The Switch

Advance warning: this post contains nothing but gear p***, since it is an equipment-centric post after all…and I make no apologies whatsoever for that. And that thing isn’t a carrying handle, it’s the HTS.

I suppose many of you will have seen this coming: I have switched primary camera systems from Nikon to Hasselblad. And you’ll also probably know I’ve got extremely good reasons for doing so, even more so since this is a significant commitment by any standards. I’ve also had several conversations with various people in the organisation that have convinced me that their future is also going to be pretty exciting indeed. Judging by the number of people who commented and emailed me after the previous recent posts on medium format (here, here) – there’s quite a lot of curiosity and more medium format shooters here than I previously thought. Today’s post is an explanation of that rationale, and the comments will be an attempt to answer any questions from the curious.

The generation gap

For the 13 or so years I’ve been shooting seriously, I’ve always maintained one system as the go-to when you Need To Get Something Done – ostensibly, the Nikons. First a D70, then a D2H, D200 (x2), D3, D700 (x2), D800E (x2) and now D810 and various secondary bodies* – there’s always been at least one general purpose Nikon in the fleet. Up to five at one point, if I recall correctly. This has been the case, because even though specific tools may address one very narrow range of tasks better – there are frequently situations in which you simply do not know to such a degree of specificity what you might be expected to shoot. In such situations, a Swiss Army Knife is the only way to go.

*D50, D80, D90, D3100, D5100, D5500, F6, F2T. Probably more, but those are the ones I remember.

When you further take into account the necessity of being intuitively familiar with the way your camera responds to avoid missing images in critical situations, reliability, support, easy replaceability, accessories, special purpose lenses, flash, etc. – you really only have Nikon or Canon as choices in the smaller formats. None of the other small format brands have meaningful perspective control solutions, and these are necessary for my architectural and product work.

I believe that either choice is equally valid and the ‘best’ tool will come down to your individual preferences for ergonomics and perhaps specific but frequently used lenses. There are other tradeoffs, of course – but over the long run, these average out. Canon’s early superiority for noise control was lost to Nikon from the D3 generation, and Nikon lost the resolution lead last year. No doubt it will swing again in the next cycle – I do not see these as good reasons to switch systems, because they are incremental and cyclical differences. Yes, I did run a 5DSR for a while – that was because I wanted access to specific lenses that could not be mounted on a Nikon. In the long run, it was impractical to carry 1.5 systems and lenses, plus the dynamic range of that sensor and rendition were different enough to give me headaches in postprocessing.

Long time readers will know I also trialled Pentax 645Z and Sony A7RII-based systems. The 645Z ticked most of the boxes, but the lens system turned out to be simply too limiting. I would still have to carry the Nikons for work requiring perspective control. On top of that, whilst the cameras themselves have been extremely reliable both in my own experience and that of others, the limited support in my country (as far as I know, there are fewer than a dozen cameras here in total) is also problematic should something go wrong.

I think my relationship with the A7RII is sufficiently well documented to avoid having to go into detail. Suffice to say Stockholm syndrome did not set in, and whilst there are tilt shift solutions available via adaptors (Canon, because of electronic control of the diaphragm) a complete system would require the use of quite a lot of adapted lenses – with all of the attendant problems of planarity, skew and decentering which accompany the use of any adaptor. I have a whole shoebox full of adaptors that work well with some lenses but not others, and keeping this straight just turned into a nightmare. Still, there were enough interesting lenses that I was willing to try – the Zeiss Loxias, Batises and older Contax-Yashica Zeisses (2.8/35 PC Distagon, 2.8/85 Sonnar, 100-300 Vario-Sonnar specifically, and also applicable to the Canon). Whilst the setup works well on a tripod, the other limitations were too unpredictable under fast moving situations – and we’re back to carrying extra Nikons again.

But I keep coming back to medium format because it seems as though something in the way it makes me work – or the hardware itself, though the former is more likely – results in a much higher hit rate, more satisfying images and to a somewhat lesser degree, happier clients. Even if most of this is intangible, the difference is there. And once you see it, you can’t un-see it.

