Fuji GFX Mirrorless System vs Hasselblad

It looks like Hasselblad are up to it again in trying to pre-empt Fuji in releasing a new model. The last time, they released the X1d ahead of the GFX50s even though Hasselblad had no previous experience with mirrorless cameras. The end result was a Hasselblad beta model rushed into production in an effort to steal the limelight from Fuji. The Chinese DJI owned firm didn't even manage to incorporate live exposure preview when first released.

Now Hasselblad appear to be rushing a new model into production with the same 100MP Sony 4433 (44 x 33mm) medium format BSI sensor that Fuji intend to use on their GFX100. As yet too little is known about the specs of either of the two new bodies. So far the Fuji GFX system has proven to be the one that has made the biggest impact in the marketplace. It is looking like medium format has returned to the mainstream of photography once more. So much so that 35mm format can no longer be called "full frame". In fact, Phase One calls its 645 medium format cameras "full frame" and refers to 4433 medium format as a "crop sensor". I guess the terms "full frame" and "crop sensor" are just relative terms.


In any case, the following is an old G+ (now defunct) post from 2017 that I will repost comparing the GFX50s with the X1d. Since this comparison, Fuji have released the GFX 50R which is not only much more affordable but it is also much more compact, being only a little larger than the X1D.


Also of interest is the size comparison with lens mounted. The Hasselblad X1D is shown with the 65mm f/2.8 lens and the GFX50R is shown with the 63mm f/2.8 lens:


Note that the Pentax 645D (and hence 645Z) is as large as either rival, even before mounting a lens. Keep in mind that the Pentax 645 cameras are SLR designs based on the film era Pentax 645 mount. The trouble with the Pentax 645Z model is that it is not a 645 format camera, but a 4433 format camera. That means the Pentax 645 mount is one size format larger than is necessary to host a 4433 format sensor. This increases the flange distance (70.87mm), although the mount diameter remains somewhat small relative to the format (61.2mm compared to 65mm for the Fuji G mount). The excessively long flange distance on the 645Z will degrade image quality.

[Repost] A Comparison of the Fujifilm GFX 50s vs. the Hasselblad X1D-50c



We are currently seeing two rival mirrorless 4433 medium format systems being released by Fuji and Hasselblad. This is a detailed comparison based on currently known information. Updates will be added to this post as new information is released.

4433 Medium Format

There have been various medium format film sizes in the past e.g. 645, 670, 680, 690. The 645 film format was 6 x 4.5cm in dimension, although the actual image area within the film was more like 56 x 41.5mm. Currently there are only two medium format CMOS sensor sizes available: 43.8 x 32.9mm or "4433" format (being about 44 x 33m), and 53.7 x 40.3mm or "645" format (being similar to the image area of 645 film). The Leica S system has a 45 x 30mm medium format sensor but their 4530 sensor is just a minor variation on the 4433 sensor.

The two new mirrorless medium format systems from Fuji and Hasselblad are based on the Sony 4433 sensor, with its approximately 44 x 33mm size. It is the same basic sensor also found in the rival Pentax 645Z (which really should be renamed the 4433Z since it is not based on a 645 format sensor). The Sony 4433 sensor offers 14.7 stops of dynamic range:

https://www.dxomark.com/Cameras/Pentax/645Z

Fujifilm also have a sensor customisation that extends the photic saturation point of the sensor, and increase dynamic range by 1/3 stop up to a hypothetical 15 stops (this needs independent confirmation):

http://fujifilm-x.com/de/x-stories/gfx-technologies-1/

In terms of sensor size, 35mm format has a crop factor of 1.27x compared to 4433 format. By comparison, APS-C format has a 1.52x crop factor compared to 35mm small format. The step up in format size going from APS-C to 35mm small format is much larger than that going up from 35mm format to 4433 format. Relative to a 35mm small format sensor, a 4433 medium format sensor has a 0.79x crop factor. More important may be that Fuji offer a small format system with a 1.52x crop factor alongside a medium format system with a 0.79x crop factor, with the medium format GFX system having approximately half the crop factor of the small format X system.

A 35mm format sensor has a 1.36x crop factor compared to a 645 format digital sensor. The difference in size (1.36 - 1.27 = 0.09x) expressed as crop factor between 4433 and 645 format is therefore somewhat trivial from an optical engineering perspective (though certainly not from a blindly emotional or snob-factor perspective). There is also a dramatic difference in price between 4433 and 645 format digital sensors. The other advantage of 4433 format over 645 format is that it permits the lenses to be designed with larger maximum apertures, or else to be made more compact.

