Sharpest Nikon lens

Discussion in 'Nikon' started by theonetruepath, Dec 17, 2009.

  1. Yes you've seen it before... but this time I plan to actually tell you what I need it for!
    I have the AF-S Micro Nikkor 105mm 1:2.8G ED and it's super sharp on my D300. I use it almost exclusively for macro work with flash.
    Now I want to have a slightly wider angle lens (50 to 60mm) for exactly the same job. I don't need the VR feature but I do want super good sharpness at the f22 side of things. I have a DX camera but wouldn't rule out a full frame lens.
    I know I could just use the 105 on a D700 but my budget isn't that flexible :(
    So what's my best bet, I'm sure there's a lens out there much cheaper than my 105 that could do the job easily?
     
  2. How about the AFS 60mm Micro? one of my favorites.
     
  3. Since you have a camera with an onboard AF motor, hunt around for the previous version of the 60/2.8 Micro (the AF-D flavor). Razor sharp. Or, get the more recent AF-S version, if you like the way that technology focuses better - but I'll bet you focus manually much of the time anyway, right?

    I use the 60/2.8 Micro all the time on a D300, and it's a favorite. I don't really hear the new AF-S version calling to me at all, actually.
     
  4. My Nikon 55mm f2.8 Micro-Nikkor AIS manual focus lens did very well at f22 on my D700:

    [​IMG]
    December 4, 2008, Seattle, Washington
     
  5. ShunCheung

    ShunCheung Administrator

    If you really want sharpness, I would avoid f22 or even f16, especially on the DX format. Diffraction will rob some sharpness from the optics.
     
  6. Shun is correct. Most lenses reach their peak at f8. I sometimes will shoot f22 or f32 when I'm actualy trying to soften an image.
    Kent in SD
     
  7. I've been really impressed with the micro nikkor 55 f3.5. It's the only lens I have ever had that was almost too sharp to shoot people.
     
  8. Although the Pigeon and the mountains required at least f16 if not f22. Sometimes you just have to do what you have to do, and obviously a superb lens will be better at f22 than a non-superb lens.
    I only came here because I thought we were voting for the sharpest Nikon lens. My answer would have been any Nikon 200mm f2 since I have the first Ai version but I am afraid I do not know about macros. I suspect I would choose the 55/2.8 AIS but you have to be careful to find one without aperture problems, but then again at the bargain basement prices I guess it would not matter.
     
  9. I have the 60mm f/2.8 D and believe me when I say I am a novice. I also use it on a D300
    I find it an very sharp lens and in the hands of someone with experience, well, it would be incredible.
    Attached it is what I have been able to do in the learning phase -
    00VHXh-201711584.jpg
     
  10. Many have observed that the 55 and 60mm Micro Nikkors are pretty well known for declining performance at the small(16, 22)stops due to diffraction. I've had both 55's, but not the 60's, and this has been my observation as well, on film and D200/D2Xs/D3. Still, because they're so relatively inexpensive, it might be worth it to test the 55/60 lenses on your camera...see if you can extract the performance you need out of them.
    The finest performance I've seen from a 50mm so far has been the Zeiss ZF50/2(on Nikon D3). I was really surprised the first time I saw a few shots made at f/16 and f/22. I've never seen the perfomance of a any lens at any focal length decline less at f/16 and f/22. Alas, the ZF is not cheap. BTW, it's fully useable wide open as well.
     
  11. Well I found the performance of my Nikon 55mm f2.8 Micro-Nikkor nothing short of outstanding, despite people saying never shoot at f22. The Micro-Nikkor lenses are highly engineered for precision images, and you decide how it looks, don't let anybody here tell you what is what. You don't need a Zeiss to achieve good results in my opinion. Nikon lenses have always been and will always be nothing short of outstanding.
     
  12. The Nikon AFS Micro-Nikkor 60 mm f/2.8 ED G is a very, very sharp lens.
     
  13. Diffraction at or near minimum aperture hasn't always been a significant problem with the 55/3.5 Micro Nikkor on my D2H so I'll stop down to f/22 or f/32 when I need the depth of field. Depends on the subject matter. There are occasions when maximum resolution without adequate DOF appears "unsharp", so I'll stop down when needed. I haven't tried focus stacking yet, tho', which might be useful for preserving maximum possible resolution and maximum DOF. But that technique doesn't seem to lend itself well to subjects that aren't perfectly stationary.
     
