Sigma 100-400/5-6.3 OS C notes
2nd December 2019
The lens continues to deliver its optical and design qualities but the zoom creep to which I didn’t pay attention very much is more serious problem. You often need to have the lens ready at a particular focal length for some time and have to move or react in the field/terrain. The creep then gets in the way of getting the shot. The lock is useful for such lens but it only locks at one FL position (100 mm). This seems a budget-related outcome. The drawback impacts the usability of the lens directly and is a real weakness of this very good and original lens.
10th September 2019
Sigma has updated the OS operation in Standard mode via its firmware 1.03 (April). This now gives the best stabilising action for the viewfinder. So we reverted from the Dynamic to the Standard OS mode.
Sigma is also planning its first release made specifically for L-mount this year. Their activity in mirrorless designs may eventually lead to lenses compatible with other ML mounts. If they produce a 100-400 mm lens for the system it may well be of the standard speed f4.5-5.6 rather than the aperture ending f6.3 that we see in DSLR versions.
Nikon Z 6 observations (AF, exposure)
2nd December 2019
The camera is packed with clever ideas and features that allow the photographer better, more direct control over the capture. To highlight them would make a list of points. Many of them stem from the ML concept. The camera is a delight to shoot and handle. It is the best handling camera I held in my hands. Its sensor has also the best colour output integrity in processing (incl. the soft proofing) and at higher ISOs among Nikon’s 24 Mp sensors. There are problems, too. The first one contrasts (sic) with the DSLR AF capabilities and the second is about metering.
As for the autofocus
The autofocus in low light (low contrast) scenarios requires improvements. AF against high contrast background such as brightly backlit leaves needs improvement as well. These are related things and overcoming them must be difficult given the nature of CDAF component. It is an inherent weakness of the ML system. The electronics will have to do more than imitate human vision. The engineers will have to find a way of getting results the DSLRs are capable in these conditions.
Perhaps even more immediate problem is the size of the AF point in AF-C. The continues AF needs a reliable AF single point for when the accuracy is critical. This is very often with telephoto lenses. The focus point needs to be smaller, something between the pinpoint and the current size of the single point. If the pinpoint was fast focusing and working in the AF-C as well, even better (it should be), but the single point positively needs to be of a smaller size as the accurate focus is often impossible working with the current size. The somewhat crude (unintelligent) CD AF behaviour mentioned above and the size of the single point lead to a combination that is very problematic in practice when using AF-C and telephoto lenses.
Lack of 3D tracking that gets mentioned sometimes doesn’t bother me in the least as I seldom used this specific mode. It can be vital for some applications, that’s not difficult to see.
The highlight-weighted metering is a step in the right direction but in the current form, not a full implementation of the idea in question. It needs more work to become a reliable help with exposure. It should also be tunable for various degree of controlling the highlights (exposure). Unlike the AF difficulties, it shouldn’t be hard to get this right. It is long overdue, in my view, and this field engineers are certainly developing this approach to exposure further already.
21st August 2019
Nikkor 200-500/5.6 VR on ML:
Z 6 vs D750 AF-C - summary of my results so far:
The Z 6 with 200-500/5.6 VR (and FTZ adapter) is faster and more reliable for continues autofocus using the AF-ON button in all scenarios than the D750. It also has a much wider AF coverage. With TC 14 III the results are even more in favour of the Z 6 when autofocus of these two cameras is compared on erratically moving objects and switching between the subjects and between closer and more distant subjects in follow-the-subject scenarios in cluttered environments. (Use of Low Light AF mode may be needed for getting the results under some lighting scenarios.)
- - -
The new memory card is readily available/offered by the industry for some time. The ball is in the court of the devices manufacturers that write the relevant firmware to run the cards in their devices. There hasn’t been much urgency about bringing the cards to the end-user. Cutting the limitations in storage media of this type was also lacking a sense of urgency (on the current market). The card should offer a better value per data storage and performance possibly (where needed) than the current XQD cards.