Format size comparisons based on sensor square surface area are often quoted by manufacturers, but this figure is highly misleading and intended to fool fanboys and other uncritical buyers. If you see this, it should largely be ignored as marketing hype. The crop factor and the linear resolution difference resulting from it are much more critical figures from an optical engineering perspective. The only exception is with respect to noise. A sensor gathers light like a bucket gathers rainwater. The larger the bucket, the more capacity it has to gather rainwater, or in the case of a sensor, to gather light. This capacity is related to squared surface area. The more the light gathering capacity, the better the signal/noise ratio.

The greater squared surface area permits a 4433 sensor to have larger sized pixels than an equivalent 35mm small format sensor. The pixel size of the Sony 4433 51MP sensor is 5.3μm, which is 1.7x larger than the 4.1 μm pixels on a 50MP 35mm small format sensor such as that found on the Canon 5Ds. This theoretically improves the signal/noise ratio.

Availability and Professional Support Services

The GFX 50s will be available as of late February 2017. Although Hasselblad managed to beat Fuji to the announcement of their system, they have been dogged by a series of prolonged set-backs and delays in making their X1D-50c widely available, leaving a large number of buyers who placed pre-orders in the lurch. Fuji look like they could supply the market with their finished model in decent numbers sooner than Hasselblad, although Fujifilm too have reported that orders have exceeded expectations and that there will be a shortage of units for about a month after release date.

Fujifilm have deeper pockets in the R&D and manufacturing department than the cash-strapped, near-bankrupt Hasselblad, which means Fuji are less likely to rush a bug ridden unfinished beta product first to market just for marketing purposes. When there have been bugs in the past, Fuji have been exemplary in how quickly they have responded in promptly fixing issues.

Concordant with its better financial position, Fujifilm look like they will launch a professional support service for the GFX 50s:

http://www.fujirumors.com/fujifilm-will-launch-professional-support-service-along-gfx-camera-new-source-pentax-645z-vs-fuji-gfx-size-comparison/

One of the major reasons for the dominance of Canon in professional photography is their legendary professional support programme. The availability of a professional support service alone would be reason for many in the target pro demographic to instantly choose the Fuji GFX50s over the Hasselblad.

The Fuji G Mount vs the Hasselblad X Mount

Both of these mounts are new. Far more than with DSLR mounts, the ideal dimensions of mirrorless mounts remain somewhat uncertain. So the creation of every new mirrorless mount tends to be a brave new experiment. It cannot be automatically assumed that every experimental mirrorless mount will turn out to be optimal in the long run. For reasons unknown, the vast majority of commentators have unquestioningly leapt to the grossly unfounded assumption that all can be presumed to be well and good with the optical foundations of the two mounts.

The flange distance of the G mount is 26.7mm. By comparison, the X mount has a flange distance of approximately 20mm. This difference of around 6.7mm is very striking indeed. The shorter the flange distance, the closer the sensor is to the rear exit pupil, and the better the image should be. So far it may seem that Hasselblad is ahead.

The interesting thing about the G mount flange distance is that we would predict, based on the flange distance of the Fujifilm APS-C X mount, that the G mount should have a flange distance of approximately 34mm:

Fuji X mount flange distance = 17.7mm
Multiply x 1.52 crop factor to get 35mm small format equivalence = 26.9mm
Multiply x 1.27 crop factor again to get 4433 format equivalence = 34mm

Yet at 26.7mm, the new Fuji G mount has a much shorter flange distance than the predicated 34mm distance based on these crude calculations. The G mount flange distance of 26.7mm is the almost the same as that of the 35mm small format Leica M mount at 27.8mm, which is also a "mirrorless" mount because rangefinders lack a mirror box in their optical pathway. The Fuji G mount has a flange distance one format "too small". Even more astonishingly, the Hasselblad X mount has a flange distance even shorter than the Fuji G mount—at 20mm it is so ridiculously short it is virtually the same as that of the micro 4/3 mount flange distance of 19mm.