  14. In my experience the "sharpness" at f22 in macro work is about the same for all real macro lenses.
    While differences in resolution (between lenses) become apparent at f4.0 to f5.6 you are, as said above, limited by diffraction (and not the lens) at the small iris openings, especially with the sensor of the D300.
    I suggest to get an old 55mm Micro Nikkor, perhaps an f3.5 or an early f2.8 version, usually a very cheap buy. If in the future "sharpest" is perhaps not the most important criterion for a lens or an image you have one of the best macro lenses available that will give you flat field and excellent resolution when shooting near f5.6.
     
  15. Shun Cheung [​IMG] [​IMG] [​IMG] , Dec 17, 2009; 09:38 p.m.
    If you really want sharpness, I would avoid f22 or even f16, especially on the DX format. Diffraction will rob some sharpness from the optics.
    I've been really impressed with the micro nikkor 55 f3.5. It's the only lens I have ever had that was almost too sharp to shoot people.
    Kent Staubus [​IMG] , Dec 17, 2009; 10:07 p.m.
    Kent in SD
    Shun is correct. Most lenses reach their peak at f8. I sometimes will shoot f22 or f32 when I'm actualy trying to soften an image.​
    My photography tends toward the forensic and away from the artistic. When I shoot a flower or tropical fish (my main subjects generally) I want the whole flower or fish in focus. Since no lens on the planet has a DOF high enough to keep the flower sharp as well as the background at closest focus, I don't have to worry about 'artistically blurring' the background. F22 and flash on the subject not only blur the background, they generally darken it for you as well.
    So you can see why I want to use f22... DOF on most lenses at closest focus is horrendously low, so I need f22 at a minimum. F57 would be my first choice, but physics ensures that all lenses turn to mush at such small apertures. Certainly the 105 is still reasonable at f22 but rubbish any higher.
    I like the sound of the Zeiss, but it looks like it costs more than my 105.
    I will try to find an AF-D 60/2.8 Micro.
    My thanks to all contributors to this thread.
     
  16. I can see why you'd want f22 as well. In fact, for some reason I get acceptable (but not sharpness-optimal) results at f22 with my 55mm f3.5 on a D50, but I get unacceptable (non-soft) results with any of my other lenses at f22.
    You are going to use a lens at f22, and even those that are acceptable at that aperture are not at their best, you don't need the sharpest lens, you need one that will at least behave itself at f22. I'd stick with a micro. The 55mm f2.8 and f3.5 variants, and both the 60mm f2.8s are good micros.
    I normally say the sharpest lens you have is a sturdy tripod, but with flash, less important.
     
  17. Patrick Hamlyn When I shoot a flower or tropical fish (my main subjects generally) I want the whole flower or fish in focus.

    It sounds like you may be better off with a PC macro lens such as the 45 or 85 if you are looking to keep the entire flower or fish in focus.
     
  18. ShunCheung

    ShunCheung Administrator

    Patrick, it makes no difference what your photography interests are; the same physics still applies: if you need the depth of field at f22, you sacrifice sharpness. You simply cannot have it both ways as your original post suggests.
    However, there is now a different option: focus stacking -- http://en.wikipedia.org/wiki/Focus_stacking
    You can bracket your focus and capture multiple images with the focus at different points and then use software to merge the different sharp areas from those images into one final image with a lot of "depth of field." But since there are multiple original images, if your subject moves (e.g. flower blowing in the wind) or your light source is not constant, it makes the process difficult.
     
  19. Indeed. One of the new PC-E's is THE tool for that job.
     
  20. Getting back to the original question, I don't know if this may help but I have been experimenting with some older Nikon manual prime lenses (AI, AIS) and the sharpness I've been getting from a 28mm (2.8), 50mm (1.8), and 105mm (2.5) are unbelievably sharp on my d200. Sharper than any of the DX and VR lenses I have used. I know a lot of younger photographers don't have the patience for manual focus and rely heavily on the camera to do it for them. Maybe it's just that I have always had good eyesight, but even when I focus visually through the viewfinder (in manual mode) I've learned that the focus indicator isn't always giving you accurate results. I ignore it and trust my eyes instead. If you have great eyesight I recommend buying a few good manual primes. They're so affordable now than everyone has been moving to digital.
     
  21. If you can't use a PC lens then you still have some other choices. You can get a bellows like the PB-4 which has movements in its front standard and/or you can consider using enlarging lenses. The longer ones are rather slow to begin with and may stop down to f/32 or f/45 so using them at f/22 wouldn't cause as much diffraction. The shorter PC type lenses which have longer back focus can be used with a little extension without getting you too close for the magnification you need. I have a 55/2.8 AIS as well as a number of 55/3.5 Nikkors from the black front Micro Nikkor Auto with the compensating feature to the AI. They are not at their best at f/22 but are still acceptably good for most purposes. I sometimes use a Minolta Auto Bellows III with a Minolta X-700 and an enlarging lens. By using a lens made for enlarging MF I have extra coverage for the movements which that bellows offers.
     
