To Z7, or not to Z7, or Sony?

After monitoring a Nikon flash sequence on a storage oscilloscope, I was amazed and appalled by how long CLS/i-TTL spends in pumping out 'morse code' before the flash that actually illuminates the subject goes off. I thought the 'scope trace would never stop scrolling before it got to the main event.

Hmm. I find that if using multiple TTL groups, the preflashing sequence does take a while and it does make it harder to time shots. However, if the on-camera flash is TTL and the remotes are manual, then the preflashing takes less time (and this is usually the way to more predictable results than having multiple TTL groups).

Each flash group that is set to TTL mode emits a series of test flashes of different flash energies to figure out each group's influence on the illumination of the subject or scene. But if some of the groups are manual, then they don't need to be tested at multiple energies. If you have only one TTL flash group, the exposure happens pretty quickly.

Of course, all that can be avoided by using old-fashioned Auto-Aperture or manual mode.

Of course.

Don't even get me started on the interminable delay if redeye reduction is employed!

It's meant to handle a situation where there is an on-camera direct flash on the subject with a configuration where the distance to subject is quite long relative to the distance from the flash head to the optical axis, but I never use flash in that way, so I never have redeye reduction on.

But if I have to buy all new lenses, then the Z7 has no advantage for me whatsoever. And Sony has a far bigger selection for me to choose from, at a range of price levels.

Well, you don't "have" to buy all new lenses, you can use some of your old lenses via adapters, and get some Z lenses to cover situations which require the best autofocus performance.

That "so far" is a significant clause in your second sentence!

Well, I doubt the lens quality will go down from what it currently is. Nikon have been consistently coming up with optical innovations and new technology that improves image quality and this is happening on the DSLR side as well. There have been a few flops or lenses that received mixed response from reviewers (e.g. 105 VR, 80-40 AF-S and 58/1.4 have mixed reviews), but Nikon seem intent on producing higher quality lenses for the Z system, and so far they have been impressive.

bgelfand just recently shared an article on dpreview with regards to major Camera manufactures, and how their openness of the lens mount, contributes to lens availability

Sony having essentially open lens mount spec, helped tremendously with quality 3rd party lenses.

Quality 3rd party lens availability, in turn, it seems, brings in more hobbyist/advance amateur crowd.

That, in turn, creates, at least, a perception of long-term system-viability, so people are willing to experiment and invest more

(as more than one company builds products for the given mount).

So overall, it seems that Sony and L-mount full frame mirroless are better investment for a more people.

This is my 'pseudo-scientific' explanation of why folks migrating to Sony.

We have D800 and Samsung NX. I felt that Samsung Nx was a bad investment, even though a technically good camera.

If I am given a choice, i will not invest into a non-open mount system.

I was very interested in Nikon Z7/Z6 but will now hold off, because I am not comfortable investing into a close-mount system.

I do not think technical specs (that are relevant to my usage patterns) between Sony and Z-sytems are that much different to justify supporting no-competition-allowed lens mount.

I think a lot of people are going for the Sony now. A good number from Nikon and Canon users too. With the new EOS-R5 perhaps Canon can stop that but many Nikon users are going Sony when they consider mirrorless. I personally would pick the Z7 instead of the A7RIV. All the Z lenses so far are good lenses and price aren't that expensive. (of course not the 58mm f/0.95 and I wouldn't want it either if the price is $1500). I really dislike the fact that Nikon made it f./0.95 instead of f/1.0.

Sony have a 4-year head start in the development of full frame mirrorless and lenses for it (nine years for APS-C mirrorless), I believe there are now 60 Sony E-mount lenses (of which some are APS-C). Nikon have 10 (12 should be out soon). Clearly Nikon need to catch up to a more complete lens lineup and then we can start looking at the relative positions of the brands in the market. It would be premature to judge Nikon's Z mount's success on early adopters and the first-generation cameras.

Sony have a technological lead in autofocus for full-frame mirrorless cameras over its competitors, and they have a relatively complete lens lineup. These are most likely the reasons customers are driven to their product line. Nikon have their own advantages including ergonomics, viewfinder quality (IMO), weather sealing (see Roger Cicala's and others' reports on that), and arguably they have a slight advantage in the mount geometry that is probably the reason they have less corner sharpness drop in the lenses. Sony's mount isn't technically open (where is a link to the specifications? Not publically available) but they allow some third-party access. Third parties will most likely reverse engineer the Z and RF mounts over time once the mounts obtain some reasonable market share. Personally I specifically want to use Nikon lenses and a consistent rendering and color across focal lengths and I wouldn't be buying other lenses unless for some special purposes (you can still get an adapter and mount your favorite macro rig).

