1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

D800, mirror slap and AF-P lenses

Discussion in 'Nikon Chat' started by ChrisNewman, Jul 20, 2017.

  1. ChrisNewman

    ChrisNewman Well-Known Member

    It’s not just my shaky hands - I think it must be mirror slap

    I’ve checked out the sharpness of my AF-S DX VR 55-200/4-5.6G IF-ED on my D800 with a series of shots with the camera mounted on the Velbon CX 440 that was my free gift when I first subscribed to AP, and I had stopped using because its plastic 3-way head, in particular, lacked rigidity.

    I mounted an AP test chart on our kitchen window, and set the camera where the lens wouldn’t quite resolve the closest end of the lines at 200 mm. I set f/8, and started at 1/1000 sec, using a remote control, first taking my usual sequence of 3 bracketed exposures at continuous high speed, with VR switched on. Then I set 3 sec exposure delay, and took another 3 shots, and then turned off VR for a further 3 shots using Live view (to check the best the lens could achieve). I repeated this sequence at ⅓ stop slower shutter speeds to 1/60 sec. I set auto ISO, and tried to maintain fairly low ISO sensitivities by starting in sunshine, and continuing when the chart was shaded or under cloud.

    Right from 1/1000 sec the instant shots were inferior to those with exposure delay, and as might be expected, Live view gave the best results.

    Even at 1/500 sec the resolution on the instant shots had begun to deteriorate, and by 1/320 sec, which I would have expected to be pretty safe for hand-holding a 200mm VR lens, results were very poor.

    The resolution of phase detection with 3 sec mirror delay began to deteriorate by 1/250 sec (I wonder whether this was due to the tripod head continuing to bounce from mirror slap, whist hand-holding might be better damped).

    Live view gave good results until exposures approached 1/100 sec. With the weather this end of the summer, I’d had a long wait for a sunny afternoon, but it was breezy, and it was seeing the camera moving in the wind that had led me to abandon the CX 440 tripod.

    So I now know the kick I feel when shooting my D800 is enough to affect shots with my DX VR 55-200, and I assume this must be due to mirror slap, but at least I know what to expect. I’d be grateful for any suggestions for combating it other than using very high shutter speeds if the light is adequate, or trying to shoot with exposure delay or in Live view. At least I now know what to expect with that lens

    I’d also be interested to hear from anyone else with experience of using a D800 with very lightweight telephoto (my 55-200mm is listed as 335g).

    With thanks,

  2. nimbus

    nimbus Well-Known Member

    Vibration Reduction systems are supposed to be switched off when using a tripod. It apparently confuses them and causes them to induce shake.
  3. Craig20264

    Craig20264 Well-Known Member

  4. cliveva

    cliveva Well-Known Member

    I have found less need for sharping when using a 24mmf1.4 compared to the 24-120f4VR used hand held, I had always put this down to the 24 being twice the price.
    At slow shutter speeds I can hear the slap, so immediately boost the ISO and retake, as I have learnt from experience that it will have shake induced softness.
    How soft are your images?
    As I can take pin sharp images with a 70-200mmf2.8VRII, on and off a tripod, I wonder if it is your lens/camera combination, or even a faulty damper on the mirror?
    I must admit it is a challenge to shoot sharp and multiply the usual focal length / shutter speed equation , by 3; i.e. 60mm/1/200, for safety!
    I would be interested in how you would get on, using a pro lens, in your shooting experiment.
  5. nimbus

    nimbus Well-Known Member

    A modern prime should be sharper than a modern zoom, after all it is designed for and functions at a single focal length, not a great many different ones which may range from a wide angle to quite a long telephoto. The zoom has improved massively, but will always have greater compromises in it's design and production.

    I certainly find it easier to take pictures at slower shutter speeds with a CSC than with anything that has a flapping mirror.
  6. ChrisNewman

    ChrisNewman Well-Known Member

    That’s why I turned off VR for the Live view shots to check that the camera and lens could give decent results on a floppy tripod provided the system wasn’t disturbed by mirror movement.

    I only opened this thread because I was pretty sure my hand-held results from the 55-200mm lens were being degraded by something more than camera shake, and thought there was a possibility that someone might have some advice that was new to me, but some of the first responses, including yours, seemed unconvinced that my hand-holding steadiness and technique weren’t the main cause. I did this testing just to make certain that hand-holding wasn’t the cause, and the best way I could think of was to use a support that wouldn’t generate any movement itself, but unlike a top-of-the-range heavyweight tripod and head, wouldn’t be able to prevent any movement the camera generated. So it seemed logical to use the VR, and three shots in quick succession for exposure bracketing, that I normally use for hand-held shooting.

