As I say, I find that the rest of my lenses behave precisely as you predict -- the 105VR is the only anomaly. My question is whether the four you tried showed results at 1:1 more like the first images in the other thread, or the ones I posted towards the end:
    (It's certainly possible for four lenses to come from the same substandard batch...)
  17. I have both the 105/2.5 AIS and the 105/2.8 VR. I find the VR helpful, but not at close focusing distances (1:4 - 1:1). The resolution of both is superb, but the AIS seems to have more contrast with images having just a bit more pop. The VR lens is also rather large and less ergonomic, but then this is a bit of apples and oranges - non VR, non AF vs VR and AFs, I wouldn't expect it to be smaller.
  18. Douglas: if you post jpgs from the camera, the EXPEED engine already has removed the CA. So I'm not surprised you didn't see the problem. However, not using jpgs and not using Nikon software, and issues become more obvious.
    I find it *very* unlikely that 4 lenses independently obtained over a period over 2.5 years should come from the same, allegedly bad, batch. This is a high-volume item in Nikon's lens line-up. Plenty people report seeing similar issues that I have reported. There is a difference between a positive and a negative observation too. Not seeing something does not entail it is not present. Observe something and you know the issue is there.
    Anyway, I'm now using alternative lenses that do not show any of the problems I saw with the 105 VR and these lenses do behave impeccably even on the D3X. No axial or lateral CA even when processed through third-party software. Remarkable image quality and bokeh too.
  19. Oh, there's no question that chromatic aberration is an issue. I don't in fact use Nikon software, or shoot jpg -- it's all RAW, converted in Lightroom. (I somehow exported those files as jpgs, and I have no idea why. The ones at the top of this page, however, are directly from the RAW files, and not corrected.)
    Still: the Zeiss 100/2 seems to be considerably worse with regard to both lateral and longitudinal CA, and the purple fringing is a nightmare (whereas it's rarely present in the 105VR). I'd in fact love to give the Voigtlander 125 a try, for this reason -- or a converted Leica APO Macro.
    I'd say I see the same level of CA in most of my lenses, frankly (including the 105/2.5 Ai-s) -- it's not a deal breaker. The issue I have is not so much with your review, but with the images posted at the top of that other thread, which were *way* softer than they should have been. Sharpness simply isn't an issue with this lens -- or it shouldn't be. This is where I'm fairly sure there must be sample variation; you can't really fool MTF software into making a lens seem sharper than it really is, and too many rigorous tests have duplicated the performance that I experience.
  20. Bjorn, if you're still checking into this ancient thread, it would be good to have your response. I just tripped over this more recent post from you on, which seems to suggest that those of us raving about this lens aren't in fact hallucinating:
    "Recently, I retested another 105 VR as part of a bigger test setup of "macro" lenses for a Norwegian magazine (the test comprised Nikkor and Canon lenses, plus Sigma). This sample performed markedly better than the lens I had rejected three years ago. True, the basic issues with some axial CA and a loss of detail in the extreme close-up range still were detectable, and focus breathing is, like the Canon 100 Macro models, a huge problem if you use the lens on a tripod for close-ups. However, the overall sharpness of this copy was at the same level as the new Canon 100 EF L USM and the Sigma 105 EX DG HSM."

    Frankly, having lived with this lens for a long time now, I'd say the *only* issue I have with it is longitudinal CA. And it has that flaw controlled at least as well as any of my other lenses. It would seem that you need a Voigtlander 125mm or a Leica APO-Macro to vanquish this problem completely -- I've seen no evidence that any of the recent Micro-Nikkors or Zeiss Makros are an improvement.
    Focus breathing: sure, that's going to be an issue with this design, and it wouldn't be my first choice for macro work on a tripod. But for handheld macros and portraits, I haven't seen anything better. In fact, the 105VR has the best O-O-F characteristics of any lens I own, or have owned: and I've had 3rd- and 4th-generation pre-aspherical 35mm M-Summicrons; the Canon 135/2L; the screw-mount Canon 50/1.4 -- many of the world's most famous lenses in this department. The *only* lens of mine that outperforms it in this respect is the large-format Rodenstock 150/5.6 Sironar-N -- a piece of glass that I've found can do no wrong when it comes to blur.