Mainly it is Zeiss who have taken advantage of their long relationship with Sony and supplied additional lenses for the E mount system. I have used many F mount Zeiss lenses as there was a time where I felt Nikon was moving too slowly to introduce modern full-frame wide angle primes for the F mount and I got into Zeiss lenses, which, however, I never really got to like. One of my main complaint is that they tend to block shadows into complete blackness with no detail very easily. Here is an example with the 21mm Milvus:

Untitled

I don't have the matching photo with a Nikkor in this case as I was in a hurry but my experience is that there would be detail in the shadows if I had taken this shot with a Nikkor. I'll use another person's example to illustrate what I think is beautiful in the handling of backlit situations with Nikon lenses:

Nikon Z 14-30mm f/4 S Review - Optical Features (Page 3 of 7)

search for the photo with caption "No flare! Taken at 14mm, f/16." It shows detail in the shadow side of the rocks and in my opinion in a beautiful manner. This is the kind of behavior I want from my lenses. (The photographylife article is actually quite critical of the 14-30 Z but I believe the issue with sample variation in the corners is mainly in the area which is not included in the image after distortion correction in software, though this is speculative on my part as I have not evaluated this lens.) The earlier 14-24/2.8 Nikkor for F mount would not handle shots with bright point light sources in the frame as well and would show prolific ghosting, so the new 14-30/4 appears to be an improvement for this type of a shot. It is also unusually compact for a zoom that goes to 14mm in full frame and takes regular front filters.

I think 5-10 years from now the Z and RF mounts will have similar lens availability as E mount. There might not be as many third-party AF lenses, though. But for me the Nikkors are the main thing I buy the camera system to use. I know there are a lot of people who want to mix lenses from various makers, but that's not for me (I am saying this from some experience). If I buy another brand's camera system, that's a different thing and then I would do whole projects with that system (again with the same manufacturer's lenses) to get a consistent look to the images. I like Canon's 50/1.2 RF from the images I have seen from it very much, but I would prefer consistent apertures across focal lengths and they offer f/1.2 only at 50mm and 85mm so far. And it would be way too expensive for me to get a full line of f/1.2's. Not to mention it would completely throw away any hope of compacness and a lightweight bag. For Nikon I really like that they started out by making a line of f/1.8 primes that satisfies my desire for consistent apertures across focal lengths typically used for event photography and portraiture. The f/1.8's are certainly going to be less expensive to get than f/1.2's or f/0.95's or f/0.7's that they might make to satisfy exotic needs and deep pockets.

I have used many F mount Zeiss lenses as there was a time where I felt Nikon was moving too slowly to introduce modern full-frame wide angle primes for the F mount and I got into Zeiss lenses, which, however, I never really got to like. One of my main complaint is that they tend to block shadows into complete blackness with no detail very easily.

I tend to understand this as some lenses having more internal reflections, which reduces overall contrast and lifts the blacks past the toe of a film response curve. I'm not sure how much that's still a desirable feature in digital, although I know some raw files still have a non-zero black. Is this what you mean, or is there another feature of the lens I'm failing to understand?

I tend to understand this as some lenses having more internal reflections, which reduces overall contrast and lifts the blacks past the toe of a film response curve. I'm not sure how much that's still a desirable feature in digital, although I know some raw files still have a non-zero black. Is this what you mean, or is there another feature of the lens I'm failing to understand?

My point is when the shadows are rendered so dark, there is no possibility to recover the detail that is there in the subject, and any attempt to do so will produce extremely noisy results. I've experienced this in practice when I've attempted to use Zeiss lenses for indoor available light photography, where dynamic range is already reduced at high ISO. Rather quickly actually I learned that I could get better results with Nikkors that aren't losing those details into noise. The darker the shadows are the lower the luminosity-to-noise ratio is and if you bring the shadow tones into light, the noise is amplified.

Of course, if one desires ultra-high contrast results then by all means. I like the more gentle rendering of shadows by most nano-coated Nikkors.

if one desires ultra-high contrast results

or, if it's not moving, a suitable HDR bracket....;)

Dimly lit church interiors with sunbeams is a case in point. Blending them can be tricky without loosing something of the 'depth' resulting in a kinda plastic looking image.