  7. nimbus

    nimbus Well-Known Member

    I certainly consider that vibration can be induced by the mirror system of SLR type cameras, it's an Achilles Heel and an advantage of CSCs which lack a mirror and all the mechanicals to operate it. Sony SLTs as you are probably aware use a fixed translucent mirror and an EVF, so are a half-way house, which also eliminates this issue. I suppose it is likely that the effect of any mirror system induced vibration is made more apparent by higher resolution cameras such as the D800.
  8. ChrisNewman

    ChrisNewman Well-Known Member

    I’ve uploaded crops of some out-of-camera JPEGs. It was disappointing photos of a heron that posed obligingly on the opposite bank of the canal that prompted me to look for a reason for the poor results I was getting from my 55-200mm. I’ve uploaded one of the better ones (at 1/320 sec). DSC_7608 Heron by Ovaltine Court, beside the Grand Union Canal, Kings Langley.JPG
    I've also uploaded a typical example of each shooting condition from my testing, also at 1/320 sec; instant DSC_8674 Phase detection, 0 sec delay, VR ON, 320th sec, Exposure comp +⅔ EV, ISO 500.JPG Hand-held with 3 sec exposure delay DSC_8667 Phase detection, 3 sec delay, VR ON, 320th sec, Exposure comp +1 EV, ISO 320.JPG and Live view DSC_8680 Contrast detection, 3 sec delay, VR OFF, 320th sec, Exposure comp +⅔ EV ISO 400.JPG (The instant phase detection shots on the floppy tripod didn’t get any worse below 1/320 sec; I don’t know whether that’s the frequency when VR becomes effective, or something to do with the mirror slap.) You can see the Live view shot isn’t that far from pixel sharp, but I’d be happy to get quality equivalent to the 3-second mirror delay when hand-holding.

    I usually set Auto ISO to match the shutter speed to the focal length of the lens. (I think the camera might up the shutter speed to 1/320 sec for the narrower field of view of the DX 55-200mm, although in principle I’d rather match sharpness to the pixel level than the field of view.) With my other stabilized lenses I find most hand-held shots are only slightly less sharp than I’d expect with a tripod and Live view, and I settle for that. (There are some where I wasn’t as steady as usual, or the VR doesn’t seem to have steadied one of the three shots in the bracketing sequence properly, and I get poor results if I forget to switch the stabilization on.) So I don’t have much incentive to repeat the tests with other lenses. My Sigma 150-500mm should be most sensitive to changes of the camera pitch, but I rarely use it hand-held, it’s rather soft anyway at 500 mm, and I think most significantly it’s longer and far heavier than my 55-200mm, so it’s moment of inertia will be many times that of the D800 and 55-200mm combination. My 85mm DX macro is close to the 55-200mm in size and weight, but it’s focal length is less than half, and macro shots usually involve an awkward position and poor light, whilst VR isn’t supposed to be as effective at close range, so as well as getting sharper results than at 200 mm with the 55-200mm, I’m probably more forgiving of that lens.

    Your 70-200mmf2.8VRII is another long, heavy lens with a huge moment of inertia to resist changes of pitch in the camera from mirror slap. I think I’ve heard that the D810 has a softer mirror action than the D800, despite a faster shooting rate, and I’ve read that Canon went to great lengths to soften the mirror action on their 5DS cameras. I guess Nikon just put an “ordinary” shutter in the D800, and then found mirror slap could limit its high-resolution performance, and tried to address the problem in the D810 (what will the “fast, high-resolution D850 have?), whilst my lightweight, long focal length 55-200mm suffers more than most lenses. (I think I used to get rather better hand-held results with it on my D90, with its smaller mirror, despite a lower pixel count).

    Last edited: Aug 22, 2017
  9. cliveva

    cliveva Well-Known Member

    Your Heron looks similar to results I may get with my 70-300mmf4-5.6VR, much poorer than the 70-200f2.8, but I shoot RAW and with judicious use of sharpening can retrieve acceptable sharpness. If I want to go light I know there will be more to do in the "dark room", if I know what I will be able to shoot, I will take the heavy stuff!!!
    Interesting thought about the size of lens dampening the vibration, yet the 24mmf1.4 appears sharper than my 60mmf2.8 micro, when you look at the pixel level. So I had put this down to the quality of the lens, affecting the image quality.
  10. ChrisNewman

    ChrisNewman Well-Known Member

    Yes, I was on a group walk when I saw the heron (I did catch the others up eventually!) If, instead, I’d set out expecting to shoot a heron, I’d have taken my Sigma 150-500mm and a tripod. But I would like to be able to get decent results hand-held from a telephoto lens light enough to carry just in case. I shoot RAW + JPEG, but rarely get round to editing the RAW files. I know judicious sharpening would improve results somewhat, but it can’t recreate detail that was never captured.
    With a focal length of only 24mm, the camera would need to be turned more than twice as far to create a given level of blurring than a 60mm lens. If shots with your macro suffer more from hand-holding than those with your 24mm, I suspect that might be the reason. My Sigma 8-16mm doesn’t have image stabilization, but shake or other camera movement is hardly ever a problem except in light dim enough to need slow shutter speeds!