FE 200-600 G OSS
Sony has created 200–600 mm zoom – the FE 200-600/5.6-6.3 G OSS for its Alpha system. It is a member of a rare kind of telephoto lenses that original manufacturers didn’t enjoy designing. Or maybe they would, but the profit-based model pushes it into the background every time a weak moment gets the better of some natural designer's instinct so that an extended gain from the artificial attractivity (or lack thereof when making optically subpar expensive zooms) of other sets of products is possible.
The third-party manufacturers model is different in that, depending on the technology available to them, they look for these artificial and other spaces to fill them in with what is useful to the user first and then profit from that sort of priority. Sigma, in particular, was successful with their 150-500 and 50-500/x-6.3 telephoto zooms, not to mention the next wave of Tamron and Sigma 60/160-600/x-6.3 lenses that combined optical qualities with other features such as professional build or compactness. That changed the settled order of the OEM model because it lost its function and so the £ 1150 Nikkor 200-500/5.6 VR was born. And the switch to a carefully designed beast that wards off this competition but still encourages to look for a higher markup (more expensive) options along with it. This was only a temporary step before the need to embrace this unpopular specification popular for practical photography with both the professionals and the enthusiasts, or consumers, if you like, will be exposed further.
The FE 200-600/5.6-6.3 G OSS is such next step. It also is a mirrorless design. And it does not come as a surprise that it came from OEM - Sony. The pricing and the status of their GM lenses (the 100-400 GM contrasting with others and especially with the fractionally slower Tamron and Sigma 100-400 lenses) asked for such a move because they no longer are the only ML 135 format manufacturer. It would be reckless not to use and emphasise its position in the ML at this point.
The lens combines all the familiar traits of the telephoto of the rare kind in a good package. It is a zoom lens of a specification that is just right, in a good coat, with good internals and function for a reasonable price.
So a quick look at the points that make this lens worth attention:
It has the reach of the ideal 600 mm while being physically about 5 cm longer, 3.5 mm bulkier and 25 g heavier than 200-500/5.6 VR (the lenses without the foot). If you factor in that the lens does not extend the barrel when zooming (or focusing) the dimensions are practically the same as with the 200-500/5.6 VR.
The reach is 600 mm/f6.3, though.
It maintains constant length and some work is done on sealing
– this is practical as it helps to limit potentially serious dust and moisture problems in environments the lens is likely to face.
The price is right - £ 1800.
The lens is optically stable – wide open long end sharpness is maintained.
The background bokeh is good. The edges seem marginally more defined with less intensive blur.
But there are no problems - including the transition from OOF to focus.
The AF with and without TCs is very good. The TC 1.4 combination is also optically very good.
The Sony 1.4 TC has moderate barrel distortion which cancels out pincushion distortion of the telephoto lenses (the 200-600 G OSS has slightly higher-moderate pincushion distortion). This is the first native super-telephoto zoom for ML.
The lens also works with Nikon Z via Techart Sony E to Nikon Z AF 2 mm adapter ring.
Let’s remember that the optimum depth of field in the telephoto portrait photography is in the range of f5.6 – f10.
This is depending on the vision and exact intent of the photographer for the given scene/portrait.
Larger animals/subjects can be shot from longer distances compared to the small ones hence more DOF in the shot, but their size requires more of it for a comparable scenario. So that is not a typical case for fast apertures.
The larger apertures shine when the subject is set deep in the environment. This is a marginal use with the telephotos – though the effect of strong isolation if desired, which also is a question, is there. I have found that if the bokeh is of good quality, the need is much reduced and less desirable (the blur as a substitute isn’t needed and it can even stand in the way of more subtle renditions that a good bokeh can render).
The larger apertures can be similarly useful when this scenario (the subject is distant) is used for cropping purposes.
The dimensional rendering of in and out of focus shapes, abstracts, colours and lights of the picture can be retained with the shallower DOF.
