It turns out those April 7 rumours were true after all: there’s a new H in town. The H5 I’ve been using for the last couple of months was announced in 2012, with the CMOS version arriving mid 2014. Given the long product cycle times in medium format land, four years is not too bad between iterations. In any case, following Phase One’s XF-100MP announcement and innovative suite of features, Hasselblad would have to do something not to get left behind. The important features are here: 50MP on 44x33mm, 100MP on 54x40mm, leaf shutter up to 1/2000s with new/updated lenses, a new 3″ VGA touch panel, improved live view, RAW video out (Apple ProRes) at 1080P30 on the 50MP version and 4K30 on the 100MP version. Firewire is now USB3.0, along with HDMI out and audio IO. Oh, and no more CompactFlash – it’s now SD and CFast. Lastly, there’s a new version of the tethering and workflow software – Phocus 3.0.
Today’s post is going to be a bit more than just a spec sheet: it’s also a little analysis of the state of medium format at the moment.
Let’s start with the good bits. The body and finder ergonomics, plus the camera-side UI remain largely unchanged from the previous H-cameras; this is good for heavy users because it means minimal disruption in the way you work. All of the things we liked about the H5D still remain: the parallax-compensating True Focus gyros, the zone spot meter, the profiles, the customisability and modularity. There wasn’t much wrong with any of this to begin with – and personally, I’ve found the H5D to be one of the most transparent cameras in operation – it just gets out of your way and lets you focus on making the image. I expect the H6D to be the same. How the new back UI works out in the field remains to be seen, but given that we interact very little with the back anyway other than for reviewing images – it’s unlikely to be make or break either way. That said, with effective implementation of the touch screen’s capabilities – I think it could speed up the workflow even more.
The sensors are Sony-sourced CMOS units, which means we’re looking at a derivative of the 50MP sensor in the H5D-50c and CFV-50c for the 50MP version, with a 5.3u pixels on a 44x33mm size and 0.8 crop to 35mm or 1.3x crop to 645. Maximum ISO remains at 6400 with a claimed 14 stop dynamic range; in practice, I find it’s a bit more than this – perhaps a little over 15 when you take into account highlight recoverability. The 100MP sensor is likely the same base unit as Phase One’s XF, which means 4.6u pixels on 53.4x40mm, ISO to 12800 and a native ISO of 64 – plus a claimed 15 stops of dynamic range, since this sensor uses a newer architecture. (Remember, the 50MP sensor dates to 2014, which means it probably shares its underpinnings with the D800 generation.) It’s also worth noting the cameras now do up to 60min exposures without needing a dark frame afterwards – I’ve tried the current generation to the 34min limit, and it’s clean. After a few years of running on older sensors and seeing 35mm close the gap to medium format, I suspect we’re going to see quite a bit of clear air open up again.
Hasselblad claims existing lenses will have no problems resolving 100MP – since the photosites are only a little smaller, and only the HCD lenses don’t quite cover 54x40mm. My experience suggests this is true, though the 2.2/100 will need to be stopped down a little. The other lenses I have are more than a match for the 5.3u pixels of the 50MP camera even wide open. It’s also worth noting that there are different finder masks for the two sensor sizes, so there are no confusing crop lines to clutter up the finder.
It remains to be seen how easy it is to deploy the 100MP monster in the field – higher shutter speeds will definitely be needed than for 50MP since the angular pixel density is higher, but if the sensor’s underlying noise levels are lower, we might just be able to crank the ISO up a stop with little penalty – and much more resolution. There’s also focusing accuracy to consider, and even if the camera can get it right 100% of the time – it’s important to remember a bit of subject or photographer motion is likely to make a huge difference in apparent focal plane, not to mention the finite size of the focusing box – and lastly, the fact that apparent DOF (and critical sharpness) are actually tied to resolution. Still, I’m up for the challenge.
Live view is claimed to be improved significantly from the previous generation – which is good, because it refreshed at 30fps and would easily blow out in bright daylight, requiring an ND filter or stopping down (making critical focus tricky). I presume this has something to do with the new electronic architecture and the higher data rates required for video and 100MP. As mentioned earlier, there’s also what appears to be a tentative foray into video: if video quality is anything like the stills, I think we may be at the start of something. As expensive as the cameras are – they’re still nothing compared to the cost of a major production.