  22. Patrick, I have the 60 2.8D and it is super sharp. I note that it stops down to F32, so using it at F22 is certainly feasible, but clearly there are diffraction issues at that aperture. Because of the new AFS version the D lens is very reasonable. However, if you are thinking of getting one I'd do some research first. Not everyone likes it. It's a variable aperture lens, for one thing. Also it's not IF.
    Good to hear from you by the way!
     
  23. 50/1.2 AIS. And if that's not good enough there's always the Zeiss.
     
  24. Alastair Anderson , Dec 18, 2009; 10:43 a.m.
    Patrick, I have the 60 2.8D and it is super sharp. I note that it stops down to F32, so using it at F22 is certainly feasible, but clearly there are diffraction issues at that aperture. Because of the new AFS version the D lens is very reasonable. However, if you are thinking of getting one I'd do some research first. Not everyone likes it. It's a variable aperture lens, for one thing. Also it's not IF.
    Good to hear from you by the way!​
    Howdy Alastair!
    How goes it in icy London.
    Yes I'm a bit confussed as to what all the slightly and not-so-slightly older lenses will do on my D300. I want to set the aperture manually, focus manually and set camera mode to Aperture Priority, pop up the flash and have one, two or three SB-600 wireless flashes magically give me the correct exposure using Nikon's CLS.
    Presumably this will only work if the lens is clever enough to report back to the camera the correct aperture setting.
    And of course, it needs to keep the aperture wide open while I focus.
    I know really old lenses won't do this, so I have to be a bit careful on Ebay...
    For the people suggesting a PC-E lens... one or two *tiny* quibbles:
    1. They cost double what my 105mm cost and that wasn't cheap
    2. One of the reasons I want greater DOF is that these little fishies move so fast I can't get them focussed exactly. Trying to line up several extra tilt/shift parameters at the same time would require about three extra pairs of hands!
    And as for bellows... I have no idea what you mean, but I suspect it's not going to be practical.
    Shun Cheung [​IMG] [​IMG] [​IMG] , Dec 18, 2009; 08:54 a.m.
    Patrick, it makes no difference what your photography interests are; the same physics still applies: if you need the depth of field at f22, you sacrifice sharpness. You simply cannot have it both ways as your original post suggests.
    However, there is now a different option: focus stacking​
    I'm not trying to defeat the laws of physics, I just want a lens that performs as well as can be expected at that end of the F-range. I have little or no use for the wide-open end of the F-range so if I can avoid paying for a lens which does both well I will. Of course lens makers don't typically cater for my whims.
    Yes I have tried focus stacking but it's a still-life sort of thing. I only ever had a fish sit that still once, and I spent the time taking an 11000x3000 pixel close-up panorama instead.
     
  25. I only do a little macro work, but I *love* my vintage factory AI-converted Micro-Nikkor 55mm/3.5.
    Believe the abovementioned posts: it is very sharp. AND it has no barrel or pincushion distortion. It works great on my D300, and makes a nice portrait lens at that.
    eBay might be a sensible choice. A lot to gain, little to lose.
    00VHpc-201873684.jpg
     
  26. You might want to try a 5omm.1.8 & ext. tube as a cheap alternative!!
     
  27. "I'm not trying to defeat the laws of physics, I just want a lens that performs as well as can be expected at that end of the F-range."
    But that is exactly the point!
    If you think one lens or another will be much better while limited by diffraction you are asking for a defeat of the laws of optics. At f22 I have never seen a lens that performs noticeably better than any other lens at f22. The lens is not limiting, the diameter of the iris is the limit. So why look for a better lens at f22?
     
  28. IMHO, the early non AI, Micro-Nikkor 55/F3.5 (Compensating aperture version) is likely "the" sharpest Nikkor I own. The 2nd sharpest (also IMHO) is the non AI 50/F2 (the 6 element/4 group edition). Both work real well on a Nikon DSLR.
     
  29. my two sharpest Nikkors are ancient non AI 50mm f2 (H) and 35mm 2.8 PC late black knob model shift lens.
     
  30. mab

    mab

    Why do you want to shoot at f/22? As others have pointed out, at f/22, even the sharpest lens doesn't look that much different from a Holga (I exaggerate, but only a little). Diffraction is the primary cause of softness, not the glass or specific lens design.
    If you're forced to shoot at f/22 because you need more depth of field, you have basically two options that will let you shoot wider. You can use a camera or lens that can tilt the plane of focus to be non-parallel to the film/sensor plane (that is, a view camera or a lens like one of the Nikkor PC-E lenses). This will allow you to keep objects on almost any plane (within the tilt range allowed by the camera plus the scheimpflug rule) in the same focus, at the expense of money. Or you can shoot at lower magnification (shoot wide and crop), which yields deeper DoF, at the expense of resolution.
    Unfortunately, physics is not just a good idea, it's the law.
     