But getting a large aperture for cropping (prior or during the processing) is today needlessly taxing (bulk) and expensive way of getting the isolation for most cases.
Nikkor 24-70/2.8 S
30th June 2019
On a technical side, I ran several tests with 24-70/2.8 S for noticeable faults and then regarding sharpness and quick bokeh impressions. I am thrilled with the lens. Here I post my brief summary of 30 m distance tests on Z 6 (all lenses). It's shot on a bright calm day with the shutter speeds ranging from 1/2000s to 1/400s, remote release (FCE shutter), fixed WB, focus in the centre and routine measures such as backing the samples (apertures compared here: f2.8, f4 and f7.1).
The target and the central area at 35 mm
30% centre 24 mm (to 24/1.8G): all f stops, no difference in favour of either lens – taking turns depending on where you look, different field curvature profile.
Towards the borders and edges (away from the red rectangle): zoom clearly better at f2.8, f4 zoom better, f7.1 zoom still better a bit.
30% centre 35 mm (to Milvus 35/1.4): at f2.8 Milvus better at the periphery, the centre the same, the Milvus centre at f4, f7.1 better with its unusual monstrous bite stopped down (f7.1) and better at periphery further from the very centre.
Towards the borders and edges: apart from the upper-left corner Milvus better at f2.8, zoom puts on even performance up to the extremes; very similar comparison at f4, the zoom fares very well at extreme peripheries. At f7.1, the prime is better; the stopped-down behaviour makes the difference, incl. in the upper-left corner. My sample of 24-70/2.8 S has some irregularity to the right off the centre around Dx border area (drop in overall sharpness). This is a tough comparison against the Milvus, the result still very good to excellent. (Consistent with the output of slightly warmer glass the Milvus is warmer, not by as much as the 24/1.4G when compared to the 24-70.)
30% centre 50 mm (to 50/1.4 Art): f2.8 the same, f4 zoom better everywhere (a Milvus style definition shows at f5.6 as well as it spills towards the wider and narrower apertures), at f7.1 the lenses compare about the same.
Towards the borders and edges: at f2.8 a wash, the Sigma field curvature (sharp focus) is forward bending (towards the camera) whereas the zoom has got a flatter field. At f4 the zoom has still a flatter field and a little bit better sharpness overall; at f7.1 Sigma a tad better all over, it’s producing an extra bite peak (glowing) whereas the zoom is losing strength a touch. Sigma is noticeably warmer (more than the 24/1.8G).
70 mm characteristic in context of the above: at f2.8 the whole frame slightly less sharp, it is slight, the sharpness is evenly distributed. An interesting thing about the 70 mm: the whole frame snaps into sharp by minuscule move to f3.2, and a tad of extra bite yet is gained at f3.5. It is good to know for when shooting with various intentions and subjects. One can opt for full power isolation or for the extra sharpness with the lens still open a lot depending on the intent.
A bit of field curvature forming behaviour here – spot focus helps the extreme-left side as the field stretches, not really on the right extreme with my sample (little bit in the corners). When the spot focus extremes of the opposite sides are compared, they both show differences (are softer) compared to the spot focus for each relevant edge (as one would expect). The midfield focus patch forms in a stable fashion, ignoring largely (for practical reasons) spot focus as compared to the centre one.
A highlight of the lens resolution and contrast test at moderate distances is that it is a remarkably well made 24 to 70 mm zoom. It’s not compromised in the standard way as other zooms tend to be – on the ends of the focal length range. It actually builds on the E version when it comes to 24 mm making this FL (along with the stunning 50 mm) a highlight and including 70 mm to gain a very good if not excellent result at the long end. It’s a change to our common experience, a one that makes one think differently when using the lens. There obviously are more aspects to a lens, for example, I found out the lens isn’t compromised when it comes to close distances either, or that it maintains a usual degree (higher) geometric distortion (note that the 24/1.4G and 24-70/2.8 S were compared to each other and to other primes with distortion correction on). As always is the case with me, I am likely to add a note or two about the lens later after a longer use in field.