My point is when the shadows are rendered so dark, there is no possibility to recover the detail that is there in the subject, and any attempt to do so will produce extremely noisy results. I've experienced this in practice when I've attempted to use Zeiss lenses for indoor available light photography, where dynamic range is already reduced at high ISO. Rather quickly actually I learned that I could get better results with Nikkors that aren't losing those details into noise. The darker the shadows are the lower the luminosity-to-noise ratio is and if you bring the shadow tones into light, the noise is amplified.

Interesting - that Zeiss is probably overrated? o_O. (Not familiar with Zeiss other than their binoculars and an ancient rangefinder.)

Yes, I think different folks have different perspective.

I think people who tend to be closer to perfection while maintain practicality -- will tend to appreciate more expensive, high quality, yet closed systems.

Folks like me would tend to 'wonder around, experiment' while in the process, increasing difficulty of usage, and, probably, reducing quality of output.

Like the picture you posted with Zeiss... I liked it, and if I would take it, I would think it is great. :-).

But you are right, shadows/blackness is not gradual, and, instead unnaturally digital.

My hope is Nikon opens up their lens mount, and make it truly open.

I do not think they will loose business if they do that.

Instead, I think they will create a multi-tiered ecosystem of not just lenses, but other auxiliary devices, that may even be transformational to the market space...

...

I think 5-10 years from now the Z and RF mounts will have similar lens availability as E mount. There might not be as many third-party AF lenses, though. But for me the Nikkors are the main thing I buy the camera system to use. I know there are a lot of people who want to mix lenses from various makers, but that's not for me (I am saying this from some experience)….

It is more than clear that neither Canon nor Nikon is interested in opening their mount specifications, because that is inviting competition and hurting their own lens sales. However, in the longer run, I don't think that makes a whole lot of difference.

Sony had to invite third-party lens manufacturers in their early days because Sony was effectively starting from scratch with the E mount. The SLR business Sony took over from Konica-Minolta hardly had any customer anyway. Sony needed help to provide sufficient lens selection in the early days to get the A series off the ground. Both Canon and Nikon have plenty of SLR lenses to be adapted for mirrorless. And they didn't want to start FX mirrorless too prematurely 5, 6 years ago when DSLRs were still selling pretty well and mirrorless still had many rough edges.

In any case, enough companies have already reverse engineered Nikon's Z mount and there are quite a few third-party adapters and some lenses, but so far there is no third-party AF lenses from the major third-party suppliers such as Sigma and Tamron. If there is sufficient Z body sales so that it is profitable for Siamg, Tamron, Tokina, etc., they will definitely make Z-mount lenses. But Nikon lenses will continue to have a it of advantage with their own mount.

My point is when the shadows are rendered so dark, there is no possibility to recover the detail that is there in the subject, and any attempt to do so will produce extremely noisy results. I've experienced this in practice when I've attempted to use Zeiss lenses for indoor available light photography, where dynamic range is already reduced at high ISO. Rather quickly actually I learned that I could get better results with Nikkors that aren't losing those details into noise. The darker the shadows are the lower the luminosity-to-noise ratio is and if you bring the shadow tones into light, the noise is amplified.

That depends how you're raising the shadows though, surely? (I ask because I'm confused, not because I'm trying to be argumentative.) Devices like the Tiffen "ultra contrast filter" work, I believe, by scattering the light from the brighter parts of the frame into the darker parts of the frame. Let's assume you've got linear 14-bit raw values from 0 to 16383. A perfectly captured shadow might have blacker areas at 0-ish and brighter areas at 32-ish. If you have less global contrast in the lens (either naturally or by adding a filter), light elsewhere in the frame might lift those values to the range 1024-to-1056. You'll be more able to see them, especially if your monitor is adjusted so that some of the black levels are a bit squashed, but they're not being captured any more faithfully. Actually less so - probably the lighter value will be slightly less offset, because some of its light is contributing to the exposure of the rest of the frame.