  31. The lens you need is called "Kiron 105mm" the Lester A Dine version comes with a flash kit to mount on the front of the lens and was and is still used as a dental and forensic lens, you can only get it now secon hand since Kiron was a subsidiary of "kino Japan" and its already gone. you can get a Nikon mount copy for about US.500. and in my opinion is worth MUCH more than that. just cant get new ones any more. If you are lucky enough to find one from a retired surgeon or dentist or C.S.I, buy it, theese people tend to take very good care of their equipment.
     
  32. If you really want sharpness, I would avoid f22 or even f16, especially on the DX format. Diffraction will rob some sharpness from the optics.​
    This is generally true. If you look at the MTF diagram for almost any lens, the peak performance tends to be around 2-3 stops above the widest aperture and 2-3 stops before the smallest aperture. That's a fact and you can verify it.
    So, to avoid stopping it down all the way and miss the sweet spot in the middle, you can use a tilt-shift lens. This will allow you to take advantage of the Scheimpfug effect. With it, you can now stop down to f/4 or f/5.6 and still get fully focused from the foreground to infinity and be able to use the peak performance of your lens in its sweet spot.
     
  33. Just to throw a few numbers in there: the 'blur disk' of a point of light due to diffraction is roughly f-stop times wave length.
    Visible light has a wave length of around 0.5 micrometer, so at f/22 you are looking at a resolution due to diffraction of 11 micrometer. Compare that to the pixel size of your camera. More exactly I think it's 1.2 * f-stop * wavelength for a round iris and an object at infinity (think: star), and twice that for 1:1 magnification.
    http://en.wikipedia.org/wiki/Diffraction
    So going to a larger image format helps (bigger pixels at same resolution), but changing the focal length doesn't make any difference, since it is the ratio of iris opening to focal length (the f-stop) that gets multiplied by the wavelength. Going to UV light (or particle beams, as in electron beam lithography) would also help, but is often not practical :)
     
  34. Icy London. This is my son (now 11 years old!)
    I know you don't want it for this but as a bonus the f2.8D 60 can work as a portrait lens. This is FX, but an effective 90mm focal length on the D300 would work pretty well for this. I like this lens because of it's sharpness; not adequately showcased here. I was using the D3 as a point and shoot, and had forgotten that the last time I used it was in low light, hence the ISO 6400.
    00VHzA-201925584.jpg
     
  35. And here's one at f22. Again, I'm aware that this is not how you want to use the lens; it's just as a contrast to the weather in Perth.
    00VHzK-201927584.jpg
     
  36. OK so here's a question. My aim here is to allow me to take macros of the bigger fish - the 105 covers small closeups extremely well. We know that switching the camera to a D700 would accomplish this extremely well by expanding the field of view by 1.5x at the same time as increasing sharpness by spreading the same 12MP over a full frame.
    Whatever I do with 55 or 60mm lenses is never going to match the sharpness that would give me.
    So my question is... assuming lenses of 'equal quality' set to say f11 to avoid too much diffraction limiting, which would give better DOF:
    1. I move back to twice the distance and use my 105
    2. I stay at the same distance but use a 55mm
    Second question: Exact same question but I'm using f22, so factor in loss of sharpness due to diffraction.
    Option 1 is inconvenient due to space limitations for fish, but generally not for flowers.
     
  37. This is too technical for me so I should probably stay out of it. But... surely spreading the same 12 MP over a a full frame would decrease sharpness, not increase it!
    I suspect that DOF for 1. would be the same as DOF for 2.
    Let's hear what the experts have to say.
     
  38. I have captured very sharp pictures with 50mm F1.8 a bargain lens, Then 60mm, and 105mm VR but to me 135mm DC F2 is the sharpest.
     
  39. Alastair Anderson , Dec 18, 2009; 09:36 p.m.
    This is too technical for me so I should probably stay out of it. But... surely spreading the same 12 MP over a a full frame would decrease sharpness, not increase it!
    I suspect that DOF for 1. would be the same as DOF for 2.
    Let's hear what the experts have to say.​
    I too suspect that 1 and 2 would be about the same.
    The resolving power of a lens is measured in lines per millimetre and if you have more millimetres you have more lines. To be precise, you have 1.5*1.5 more pixels which is 2.25 times as many 'useful' pixels. This assumes that both cameras are diffraction limited which they will be at f22.
    In other words when using the 105 at f22 on my DX camera I could just as well have a 4MP sensor (or less? I'm not sure exactly how much I'm losing to diffraction)
     
  40. The F8 or so as sharpest aperture is no longer true in many cases. The 17-55 and 24-70 both hit their peak sharpness early on, at about F4 to 5.6. You can see it over at Photozone.
     