I have now owned two medium format digital systems – Hasselblad V (CFV-39 and CFV-50C) and Pentax. I’ve also briefly used the Leica S, Phase One IQ250, and of course the Hasselblad H – in H4 and H5 incarnations. My experience with all of these systems has told me that most have significant compromises, and those are all different – pick whichever one is least of a problem for you. However, a lot of these have gone away with the transition to CMOS – and some other parameters like dynamic range have been improved even further. There was certainly a point where the D800 and co basically invalidated any argument for low end CCD-based (22, 31, 40MP) medium format – it beat them in price, image quality, and left them far behind in versatility and shooting envelope. and not only that, did things those cameras simply could not, like high frame rates, high ISO and focus tracking. However, now that the underlying sensor technology has been equalised, two things have happened: volume has driven price down quite a bit (A H5D-50C is about 40% its original launch price – $14.5k vs $25k or thereabouts) and the shooting envelope has opened up again. In the same generation of technology, bigger is always going to be better.

My underlying reason for trying any new hardware is always because I have reasonable reason to believe that it will do something my existing kit cannot: that may be resolving power, low light performance, rendition, color accuracy, speed, responsiveness, perspective control, etc. The hardware is always nothing more than a tool which must be subservient to my desire to make a certain idea into visual output. The one exception to this is when I shoot for myself – personal projects or art. Then, the camera must feel right. I have no way to quantify this, other than you will know it when you pick it up – and it varies from person to person. That was, and remains the reason I shoot with the Hasselblad V cameras – despite massive limitations like 1/500s top speed, no wide solutions on digital, and 100% manual operation – plus the newest lenses being old enough to vote. But I’ve always enjoyed the experience, and the simplicity and immediacy that comes from only really having three controls – focus, aperture and shutter – in one place. Not to mention that enormous viewfinder. (The best way to use the digital back is like an endless film back where you can change the sensitivity.)

It was for this reason I purchased a CFV-50C late last year after the price reduction – it had just crossed over the threshold into affordability. The recent garage sale was to cover the cost of acquisition; it is still by no means cheap, but not a decision I regret. The color and tonal output from the CFV-50C was far better than anything I’d used up to this point. It isn’t just the outright color accuracy, but the sensor’s ability to distinguish between very similar hues – the latter is critical in reproduction work and I find tends to be of greater concern for clients. Unusually, almost no calibration was required to achieve extremely high transparency, and the sensor’s response appeared to be similar to that of the D810 (highlight biased, long shoulder) – but with even more dynamic range. The difference between the 645Z (which uses the same sensor) and the Hasselblad versions is much like the difference between the D810 and D800E – less linear, more natural, and qualitatively better in every way. Once you experience the other one, you have a really hard time going back.

Given the limitations of the V system – this is obviously not what I am switching to. I was lent a H5D-50C while my CFV was being repaired**, and used it on several assignments. Firstly, the camera didn’t get in the way; secondly, it did things I could not do with the other hardware at my disposal: mainly to do with color accuracy and tonality/dynamic range, which are things clients can definitely see.  Of course this is not really surprising given the H5D-50C and CFV-50C share the same sensor and image processor – it was just nice to be able to do it at 22mm-e and with the option for autofocus or shift.

My clients were very happy with the look, too: ‘natural’ was the term that was constantly used. Feeling overconfident, I used it to cover the Thaipusam festival this year, too – the Q that rode shotgun as backup didn’t get used at all, nor did I feel like I missed any shots by not using it. There’s also a unique (to my knowledge) spot metering mode that allows you to select a zone (as in the zone exposure system) and use the spot to define which portion of the image falls in that brightness range. It’s the closest thing we’ve got to ETTR in a camera with an optical finder – just set it to 9 or 9.5. Overall, the results really made me feel as though this might be The One. Thinking it might be my imagination – I shot again with the D810 and Otuses recently to question my objectivity, which only confirmed that the Otuses are still the best glass out there – but the D810 isn’t matching the H5’s (or CFV’s) sensor.

**This is what I mean by professional service: if your camera has a problem, you really get another one until it is fixed. Despite the promises of NPS, I’ve never consistently had this level of service (usually, the cameras seem to be loaned out to the media and you are left hanging or self-insuring) from any other brand, and it really matters when your job and reputation are on the line – not to mention on a camera of this cost.