Nikkor 200-500/5.6 VR
2nd January 2019
Telephotos form images by bringing subject close to the camera. This changes certain aspects of the process and also the perspective of the image in a way that is unique to these lenses. Surrounding elements appear a little closer to the subject (their environment) as these also get magnified. We call it compression. Extreme settings can result in dramatic effects with objects of which apparent size in embedded in our minds such as the Moon or the Sun.
This focal length flexible telephoto zoom that goes from 200 to 500 mm (or from 280 to 700 mm with a TC 1.4) is an old-school F-mount lens. Its main feature is a maximum reach (magnification), but depending on the applications it’s the flexibility that can sometimes, or often, be essential. How often changes from case to case or even moment to moment but when there is the need the zoom function is critical for getting the shot.
This zoom lacks refinements of some higher-class lenses such as solid seals and fine mechanics, but it is a very nicely built lens that extends moderately with zooming. It operates well, though the zooming and focusing is somewhat Spartan experience. Its AF isn’t very fast on the current cameras, F or Z mount, but again, the VR works excellently hand-held and also on the tripod with the head relatively loose or fixed. For its power, it is priced very reasonably which is underlining its class within the line-up of the Nikon telephotos but also benefits that come with this package.
Optically it is a similar story. As a modern DSLR lens, it is virtually flawless for what it needs to do in the field, but it doesn’t give its owner any extras. This translates into two important properties for such a lens: very good to excellent sharpness that is consistent throughout its focal length range and the same for the quality of bokeh. The sharpest aperture (peak sharpness) is f6.3 which is only marginally sharper than f5.6. Where the lens is lacking is the border and corner resolution. But because the central region of excellent resolution is large, it is of a lesser to marginal importance if what you shoot are separate subjects - the subject shooting, as I call it. Better coverage for resolution across the image field can be found with some other new lenses, though that may come with specific limitations elsewhere. As for the border and corner resolution, it improves steadily as the zoom approaches its long end, it is worse at the short end where the drop in resolution and contrast is noticeable, improved at 300 and 350 mm and improved further with the best consistency achieved at the long end. That makes sense today although a long row of various telephotos of the previous generations will remind that pulling it off – optimising for the long end – wasn’t always straightforward. The resolution can be improved by stopping down to f8 at the short end, e.g. for a landscape shoot, one can need it. At longer focal lengths one can notice responding to stopping down (by any small increment) is better and faster with regards to the periphery. Covering this sort of inconsistency is a task reserved for finer and more expensive tools where considerably higher efforts need to be employed to manage it.
Unsurprisingly, the lens isn’t excellent at minimum focus distance as well. Something we can’t expect from a non-macro lens of this class. Neither can we expect it from every design in general. Though macro sharpness it is not, the sharpness is sufficient and puts no barrier in the way for achieving beautiful photography in the close-up realm with this lens. The lens maximum sharpness is achieved short distance out of MFD.
Besides the consistently high sharpness in the central image area, excellent bokeh is another defining aspect of this lens output. The out of focus areas play a critical role in the rendition of the telephotos to more and more people. Combination of these traits is a building block of this lens look.
Having seen colour results for some time, there isn’t anything that hits us as a deviation from an average Nikkor glass to warrant study of the colour in detail. White-balanced settings through the lens give in this case satisfactory colour that I have come to expect from a Nikkor lens without noticeable particularities that would go beyond the characteristic feel of the Nikon glass. Similar can be said about other potential optical flaws. The lens works exceptionally well with TC 1.4 III optically (we haven’t tried TC 1.4 II, but reports sound positive) on 24 MP if we ignore a slight increase in pincushion distortion above otherwise low-level distortion of the lens. I observed coloured moiré patterns in the bird’s feathers at 10 mm MTF chart position.