  41. All,
    Thanks to all of you who posted pics - they've been great! I think they've added interest to this post.
     
  42. My sharpest lens: Nikon D300 is Nikkor AIS 55/2.8 http://joergvetter.oyla.de/cgi-bin/hpm_homepage.cgi
    BR Joerg
     
  43. just try to listen. the 105 with 1.5 crop factor will become a 175 aprox, very narrrow for some purposes but on the FX format would be a much better proposition, the Kiron 105mm 2.8 is great and the voightlander125 apo lanthar is also great.
     
  44. Any micro Nikkor 55 or 60 will be ok at f22 at close range. I say ok, not great; I like my macros very sharp. If the fish is too fast, then you should probably start thinking about automatic optical triggering of your camera when the subject is in focus. Even at f22 DOF is shallow when going close.
     
  45. Jose Antonio Ramirez [​IMG] , Dec 18, 2009; 05:37 p.m.
    The lens you need is called "Kiron 105mm" the Lester A Dine version comes with a flash kit to mount on the front of the lens and was and is still used as a dental and forensic lens, you can only get it now secon hand since Kiron was a subsidiary of "kino Japan" and its already gone. you can get a Nikon mount copy for about US.500.​
    Since I'm already discarding significant sharpness from my 105mm Nikon by (1) using it at F22 and (2) using it on a DX body... Wouldn't you think that I'd be wasting my time and money trying to buy a rare and expensive Kiron 105mm, which doesn't give me the wider angle I'm looking for anyway?
    Jose Antonio Ramirez [​IMG] , Dec 19, 2009; 02:47 p.m.
    just try to listen.​
    Maybe I need to try a bit harder...
     
  46. For all those people trying to get me to stop using f22... this is the sort of shot I'm talking about.
    This is f22, and if I drop to f11 I lose too much DOF while gaining only minimal sharpness. If I increase to to say f57 the DOF is perfect but you literally can't tell because the whole thing has lost so much sharpness.
    Link is to full res pic. Might be hard to view but shows sharpness and DOF better
    [​IMG]
    http://i923.photobucket.com/albums/ad73/onetruepath/Catfish/DS2_5706LTA.jpg
     
  47. Excellent macro work Patrick. Stick to the methodology you know best.
    Highly recommend the 60mm AF-D Micro Nikkor (why did I sell mine - doh.......)
    I now use the Ai 55mm f/3.5 Micro Nikkor - is as good as the above mentioned lens IMHO, focus throw is nice and long for precise focusing.
     
  48. I'm not even going to pretend i know how to help, but wow that image looks nice.
     
  49. Patrick, I'm struggling to get my head around this. My IQ is plummeting in old age; not that there ever was too far for it to fall of course.
    I can intuitively understand that in order for you to get the field of view that you need, you have to move back from the subject if you are to use the 105 lens on the D300, that this backwards step wouldn't be necessary if it were on a D700 and therefore this would give you better resolution. However I'm battling to grasp the maths. Why 1.5*1.5? It's not a square image. And then how do you get to 4MP? Are you dividing 12 by 2.25 and then taking something away because of diffraction?
    Is there anyone else who can shed some light on this, or is it perfectly clear to everyone but me?
    Incidentally I discussed this with Alex. He says you need to experiment to see what works but suggested that if there is a wider lens that focuses close enough this may give you the best perspective. I think this is a very interesting idea.
     
  50. Alex is talking about 35mm or wider. He seems to think that the pc 35 lenses screw out a long way.
     
  51. Forgive me for musing here, but no-one seems to want to put me out of my misery!
    If I put a DX lens on my D3 I effectively have a 5MP camera, and presumably that is 12/2.25. Ok, so it doesn't matter that it's not a square image because we're talking about resolution per square centimetre or inch or whatever.
    I can accept that this somehow works in reverse when you want to go wider on a DX camera, but I need a diagram or a simple analogy or something to make it clear.
    Anybody?
     
  52. ShunCheung

    ShunCheung Administrator

    For all those people trying to get me to stop using f22...​
    Patrick, nobody is telling you to stop using f22. What you should stop doing is putting f22 and sharpest in the same sentence.
    In other words, what you are facing is a very simple trade off between the amount of depth of field and sharpness at where the focus is. That is very basic physics that we cannot violate. One way to get around it is to use stack focus to merge multiple image captures.
     