  11. Learning

    Learning Ethelred the Ill-Named

    When it becomes necessary to post photographs of test charts, haven't we got something wrong?
  12. PeteRob

    PeteRob Well-Known Member

    The heron shot looks like you are asking too much of the lens.
  13. nimbus

    nimbus Well-Known Member

    My question is what area of the frame is this from? If it is from the full-frame it is not good, but if it is from a section it is as might be expected. It actually looks oof.
  14. nimbus

    nimbus Well-Known Member

    It's a good way of testing lenses to our dissatisfaction!
  15. ChrisNewman

    ChrisNewman Well-Known Member

    cliveva asked “How soft are your images?”, and posting the test chart shots seemed the clearest response. What’s wrong with that?

  16. ChrisNewman

    ChrisNewman Well-Known Member

    The lens is DX, and I accepted the camera’s auto crop to APC-S rather than full frame, 4,800 × 3,200 px. What I uploaded is a simple crop from that to 800 × 600 px.

  17. ChrisNewman

    ChrisNewman Well-Known Member

    Please can you explain, because I don’t understand what you mean?

    I had a fairly close, extended view of a heron in good light (although I would have preferred the bird to face the other way), and wanted to take a photo. The most suitable lens I had with me was my 55-200mm telephoto, and because the heron was distant relative to its size, it was logical to use its maximum focal length of 200 mm. An aperture of f/8, one stop short of its maximum, seemed appropriate to me. The quality of the resulting shots (I took several) reminded me that my results with that lens hand-held on my D800 always fall short of what it can achieve with Live view on a tripod. The Live view shot of the test chart I uploaded shows that, without disturbance from the mirror, it can give reasonably sharp results in those conditions. But my shots of the heron, taken through the viewfinder without exposure delay, blurred out much of the detail the lens is capable of producing.

    My testing shows that the loss of quality isn’t due just to my hands shaking, or phase detection giving less accurate focusing than Live view, and I’m hoping that someone on the forum might have some advice, particularly in how to limit the effects of what seems to be mirror slap.

  18. nimbus

    nimbus Well-Known Member

    If you are experiencing mirror slap and it is causing such issues it is difficult to find anything to recommend to deal with it, in that you can't remove the cause and therefore the effect. Not helpful I am aware. Might I suggest therefore trying a different model of body, either owned by a friend or rental and ascertaining whether such action reduces it, eliminates it or mollifies you to what you regard as an acceptable extent. In other words a D610, D750 or a D810. Beyond that I could only suggest the expensive course of changing to a Sony SLT or FF CSC.
  19. ChrisNewman

    ChrisNewman Well-Known Member

    Yes, thanks, it had occurred to me that a change of kit could be an (expensive) solution. Also, if I could engineer a chance to try someone else’s D800, I could at least be certain that my mirror isn’t coarser than it should be. I only use my telephoto occasionally, so buying something new would be rather a sledge hammer to crack a nut. I value the light weight of my 55-200mm, as most days I carry it without ever taking it out of the bag, but the extra weight of a 70-300mm, especially full frame, should ballast against the mirror slap, as well as giving me welcome extra reach (see my comments in my opening post for this thread).

    I’d see a D750 as rather a downgrade, given the lower pixel count, and, I think, slightly more limited control set, and a D610 even more so, although I’d welcome the D750’s mobile LCD. I’d love to have a D810, particularly for its improved dynamic range, and highlight-weighted metering, but it’s a lot of money.

    I’d love a Sony a7RII for use on a tripod (even better if it had the tilting viewfinder of some Panasonics). But I really don’t want to lose my OVF for the hand-held shots I take most of the time (yes, I realise that it’s the OVF’s mirror causing the problem with my 55-200mm!) Also, I’m not sure I could match the abilities of my current system with a Sony fit outfit at an acceptable weight. (I believe Sony’s SLTs use an EVF, which seems to me to combine the disadvantages of DSLRs and mirrorless systems.)

    With thanks,

  20. PeteRob

    PeteRob Well-Known Member

    Assuming it is in focus, which of course it might not be, it looks like results we would get from a Canon SX40 HS at large zoom ratio. Basically the optics aren't good enough to pixel peep.

    Sorry for late reply I was travelling.

Share This Page