    I'd very much like to experiment with the new 85mm 1.4G, but from what I've seen online it doesn't have longitudinal CA tamed any better than the 105VR, and the bokeh seems -- heresy -- a touch less impressive. (Although, as an old-school Leicaphile, I can't help but be blown away by the performance in the corners at f1.4 -- this was *the* characteristic people sought in fast lenses back in the day: Noktons, etc. And it seems as if the new Nikkor has finally nailed it.)
    All this to say: there's a reason why you keep getting annoying flack from people like me over your first review of the 105VR. Might be nice to revisit it on your main page -- I'd especially like to see how it performs compared to the new 85, and the Zeiss 100/2.
  21. (Not that you don't have better things to do with your time.)
  22. My evaluation of the 105/2.8 AFS Micro-Nikkor would be substantially more positive if it hadn't been labelled "Micro". That designation sets expectations the 105 VR cannot met. It is not a "dog" in any sense ofthe phrase, however, it clearly is not the best 100/105 there is out there. Even the cheap Sigma 105 EX is demonstrably better in optical terms. Good enough, perhaps, if you do all kinds of photography with it and stay away from real close-ups. Also, one's perception of this lens will be flavoured by the criteria to which most weight is put. I for one cannot stand the annoying focus breathing and the (unnecessary) amount of axial colour, when so much better alternative optics exist. Others will be endeared by the excellent bokeh and nice colour rendition, to name an example. However, I thereafter found a solution which is sharper, has no CA either lateral or longitudinal, and with even better bokeh, so had no need for the 105 VR and accordingly, returned it.
    The sentence "you keep getting annoying flack ... over the review of the 105 VR" is meaningless. I never care about what people may think about what I observe and report. If they don't like hearing observable facts, this is their problem not mine.
  23. Just being (very modestly) amusing re: flack.
    The only fact I was observing, really, is that you'd apparently changed your mind about significant sample variation.
  24. Not at all. Some variation had to exist to explain the wide range of user responses to this particular lens. I have had hundreds of e-mails from people observing similar to what I found, and a fair share of reports that they didn't. Not all of this could be a result of different expectations alone.
    The basic optical properties of the 105 VR do remain, though, even when sample variability exists. The optics do show axial colour (longitudinal CA) and severe focus breathing. For me that's more than sufficient to stay away from this lens. It is true some other highly regarded lenses (for example, Zeiss 100/2 Makro-Planar) has issue (axial colour, also purple fringing of highlights) , but that is no excuse for the 105 VR. Canon has a recent 100/2.8 IS with optics equal or better than the Nikkor and much less of CA issues is seen with their offering. A quite cheap Sigma 105/2.8 EX equals or surpasses the 105VR in optical perfomance, has negligible CA, and little focus breathing.
    The 105 VR is good but in no way unique or outstanding in its class. I stick to my 125 APO-Lanthar as the best option - for me - as this lens excels both at distance and for close-range shooting. No CA, and focus breathing is substantially diminshed.
  25. Yes, the problem really is "its class." It's termed a Micro, and the one thing it probably isn't perfect for is macro work on a tripod. (Which is also true of the Zeiss, as it doesn't go past 1:2.)
    Measured against, for instance, the 105/2.5 AIS, the 105VR holds up beautifully. The AIS is a stunning lens in its own right, of course, but it can't compete in the corners at near distances. It also has significant CA -- as do the 105mm DC, the Canon 135/2L, and the various 85/1.4 lenses that are the real competitors here.
    The class we should be talking about is high-end short teles -- the 105VR is a class leader here. (You can consider decent macro capacity an added feature.)
  26. A lens designated Micro-Nikkor has its "class" set by default. In this respect, it disappoints. That's what I have been saying all the time. No bashing, just stating a simple fact.
    I do, however, disagree the 105VR is a class leader when it comes to high-speed medium long focal lengths (it is *not* a "tele"). A great many other alternatives should be considered. The 105 VR is, like so many multi-purpose designs, in the sitation of being applicable to many sitations and master of none. If that is the lens you need, by all means get it. For me, the 105 VR was a disappointment and it is the only Micro-Nikkor I don't have in my lens arsenal which says a lot.
    I don' think there is much more I can contribute to this thread.
  27. (Yes, I'm using "tele" in the loosest sense; it's not technically a telephoto.)
    I suppose I'm just wondering what it is that you *do* see in, for instance, the 105/2.5, that you don't see in the 105VR.
  28. I compared the 105/2.8 VR to the 125/2.5 APO-Lanthar and found the latter to be a much better lens Higher sharpness, better bokeh, no CA, little focus breathing . So that is the one I use. Nothing more complicated than so. What more can I say?