  53. Shun - Patrick my not be stating his case as accurately and tersely as possible. Let me try:
    Averaged over the range of depths that folks like Patrick are interested in having as sharp as possible, and constrained by not being able to use focus stacking because of subject movement, his solution, ie, ~ f/22, is probably optimal.
    It doesn't matter one wit if a 2 micron depth of field around the exact focus plane is a bit sharper at f/8 than at f/22, what matters practically is that if +/- 1 cm needs to be as sharp as possible, f/22 is a much better solution.
    Cheers,
    Tom M.
     
  54. Thanks Tom for stating that so clearly. That's exactly what I was trying to say.
    Now since you helped me out I think I should help Shun out:
    "You used 'sharp' and 'f22' in the same sentence! Foul!"
    Alastair:
    You seem to be almost there. It's not the same as going from FX lens to DX lens and actually losing most of your pixels, since in both cases you are using all the pixels.
    Think of the lens as projecting an image onto the sensor with a certain number of lines per millimetre, in a square grid.
    My sensor picks up less lines than yours: Even though I have a higher density of pixels per square millimetre, they aren't doing me any good. Simple as that.
     
  55. ShunCheung

    ShunCheung Administrator

    It doesn't matter one wit if a 2 micron depth of field around the exact focus plane is a bit sharper at f/8 than at f/22, what matters practically is that if +/- 1 cm needs to be as sharp as possible, f/22 is a much better solution.​
    Tom, that is perfectly fine, but in that case since diffraction is going to rob some sharpness anyway, there really is no point to worry about which macro lens is the sharpest. Therefore, the original question and this entire thread is moot.
    Walter Schroeder already explained that very well 2 days ago:
    If you think one lens or another will be much better while limited by diffraction you are asking for a defeat of the laws of optics. At f22 I have never seen a lens that performs noticeably better than any other lens at f22. The lens is not limiting, the diameter of the iris is the limit. So why look for a better lens at f22?​
     
  56. Hi Shun - I agree totally.
    What I was indirectly trying to point out was that while photographers may not be consciously aware of doing so, different individuals intuitively try to minimize a penalty or cost function that is most relevant to their particular photographic application, constraints, etc..
    For example, user #1 can avail themselves of focus stacking, so he/she probably will want to select and adjust a lens so that the peak resolution (e.g., in line pairs per mm @ (say) 50% contrast, averaged over some fraction of the frame, in the plane of perfect focus) will be as high as possible. However, a second user, #2, may need to get a usable image in one shot. Thus, user #2 might select and adjust a lens so that the average resolution within a cm of focus is optimized. Each optimization effort will lead to different results.
    Unfortunately, too many discussions of this type proceed into rancorous discord because the presence of two very different goals (as suggested above) is not recognized by the participants. Thus, one of the benefits of a thread such as this is having the participants recognize their differing goals through discussion.
    Just my $0.02,
    Tom M
    PS - For the sake of completeness, I suppose I should mention that hypothetically, it's possible that lens aberrations and other problems could be so hideous that they could overwhelm diffraction even at f/22, but among lenses of any reasonable quality, I doubt that's going to happen.
     
  57. My 105mm lens loses sharpness rapidly at smaller f-stops than f22. However it is still easily sharp enough for my requirements at f22. See my pic above, that's all I ask of a lens.
    The sense I get from everyone else on the forum (except for Jose Antonio Ramirez who thinks I need a sharper 105) is that any 55mm or 60mm Nikon lens with 'Micro' in the name will do what I want. Furthermore since I'm probably going to be focussing manually I might as well get a cheaper one.
    The sense I get from you and Walter is that any lens at all with the appropriate focal length will do.
    Well it's always been my belief that my 18-200mm DX VR zoom is a rather nice travel lens, but from the sound of it you're telling me I would be better off using that lens for macro work as well if I'm going to use f22, since the sharpness will be identical and I'll have the benefit of the zoom?
     
  58. For me this is a particularly interesting thread. I realise that for someone who stated earlier "...I should probably stay out of it," I've said more than enough already. However, I'd like to draw everyone's attention back to Patrick's question about the difference in DOF between using the 105mm lens at twice the distance or changing to a 55mm lens. We both said that we suspected that there would be no difference. Can anyone confirm this?
    Patrick, are you convinced about that?
    Also, I want to reiterate my friend Alex's observation that a wider lens may provide a better perspective. Your original question specified that you were looking for a focal length between 50 and 60mm, but I remember reading somewhere that a 21mm lens is particularly good for photographing scale models because it provides a persective that makes them look real. I'm not suggesting 21mm in this case, but I think 35mm is feasible. Perhaps the problem with this is the further loss of resolution. By the way the early pc 35mm lenses are not that expensive and the f2.8 versions work on DSLR's.
    I think you were too quick to dismiss Jeff Adler's bellows idea. You can mount almost any lens on bellows and thereby focus really close. It's a little more complicated than simply mounting a macro lens, but does afford additional possibilities.
    Perhaps you should work a couple of extra days in the new year and buy yourself a bellows unt. Or a D700. :)
     