The increase in maximum shutter speed is interesting. Some sophisticated electronic synchronisation plus improved shutter units allow 1/2000s with the new camera and new lenses, or 1/1000s with prior HxD cameras. (Note that existing lenses are still limited to 1/800s on all cameras, but can be sent in for the upgrade.) From an available light and handheld photographer’s point of view, 1/800s is borderline for anything longer than 150mm or so; even at 50MP the slightest shake can completely destroy any resolution advantage over 35mm. For the studio guys, 1/1800s at all apertures and with all lenses basically means you can overpower the sun with impunity. It’s also a clear shot across the bows of Phase One, whose 1/1600s max sync with leaf shutter lenses (not all of its lenses have leaf shutters, though) has been one of its strongest selling points for the studio pros. On top of that, a slower maximum speed of course limits possible apertures in bright light – 1/2000s and ISO 64 effectively means you can now work with an aperture two stops wider in the same light conditions as before, too.
There are basically only two complete system players left in medium format: Hasselblad and Phase One. I don’t consider Leica or Pentax to be complete systems: Leica lacks tilt shift lenses other than the 120, and Pentax lacks both leaf shutter lenses, tilt shift lenses, and a complete set of lenses whose performance is consistent and commensurate with the abilities of its sensor. Furthermore, as subjective as these things are, I don’t think the Pentax implementation of the sensor is quite as good as either Hasselblad or Phase One; its files seem to lack the depth and richness produced by the other two, or the accuracy and highlight handling of the Hasselblad.
CFast and SD – whilst I can understand why CFast is necessary for video with its 500mb/s+ IO speeds, it’s still very difficult to find cards and accessories in some parts of the world – where I live, for instance. Thankfully there’s also an SD slot.
Until now, Hasselblad lagged somewhat in sync speed and resolution, but trumped P1 on lens selection (specifically at the wide end) and perspective control options – there’s the HTS 1.5x tilt shift converter for most of the lenses 100mm and shorter – P1 requires you to buy an Alpa. P1 has some more innovative functions – the vibration-sensing shutter release and all-over touch UI, for instance, and higher overall shutter speeds (up to 1/4000s with the focal plane shutter). Their products are also significantly more expensive, but if you are a working pro able to justify medium format – it probably doesn’t matter that much. I think at this level, photographers select their tools based off necessity and support infrastructure rather than loyalty or fanboyism. If I was a studio photographer, I’d probably have gone with P1 (assuming they’d ever reply my emails, much less sell me a camera). But since a lot of my work is available-light, and needs movements, the HTS is the one killer bit of hardware for me. I also much prefer Hasselblad’s tonality and color.
The question of ‘do I need it?’ and ‘are such products even relevant?’ is probably going to be thought that follows immediate lust – the answer is quite simple: you’ll know if you do. To me, the biggest gains for MF over 35mm aren’t so much outright resolving power as dynamic range, color accuracy and handling of highlight transitions – the files are highly transparent and just don’t look ‘digital’ and ‘hard’, and that’s not something you can get anywhere else. But if you don’t have the output method to make the most of it – preferably large and/or high resolution print – then the difference may be somewhat lost.
Prior to the H6D announcement, the overall balance of things probably tipped somewhat towards Phase One: the one thing holding the fort for Hasselblad being pricing (50MP CMOS being $15k against a whopping $35k). Better is of course relative: this is the point at which we search for extra organs and children to sell: the H6D-50c has an European price of EUR 22,900, and the H6D-100c is EUR 28,900. (It’s worth noting that a) there’s a trade in program, b) the cameras have a three year warranty, and c) the 100MP version is some 40% cheaper than the Phase One equivalent.) With lenses at 5-6k a pop, this is not a casual tool. However, if we look at things in relativity – $7k Nikons and $50k Phase Ones, in a market that’s not more than 20,000 cameras a year – it puts things back in context somewhat. Reading further between the lines, the shared architecture between the CFV backs and H cameras suggests there’s probably going to be an update to the CFV, too – still not the full 6×6 square we all want, but getting ever closer. The H6D itself now brings both a superlative feature set in most areas, and very competitive pricing, too: I’d say Hasselblad has fired a shot that shows they’re firmly back in the game. MT
Ambassador benefits: I’m trying to get hold of a H6D-100c to test in the near future, the results of which will be published here. There are very few units in the wild at the moment, but I’m sure if there’s strong interest demonstrated in the comments it’ll move me up the priority list 🙂 I’ll also be publishing a closing report on my time with the H5D-50c in due course.
The Singapore Architectural Masterclass from 1-7 July 2016 is now open for booking. Click here for more details and to book.
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!
Images and content copyright Ming Thein | mingthein.com 2012 onwards. All rights reserved
More info on Hasselblad cameras and lenses can be found here.