If you want the brighter bits of the shadow to hit 1024 and they were captured at a level of 32, you can do that either by scaling the numbers by a factor of 32, which stretches the shadows apart, or by adding an offset of 992 to everything, which makes the blacks less black. You can do both of those in post-processing with a basic slider, though; I'd rather have closer to the original values captured. If the contrast of the capture is reduced, you're getting less detail between the shadow levels, not more. I'm not aware of there being a way for a lens to adjust the "gamma" of an image such that the blacks stay black but the shadows are lightened relative to the highlights; am I missing something? (Edit: I mean universally, before someone realises I've forgotten ND grads.)

The exception (and reason I could be talking rubbish) might be, as I suggested, that I think the raw black value encoded by Nikon might not actually be 0 - the blackest blacks might be clamped. This is from a vague recollection of a discussion of how Nikon, Sony et al. cook their raw files. (I'd really just like the numbers to the best of the ability of the hardware, and then pass on some parameters to the decoder so the software has as much detail as possible to work with. But this doesn't seem to be the way.) It's absolutely the case for film - since a negative needs a certain amount of light before it starts reacting, there's something to be said for optically raising the blacks (or in extreme cases doing a brief "pre-exposure" to sensitise the film). For the most part, though, digital sensors ought to be fairly linear, so the farther apart you put the brightest white and darkest black in the sensor's exposure range, the more recorded values you'll have to play with everywhere in the frame. (E-TTR, Nikon.) Reducing the contrast range you capture ought to be reducing the number of values the sensor can record, and that should never be a good thing.

Of course, if one desires ultra-high contrast results then by all means. I like the more gentle rendering of shadows by most nano-coated Nikkors.

Oh I don't - I hate blocked blacks and highlights. Slightly to my shame as someone who wrote a specification covering the subject, I don't have an HDR monitor - but I absolutely don't like areas where detail has been lost. That's the difference between a negative and a print, though: just because I want to lighten a shadow region for consumption (and yes, since the eye works with ratios of brightness, spreading the same number of colour gradations over a higher base brightness will make them less visibly different) doesn't mean I needed it to be captured that way.

My point is when the shadows are rendered so dark, there is no possibility to recover the detail that is there in the subject, and any attempt to do so will produce extremely noisy results. I've experienced this in practice when I've attempted to use Zeiss lenses for indoor available light photography, where dynamic range is already reduced at high ISO. Rather quickly actually I learned that I could get better results with Nikkors that aren't losing those details into noise. The darker the shadows are the lower the luminosity-to-noise ratio is and if you bring the shadow tones into light, the noise is amplified.

Just wondering: when the [Zeiss] shadows are so "dark" that nothing can be recovered even from the raw file without extensive noise, how does the RGB show up on the histogram? Clipping? 0,0,0 or not so bad? Or, if it does not clip on the histogram but too dark anyway? How does Nikon compare on the histogram?

I have been very pleased with results from Zeiss lenses for the Sony. While contrasty, they capture shadow detail quite well. Their rendering is relatively consistent across all focal lengths in the series, and to a large extent, between series as well (e.g., Batis and Loxia). Milvus lenses for Nikon and Canon DSLRs are held in high regard optically. Since they are manual mavericks in a herd of AF lenses, and expensive, their acceptance has been somewhat subdued. With the Otus line, also manual, Zeiss spared no expense in making them as aberration and distortion free as possible, and pass that expense on to the customer.

Zeiss is very lens-suite oriented for both still and cinema lenses. Sony and Nikon have responded to the popularity of prime lenses with their own brand emblazened. My impression is that they tend to be one-of-kind designs, each one by a separate engineering team, with little though to consistency. They are also large and heavy, while my preference runs in the opposite direction.

I suspect the Nikon mount, particularly the electronics, will remain proprietary, as they have since. the beginning. Foreign lenses will remain manual, except for a few brave hackers. If Sony had done that in 2013, there probably would not be a separate Sony/Minolta forum in PNet today.

Sony A7Riii + Zeiss Loxia 50/2 Planar, bracketed -2,0 and +2 stops.

Middle (0) exposure, unedited - still plenty of shadow detail if you want.

Just wondering: when the [Zeiss] shadows are so "dark" that nothing can be recovered even from the raw file without extensive noise, how does the RGB show up on the histogram? Clipping? 0,0,0 or not so bad? Or, if it does not clip on the histogram but too dark anyway? How does Nikon compare on the histogram?

Maybe not an easy thing to analyze - not a fair question. Anyhow, I try to "over-expose" these days while making sure the white is not blown out. The raw file retains better info this way.