The focusing takes some hit there but the resulting combination yields 700 mm and sharp f8. This focal length allows for effective reach and improves the rate of discrete shots of the wildlife. Combined with the 1.4 TC the lens has no peer within Nikon line-up and beyond. The 200-500/5.6 VR is close to perfect for photography on the move where one needs to react, follow the subject and operate lens quickly (as opposed to shooting prepared, e.g. in a hide). There were rumours that Tamron was involved in the design of the 200-500/5.6 VR, we may never know. Considering the role Nikon had to play in bringing the lens to the finale it would be more curiosity than meaningful information.
Lens set to 600 mm:
A slate-coloured hawk watching with interest anhinga struggling to swallow a fish she caught
(the hawk attacked and tried to steal the fish).
The 390 mm for the same (proportions of) framing:
The attention shifts to us as we drift in the canoe closer unintentionally; the bird puts more distance between us shortly.
This we consider a good portrait of the hawk, but the outcome is exception rather than rule which says that getting too close to the wildlife reduces the chance for capturing natural poses and expressions.
Note that neither shot would be possible without the ability to zoom.
The lens has its weak points. Its versatility is lessened by weak border resolution and focusing with the lens is a bit like its cumbersome zooming. However, the choice of strong points and their implementation make it a remarkable tool. And it is just as useful in a wide range of scenarios and conditions.
It will deliver highly for the picture quality in particular but will remind on occasion that you may want to pay for reliability with relativities such as cash, risk, zooming function, bulk and weight and it may well pay off - or not. As with any lens of technically lower specifications made to high standards and executed in superior design, the compact 200-500/5.6 VR (plus the TC 1.4 III option) has its cons and pros arranged in very strict relations. It means that if one needs the package the lens is offering, they have minimal options to get close to it elsewhere. It’s one of the highlights in F-mount line-up.
Crop to MFT; Crimson-crested woodpecker in a tree bending over the river
working on the nesting hole (1/640, drifting downstream):
Using longer focal lengths, one can take advantage of photographing subjects that are high above the ground at flatter (narrower) angle,
so the subject appears more level with the lens. The reach is effectively lifting the shooter off the ground:
Sigma 100-400/5-6.3 OS C
2nd January 2019
Nikon has many F-mount telephotos in its line-up of professional, prosumer and enthusiast orientation, including the new releases and high-performance compact PF lenses. We can expect telephoto lenses to appear in Z-mount as well and similar to the situation with the 200-500/5.6 VR in F-mount, we can expect a 100-400 lens of a variable aperture that would reflect the appearance of new lenses from Sigma and Tamron. They manufactured 400 mm zooms that are easier to handhold and operate at this focal length as they are noticeably lighter, smaller and portable. Z-mount lenses tend to benefit from its larger mount, and one of the ways where this can be reflected is the size of the lens. Use of versatile zooms capable of and often used at 400 mm for close-up shot places demands on following of erratically and rapidly moving subjects, precise focusing and timing. A compact 400 mm lens helps with this and is allowing to handhold the lens for longer, have it ready for a successful shot for longer and the lens can be taken to locations where you would previously consider only shorter lenses. These advantages far outweigh fractionally brighter lens in the field, especially for close-up shots where the depth of field is very shallow, which is the typical and main application for this lens.
This is the first thing Sigma did well with this lens; taking full advantage of a fractionally slower lens that could be then transformed on essential fronts for the user.
The second good thing is that the lens is strictly optimised for wide open across the image filed. There are settings, including the borders, where wide open the lens resolves more than at f8 on spot focus. 400 mm zoom lens must be optimised for wide open sharpness, and this is what Sigma did. As for the curvature of field, the centre focus field curvature is always covered by Nikon D600 focus points spread (a modest spread) amounting to about 33% of flat focus plane at 100 and 200 mm and 29% at 400 mm with a slow, gradual increase of the curving towards the edges.