  59. I think Alastair understands perfectly the point but I find sometimes this threads to end up on endless pointless discussions, if I tell the OP wha is the "sharpest lens with the better workable distance for his nikon mount camera" then it becomes a subject of budget or something else. the question would have been better formulated as: Sharpest and least expensive nikon lens for macro work"
    I echo this proposition keep what you have get some bellows or a d700.
     
  60. Patrick Hamlyn @ Dec 18, 2009; 09:11 p.m. asked:

    "...So my question is... assuming lenses of 'equal quality' set to say f11 to avoid too much diffraction limiting, which would give better DOF:

    1. I move back to twice the distance and use my 105
    2. I stay at the same distance but use a 55mm "

    I may be wrong, but I couldn't find anyone that actually addressed the above question which was asked quite early in this thread, and is quite informative, so let me give it a shot.
    To address questions like this, the on-line depth-of-field calculators are a tremendous resource. I prefer http://www.dofmaster.com/dofjs.html, but there are others.
    To make the ratio of the lens focal lengths and distances both exactly a factor of two, I'm going to assume that the two lenses being compared are a 50mm and a 100mm, both mounted on Patrick's D300. Lets first do the DOF calculations for distances of 1 ft and 2 ft. I enter the above items and the aperture (= f/11) into the on-line calculator mentioned above and see that the total DOF is 0.05 feet for both lenses -- ie, *EXACTLY THE SAME*.
    Next, lets re-do the calculation for a 50mm lens at 5 feet and a 100mm lens at 10 feet, with all other assumptions the same as before. In this case the total DOF comes out to 1.36 feet and 1.34 feet, respectively, so we are starting to see a slight difference between their depths of field.
    Perhaps differences in DOF under our assumption of constant image size show up at really long distances, so, as our final calculation, lets re-do the numbers assuming a 50mm lens at 50 ft and a 100mm lens at 100 feet, again, both at f/11 and on a d300 body. In this case the total DOF comes out to infinite and 260.7 feet, respectively. Now, there is clearly a big difference between the DOF of two lenses. Obviously, the reason for this is that with subjects this far away, the subject is within the hyperfocal range of the 50mm lens at f/11.
    The bottom line is that for a constant reproduction ratio (ie, holding the size of the subject on the sensor constant), there will be essentially no differences in DOF between the two lenses when used in the macro range. Of course, there will be other differences between them. For example the ratio of the diameters of the blur circles of a point at infinity for the two lenses will be very different when used in the macro range. The 100 will give larger blur circles and take in less of the background, but that's not the question under discussion.
    In any case, at least some of the questions raised in this thread can be answered by using a DOF calculator. Specifically, an example of a question that can be answered is that of DOF changes when switching from a d300 to a d700-sized sensor. Questions that the DOF calculator can not answer include those dealing with no-ideal lens phenomena, eg, aberrations, diffraction, etc.
    HTH,
    Tom M
    PS - Beware of lenses, especially modern ones, whose true focal lengths decrease as you focus on closer objects.
     
  61. The answer is that at the same detail magnification, depth of field will be virtually identical in your example. Perspective and background rendition will not be identical. The treatment to show this can be found in any textbook on optics.
    However, using a DOF calculator cannot in itself "prove" anything. It is a model and many of the simplified assumptions of the model break down for the close-up range.
     
  62. Bjorn, I never said that using a DoF calculator will prove anything. OTOH, the DoF calculator is, exactly as I said, an incredibly useful resource for folks who are not likely to open an optics textbook or delve into the approximations used in such calculators.
    Don't get me wrong, when it is possible, I'm certainly not against a more technical approach as I ran a program (between 6 and 15 Ph.D.s) in laser optics at one of the national labs here in the US for 12 years. It's just that when dealing with less technical folks, allowing them to play with the numbers (albeit approximate), can be incredibly instructive.
    Cheers,
    Tom M.
     
  63. The main problem is that the DOF calculator can be used outside its parameter domain and potentially misleading conclusions thus are drawn. Most people don't even realise it is a (geometric) model but they are happily using the output as arguments in a debate. I've seen this too many times to list them all.
    Stating the DOF calculator *predicts* something is better. Verifying the prediction is the next step and may not always be possible.
     