  29. Oy. I give up. This is a non-conversation. YES: we all agree that there are a few lenses out there -- unavailable on the open market -- that outperform the lenses available to mere mortals. I'd like a Cosina 125/2.5 APO-Lanthar. I'd also like a Contax-mount Zeiss 85/1.2. (Yes, it exists, but nobody I know has ever seen one). And unicorns, I'm fairly sure, are superior to horses -- but I won't be able to test that hypothesis.
    I really never meant to be even slightly confrontational -- although you assumed that from the start -- but this is just silly. I talk about the 105/2.5 -- you respond with the Cosina. I ask about various 85s, the 105 DC, the Canon 135L, and your response? The Cosina is superior. How about the Zeiss 100/2? How does that compare to the 105VR? In that matchup, you'd recommend the Cosina.
    These seem to me fair questions, given that yours are by far the most prominent Nikkor reviews on the web. You don't have to answer them -- of course you don't -- but these non-answers are just weird. Is there ANY reason to believe that the 105VR is inferior to any of these lenses: the AIS 105/2.5, the three iterations of the 85/1.4, the 105 DC, the Canon 135L, the Zeiss 100? These are the ones I've been discussing. (All of which are, for what it's worth, more expensive than the 105VR.) These are mostly lenses you've given your highest rating. So my question -- which you've completely ignored -- is what is it that you find inferior about the 105VR to *these* lenses. All of which demonstrate chromatic aberration (some of them egregiously). Focus breathing in this matchup is a non-issue. I'll take your word for it that the Cosina APO-Lanthar is a superior product. As is the Coastal Optics. As are the Lamborghini Miura and the Cray supercomputer.
    In fact, this entire non-conversation has gotten off to a non-start. It was a response to posted images by a member here -- purporting to demonstrate the inferiority of the 105VR -- and I proposed that these images were completely atypical. And I politely requested your input. To which you've responded: the APO-Lanthar is very sharp.
  30. So you want me to produce a lie, telling that some lens is better (or worse) according to reports I already published? I don't understand your motivation for pursuing the topic in this manner, and this might show.
    I also told that a Canon 100 IS Macro or Sigma 105 EX is at least the equal, if not better, than the 105 VR. This goes against the notion that the 105 VR "is a leader in its class", or not? The 105VR is good as an overall lens, yes, outstanding, no. As a Micro-Nikkor it fails on criteria important for shooting close-ups. These are the nuances you don't seem to grasp.
    Cosina has made Mk.2 of several of the earlier Lanthars and given the cult status gained by the 125 APO, I would not be the slightest surprised if it re-appeared on the market.
    High-contrast, low-detail targets rendered at moderate magnification such as the examples at the beginning of this thread are easy match for any lens designed for close flocusing. They are not very informative. Try a knife-edge test or Siemens star, or nature subjects with their inherently lower contrast and continuous tonal range.
    No further comments from my side. You are absolutely correct, there are better ways of wasting one's time.
  31. Nobody's asking you to lie. But it would not have been a waste of time if you'd addressed the question; it would in fact have been extremely interesting. We're clearly through here, but the question was -- for future reference:
    "Is there ANY reason to believe that the 105VR is inferior to any of these lenses: the AIS 105/2.5, the three iterations of the 85/1.4, the 105 DC, the Canon 135L, the Zeiss 100? These are the ones I've been discussing."
  32. Hm. Reading over this last few exchanges, I realize that I come across as a bit rude. So I apologize.
    Part of it was my reaction to the "tele" gotcha. Yes, like most lens aficionados, I'm aware of the fact that "telephoto" technically designates a very specific kind of long lens, and that the term is used improperly by most photographers.
    That said: language changes. It is arguably not simply pedantic but *wrong* to nail somebody for using the word "tele" to designate any lens with a long focal length. The word is now used in that way by nearly everyone. It is in fact how Nikon refers to its long lenses in the official literature:
    (None of the lenses here is an actual telephoto, according to the precise definition: a lens that is physically shorter than its focal length.)
    Canon has also embraced the new usage:
    You'll find this (and not the more technical) definition in most dictionaries. So, unfortunately -- according to all but the strictest prescriptive grammarians -- that is what the term now *means*. Anyway, I got a bit prickly: yes, I know what a telephoto is (or used to be).
    All of this got in the way of what I intended to be a civilized discussion of a specific piece of glass. Apologies.

Share This Page