I began to seriously a) run economics and ROI calculations and b) evaluate the whole system for completeness to the edges of the envelope to see if a switch was feasible. I would not have considered it if the H5D-50C price hadn’t recently been cut significantly; if the MF players want to stay in the game for the long run, drastic action like this might be needed – the number of pros and studios here who’ve gone to FF instead of upgrading their MF systems is legion, at least in my part of the world. But there are also others who are relatively new to the game like myself, who are going the other way. In my experience, Hasselblad’s two strengths over the other MF solutions are the aforementioned color accuracy – which hugely minimises time wasted in PS for critical product work – and a little gadget called the HTS. The HTS goes between body and lens, adds a 1.5x crop factor to expand the lens’ image circle, but gives almost every lens tilt and shift capability. Lenses were designed with this in mind, which means the optics play nice together. In practice, instead of carrying both fast or AF 24/45/85mm lenses and the PCE set (which I frequently do on assignment) – I can now carry one set of lenses and the HTS. There is actually an appreciable weight saving, too.

After a huge amount of thinking, scouring of historical shoot packing lists, EXIF data and final client deliverables – purchasing mistakes are going to be highly costly – I came to the conclusion I could get away with four lenses: the 24mm, 28mm, 50mm, and 100mm; HTS, 1.7x TC and V-H adaptor. I cover roughly (in 35mm-e terms) – 19, 22, 40, 80, 135mm with AF, and 30, 60, 120 and 200mm with movements, and flash sync to 1/800s. The V adaptor adds 50, 80, 120 macro, 150 and 250mm lenses – again with flash sync to 1/500s. I can even pack the V body as a spare. (There are also various unhealthy combinations involving all three adaptors, but let’s not go there.) If you’re wondering why 24 and 28 – the 24 will pretty much live on the HTS for architectural and interior work, and after combined crop factors is a ~27mm. The 28 is 33mm, which isn’t wide enough most of the time. The 28 is a more friendly focal length for handheld work, and also 2/3 stop faster.

At this point, I also have an announcement to make. For the first time, I have formally tied up with a brand: I have been appointed a global Hasselblad Ambassador for 2016 (official announcement here), for which I have to thank the team at Hasselblad (specifically, William Penrice and Anders Espersen) and Shriro in Hong Kong and Malaysia for their support and for making all of this possible.

I’m excited because I think a) image quality really takes a step up, and b) my clients appreciate the difference. I enjoyed shooting with the CFV-50C, and wanted that image quality all the time – with an even wider envelope. (I still of course enjoy shooting with the CFV, but that’s now my ‘weekend camera’.) The H5D-50C gives me that, and saves my back somewhat. For the curious – one Nikon and the more specialised lenses (Otuses, 20/1.4. 180 APO) will stay for the time being during the transition. I am not suggesting even for a moment that the H system (or even medium format for that matter) is a perfect solution for everybody – far from it, given the frightening cost, limited AF and frame rates and overall size – but I do believe I’ve found a better fit for the way I shoot and my own professional and personal needs. MT

Coda: Contrary to popular opinion, I don’t think medium format is dying because of improvements in 35 FF – far from it. If anything, it’s the opposite: as awareness of precisely what improvements can be had at the top end increases, the demands are getting greater, at least on the client side. And it’s important to remember that for a company like Sony that manufactures sensors at all sizes, architecture developed for smaller sensors can easily applied to larger ones – if anything, there are some great economies of scale to be had here since a new sensor does not have to be developed from scratch. Ergo, what we gain from 35 FF and smaller applies to MF, too. Remember, BSI CMOS technology was really driven by camera phone imagers before it becomes mainstream in larger formats…

The equipment mentioned in this post is available here from B&H in the US, or Shriro in Asia. More information is available at www.hasselblad.com
Hasselblad H5D-50C (with wifi)
Hasselblad HCD 4.8/24
Hasselblad HCD 4/28
Hasselblad HC 3.5/50 II
Hasseblad HC 2.2/100
Hasselblad 1.7 TC
Hasselblad HTS 1.5
Hasselblad CFV-50C

__________________

Visit the Teaching Store to up your photographic game – including workshop and Photoshop Workflow videos and the customized Email School of Photography. You can also support the site by purchasing from B&H and Amazon – thanks!

We are also on Facebook and there is a curated reader Flickr pool.

Images and content copyright Ming Thein | mingthein.com 2012 onwards. All rights reserved