  64. Thanks Tom that's useful - so they'll be about the same. But now I have to try and get my head around the fact that the DOF increases significantly if I switch to the D700.
    Example: Closest focus is around 32cm. So I stuck all that into the DOF calculator you linked. It says for the D300 focussed range is from 0.28cm in front to 0.28cm behind, ie depth of 0.56cm.
    Switch to the D700 and it goes up to from 0.43cm in front to 0.42cm behind, for a total depth of 0.85cm. Ignoring any losses due to diffraction for the moment.
    Whatever the reason, apart from increasing my own 'circle of confusion', it's yet another argument in favour of the D700...
    The only reason I rejected bellows is that 99% of my macro work is 'freehand' ie camera in right hand and flash in left hand, chasing after little fishies/froggies/insects/flowers. It just struck me that anything involving bellows would require focus rails and a tripod or more likely a bench.
    Alastair's (or Alex's) pointing out that there will be a significant perspective change with say a 35mm Macro lens is interesting too, and I can test that up to a point with say my 18-200 lens, although I may be crippled by 'closest focus' limitations.
    It would be interesting to see if there is a noticeable distortion of say a fish or a frog the same way that a wide angle portrait lens distorts a face.
     
  65. Here's a comparison of f8 versus f22 for the 105mm Nikon Micro. Embedded pics are half resolution, links below give full resolution. I cropped them both down quite a bit as well, no other processing at all. f8 is noticeably sharper on the focussed plane. Scratches and dirty spots on the glass are noticeably more visible on f22. But f8 loses significant detail on the fins and barbles around the mouth etc.
    I would love to see a comparison picture taken with a D700, with say f16 added as well.
    f22
    [​IMG]
    f8
    [​IMG]
    f22 full res
    http://i923.photobucket.com/albums/ad73/onetruepath/Catfish/DS2_4999C.jpg
    f8 full res
    http://i923.photobucket.com/albums/ad73/onetruepath/Catfish/DS2_5002C.jpg
     
  66. Hi Patrick -
    I'm glad you found the DoF calculator useful.
    The reason that the DoF shown by this calculator goes up when you switch from an APS sensor to a full-frame sensor (assuming the same lens and constant f-number) is almost certainly that they are using one particular convention for the diameter of the circle of confusion which defines "in-focus" vs "out-of-focus". Specifically, this convention assumes that the threshold CoC diameter is a fixed percentage of the sensor size, typically, the diagonal of the sensor in mm divided by 1500. Thus, when you go to a sensor, say, 1.5x larger, circles of confusion 1.5x larger in diameter are accepted as being in-focus. This leads to a larger DoF.
    You can check the effect of size of the CoC on the DoF using the same calculator. Instead of specifying a camera or sensor, scroll all the way to the bottom of the same list and you will see options for a wide range of circle of confusion diameters.
    There is a great, very readable article in Wikipedia on the topic of "circles of confusion": http://en.wikipedia.org/wiki/Circle_of_confusion. There are many subtleties in selecting an appropriate and useful standard for the threshold diameter of the CoC, many of which (including alternate standards) are discussed in the Wikipedia article. For example, when comparing two different size sensors, are you going to enlarge both resultant images to the same final size, or are you going to enlarge them both by the same numerical factor (e.g., 5x, 10x, ?). Each scenario leads to a different but equally valid definition of the threshold CoC. If you are interested in this topic, I strongly suggest that you read the wikipedia article closely.
    Finally, w.r.t. using short FL lenses for macro work, I fully agree with the previous person who stated that the perspective given by a short FL lens can make many small subjects look more realistic. I know that somewhere I have a great shot illustrating this, but I must not have keyworded it, so I can't lay my hands on it at the moment. Basically, it consists of two shots of the objects under the Xmas tree, one taken with my 14/2.8, and the other taken with my 105/2.8 VR lens. The 14 mm lens produced a surprisingly pleasant and realistic perspective. If I run into that image, I'll post it.
    HTH,
    Tom M
     
  67. As a photographer who has used various medium format cameras including Hasselblad, varous Zeiss lenses and Schneider larger format lenses, very few Nikon lenses come anywhere close to great lenses like the Planar, Apo Lanthar or Apo Symmar. Nikon lenses maybe sharp, as in the 50 mm 1.4, but the distortions, vignetting, etc. all compromise. Furthermore, all of Nikon's lenses have a curved field. The Planar is the ultimate wedding lens. It focuses evenly across a flat field an has totally predictable dof characteristics with beautiful transitions. Nikon lenses also suffer from rather harsh colour and shadows. For the best lenses on a Nikon? Use the pro lenses and a few Tamron or Sigma versions.
     

Share This Page