The wide open optimisation is continued by bokeh, the third thing done right with this lens. The bokeh drawing is less fat, it doesn’t blur as large and intensively, but that is a standard adopted more widely today, often resulting in superior transitions and quality of the blur as is the case with the 100-400 OS C. The lens is optimised for background bokeh. This results in very calm backgrounds where other lenses render more structured blur and double lines. Foreground bokeh, on the other hand, can be edgy and jittery in some settings with some elements, adding a “photographic spark” to the overall painterly rendition. The outcome is mostly neutral, but where applicable, I find the combination pleasant and interesting for about 80% cases whereas for the rest I would consider calmer representation for the foreground elements.
Elements of the foreground combined with the intensity of the blur lead to a smooth drawing of out of focus areas:
Added spark in the foreground that is positive for the image:
The lens has a slight red tinge we know from other Sigma lenses (e.g. 50-150/2.8 OS II). The colour set through the lens (white balance) is void of the shade, and the colour overall is exquisite, also in direct comparisons with other lenses. This isn’t a given as each lens’ glass colour profile is interacting with WB-setting methods uniquely. Despite the red tone, the glass is on the cooler side, perhaps surprisingly. In other words, the red tinge isn’t followed by the underlying warm (yellow) tone some Sigma lenses have (24-35/2, 50/1.4 Arts). Though we are not exactly keen on more noticeable colour deviations, we are unaffected by them in practice. For similar concerns people can consider, among other things like the white balance, using the lens calibration application that X-Rite rolled out recently.
As for the operation, I shall be brief. The only problem I found is focusing on the subject that is approaching towards the camera. The number of keepers is noticeably lower as the camera struggles to focus the lens in this particular scenario effectively. It seems to be the lens problem as this has been reported widely including the Canon DSLRs. The focus is otherwise very good in a wide range of situations we encounter in the field. Overcomplicated settings and description of OS modes should be mentioned as we found out that shooting on Dynamic Mode works best because this mode is stabilising the viewfinder well, which is very important with this lens, while very effective for the shot. The mode has no adverse effect on anything one can shoot (including while zooming).
This is a well built lens and to my mind, it deserves seals with which Tamron seems to have equipped their later coming version of very much the same lens. Still, the lens is inexpensive (as is the Tamron lens), so one can view its exchange as a valid option after a time, depending on the usage of the lens etc. In my experience though, photographers don’t like to part with lenses they know well for reasons of maintenance, so they are likely to use the Sigma’s service in many cases and hope it will come back in a similar form they were familiar with. With the help of the seals, the lens is preserved better for longer. So I would venture a guess that it makes sense for most that the high-quality slow lenses have adequate seals in their construction to underline the quality, even at the cost of some price increase. As I said the build is very good and when I first handled the lens this marked the moment I really wished the lens performed. That is I was impressed. My attitude was quite reserved at first, and I wasn’t sure about purchasing the lens unless it was able to replace reasonably 200-500/5.6 VR and 70-200/4 VR lenses. That means both the landscape shooting and the subject shooting close to standards of these Nikkors. That is why I had doubts. As I mentioned in several conversations, we had three samples of the lens. The first sample was very slightly decentred which I brought to the attention of the seller. It was mild and more borderline decentred than a clear example of decentring, but my landscape requirements needed better. The replacement arrived with a squashed corner of the box. I still tested the lens but asked for delivery with box intact. I didn’t discern a fault with the lens during a brief test (impact related or otherwise) but the third lens was slightly better at its long end, and it was the lens I kept. It was only later, after some use, when we started fully appreciate this lens for what it is and the capability of the package it offers and delivers and I have to say, we would now be keener on purchasing it today. Sigma has moved things on the front that helps photographers, and that is influential in photography again. To some extent, they have to do that if as a third-party manufacturer they want to be at the forefront of events in the field. We can thank them for their bold, often nonconformist efforts and not least for the creation of clever, innovative tools that move boundaries in what is possible.