The short and the long of it Using the lens toward the long end requires a more demanding technique from that which works well at the short end.
Challenges at the long end I recently tested and reviewed in this blog the Panasonic FZ200 which has the same angle of view at the long end as the 100-300mm on M43. Both kits present similar challenges at the long end making sharp pictures more difficult to achieve than they are at the short end. The problem is that everything is working against sharpness as the lens is zoomed out and subject distances increase. The main issues appear to be:
* Even with impeccable technique in ideal conditions and absolutely no camera shake at all,
b) Optimum resolution is achieved at f8 whereas at the short end peak performance comes in at about f4.5.
* The effect of camera shake is greatly magnified at the long end and
* OIS is not able to counteract the effects of fast jitter typically produced when hand holding or even on a less than sturdy tripod. Panasonic OIS works well to stabilise the appearance of the preview image in the EVF mainly due to it's ability to compensate for relatively slow camera/lens movements.
* Atmospheric haze and heat distortion are significant detractors from image clarity as distance from the camera increases.
* Autofocus is a little less confident at the long end but simultaneously greater AF precision is required due to the small depth of field at long focal lengths.
* My tests show that the standard rule for minimum safe hand held shutter speed is too slow by about one EV step at maximum zoom. So the rule would indicate that a hand held shutter speed of 1/600 sec should be safe but I find that reliable sharpness hand held requires a shutter speed of 1/1250 or faster. Even at 1/1000 sharpness is not consistent at the long end. With slower shutter speeds I will get an occasional sharp frame but lots of unsharp ones.
Same subject as the photo below, different day and slightly different camera position. 100-300mm at 300mm, 1/500 sec, f8, cable release, E-Shutter, sturdy tripod. Sharp, clear, no artefacts. |
100-300 at 300mm. Handheld, 1/1000 sec at f5.6, E-Shutter. Look at the top row of windows. The frames appear to be bent. This is one of the E-Shutter artefacts referred to in the text. |
Optical performance
Resolution and contrast are excellent to outstanding at the short end. I rate the 100-300mm at 100mm as equal to the 35-100mm f2.8 with both at f4. It really is a top performer at this focal length right from maximum aperture.
In the middle of the focal length range, 150-200mm, it is excellent. On my tests the 100-300mm has better resolution than either the 45-150mm or 14-140mm at 140mm. Again there is little to be gained from stopping down apart from reduced corner shading.
Towards the 300mm long end resolution and contrast drop a little from excellent to very good. There is a definite improvement as the aperture is stopped down from f5.6 to f8. I found no benefit to further reduction in the lens aperture in fact on several of my test sequences resolution fell slightly at apertures in the f9-f10 range.
Corner shading is obvious at wide apertures, becoming less so as the aperture is decreased.
Flare is generally well controlled especially if the lens hood is ued which I recommend at all times. Veiling flare becomes evident if the lens is pointed directly into the sun or sunlight reflected off water. However shadow details are quite well preserved and can be revealed in Camera Raw.
Chromatic aberration and Purple fringing CA is corrected in Panasonic cameras so is not usually an issue. However small amounts of both CA and PF can appear at bright, high contrast edges. They are easily corrected in Camera Raw.
Distortion is minimal at al focal lengths.
Centering My copy appears to be well centered with no significant asymmetric unsharpness at any focal length as long as no pressure is applied to the inner barrel.
Bokeh is generally smooth although at some focal length/focus distance combinations I did see some tendency to a jittery appearance in out of focus backgrounds.
Artefacts with E-Shutter The following phenomena are mostly seen at maximum zoom. If the E-shutter is used handheld some curious artefacts can appear in the image. One type I have seen at low shutter speeds is horizontal bands of blurring across the image (landscape orientation). Another, seen at higher shutter speeds is a horizontal band in which vertical subject elements are distorted in wavy fashion. I attribute both these artefacts to camera/lens movement occuring while the E-Shutter is scanning a strip of pixels. It takes 0.1 seconds to scan the whole frame which is a long time in relationship to the shutter speeds required for hand holding a long lens.
How to get sharp pictures at the long end
Settings: Mechanical shutter, speed 1/1250 or faster, let the ISO come up to allow f8. OIS on (IBIS if Olympus).
Technique: Auto focus very carefully on a clearly defined part of the subject, make sure there is nothing in front of, behind or near the desired focus point to cause misfocus. I find that AF is more reliable than MF on the GH3, but even on a camera with peaking, hand held MF will be difficult as the enlarged image is jumping about so much. Do your relaxation exercises before shooting. Consciously relax the arms and hands. Adopt a comfortable posture. Breath in then out and at the point of full exhalation gently squeeze the shutter button.
Tripod mounted Key message: Keep the camera/lens absolutely still during the exposure.
Equipment: Use a sturdy tripod. My experiments with a lightweight tripod showed it is not stable enough. Wind will cause havoc even with a solid tripod. Use the aftermarket Collar/mount for portrait orientation. Trigger the shutter with a cable release or smart phone. If you use timer delay I suggest 10 seconds. The preview image appears stable on the monitor after 2 seconds but my results with 2 second timer have been inconsistent so I think there is still some residual vibration happening there which is not problematic at shorter focal lengths.
Settings: E-Shutter on, unless the exposure time is longer than 1 second in which case use shutter delay of at least 2 seconds (Panasonic) or antishock (Olympus). OIS/IBIS off.
Technique: For distant subjects choose a clear cool calm day or time of day with low air pollution.
Summary My original, very likely inadequate, evaluation of the Panasonic 100-300mm lens was that it was quite good at the short end and a bit soft at the long end.
My re evaluation reveals a much better lens which is really excellent at the short end and very good at the long end. The second copy may have been better than the first but I believe the main reason for the different result second time round is that I used more careful technique. The lens represents outstanding value for money and is one of the reasons for the M43 system's appeal.
My wish list for the Mk2 version I think the basic concept, focal length and aperture range are just fine. Some contributors to user forums have wished for a wider aperture and/or longer focal length. But those things would make the lens larger, heavier, more expensive and therefore less appealing to most users. So my wish list is:
* Same focal length and aperture range. Of course if the designers came up with a way to make the lens more compact that would be a bonus.
* A built in rotating tripod collar standard with every lens.
* Faster burst rate in continuous shooting. I think this will become a more pressing issue when users start mounting the 100-300mm on their GH4's. There are already reports that the advertised frame rate is not achievable with this lens.
* Some improvement in resolution/contrast at the long end would be welcome.
* Some improvement in zoom smoothness would be nice also.
* As a user I would prefer a lens with internal zooming like the 35-100mm f2.8. I would opt for this even if it meant the lens were slightly longer in the camera bag. A lens with internal zoom is easier to work with both on the tripod and hand held and is not subject to mechanical decentering by external force.
Tidak ada komentar:
Posting Komentar