This follows a trend towards increasingly short flange distances started when mirrorless APS-C mounts flange distances were all made shorter than that of the M4/3 mount. This trend continued with the Sony FE mount, which was originally an APS-C mount (NEX mount), but which was retrospectively converted into a 35mm small format mount. That means that the FE mount is a 35mm small format mount with APS-C dimensions, including an APS-C flange distance of 18mm (comparable to the Fuji X APS-C mount 17.7mm flange distance). Sony thus continued a trend towards creating newer mounts with a flange distance deliberately set to be one format "too short" compared to previous generation mirrorless mounts.

As already stated, shorter flange distances permit great acutance, and that is the reason for this new trend in mount design. The shorter the flange distance, the better the image quality, and both the G mount and X mount have dramatically shorter flange distances than the huge 70.87 mm flange distance of the Pentax 645 mount.

The Hasselblad X mount flange distance is 6.7mm shorter than the Fuji G mount, and is therefore theoretically able to provide better image quality. The shorter the better—it seems so simple, doesn't it???? In fact, why didn't Hasselblad go in for the kill by making the flange distance almost zero...meaning that the light would hit the sensor corners horizontally sideways.

Corner Angle of Light Incidence

Here is the catch with making flange distance "too short". The shorter the flange distance, the steeper the corner angle of light incidence. This creates a problem with modern digital sensors because if you look at a digital sensor under a microscope it looks like a city with lots of high-rise sky scrapers. At midday, with the sun straight above, it illuminates the city well. However, at sunset or sunrise, the light hits the "buildings" at a steep angle, casting deep shadows. That's what it's like at the corners of the sensor when the flange distance is too short: you get shadows cast. Digital sensors only fully register light when it hits the sensor nearly square on, at close to right angles. If the angle of light incidence is too steep, as it can get at the corners, the sensor fails to register light information. The result is loss of light intensity (shading) and loss of resolution.

So there is your answer to why Hasselblad couldn't just reduce the flange distance to zero. It would mean that light hits the sensor corners at too steep an angle. It would have caused dreadful corner performance, however stellar the centre acutance.

Although the problem of angle of light incidence on digital sensors is poorly appreciated amongst photographers, it is universally regarded as a critical issue amongst optical engineers, making it a major area of photonics research. The ultimate solution that would allow ultra short flange distances, while allowing light to hit the corners at optimal angles, would be to make curved sensors to reduce the angle of incidence of light in the corners:

https://www.dpreview.com/articles/6229436014/sony-s-curved-sensors-may-allow-for-simpler-lenses-and-better-images

Curved sensors are still in the experimental phase, and it does not look like they will become a feasible option for some years to come. In the meanwhile, sensors look like their vertical structure is going to become more built up and complex with the advent of stacked sensors: the sky-scrapers are getting taller and casting deeper shadows. That means, for now, ultra short flange distances look set to become an ever greater headache for optical engineers. Meanwhile the lay public seem to be in gross denial over the very existence of this issue. Sadly, burying your head in the sand does not fix a very major problem.

Offsetting Steep Angle of Corner Light Incidence

Fujifilm has dug into their deeper R&D pockets to develop novel technology to overcome the issue associated with the steeper angle of light incidence in the sensor corners. Such R&D always tends to be costly, and no company would invest funds into something unless there were meaningful concern that it posed a serious obstacle. Fuji have developed a novel micro-lens and silicone process to allow the sensor to tolerate a steeper angle of light incidence. We have seen before that Sony's makeshift solution to the problem was to add BSI to their sensors, even if that means sacrificing dynamic range.

Fuji are doing something similar to what researchers in a recent Nature Photonics paper described regarding the use of a novel black silicon sensor to improve the tolerance to steeper angles of light incidence:

http://www.nature.com/nphoton/journal/v10/n12/abs/nphoton.2016.226.html

What Fujifilm offer in the GFX 50s is a heavily customised Sony sensor based on similar optical engineering principles allowing a sensor with a modified silicon process to tolerate steeper incident angles without having to go as far as to find a way to manufacture curved sensors. This also reflects the greater time and money Fuji have had to invest in the optical engineering R&D of short flange distance mirrorless mounts. Such modifications are clearly not standard features of the Sony 4433 lens, or else Fujifilm would never have sunk a considerable amount of R&D funds into developing such technology.

The customised modification to the micro-lens and silicone process replaces the X-Trans RGB filter customisation found in Fujifilm APS-C cameras. Although the X-Trans filter reduces moiré on small format sensors, 4433 format does not suffer much moiré, and Fujifilm claim the benefits of adding the X-trans RGB filter to a 4433 sensor are only modest while adding a further 20-30% increase in data handling burden to the processor:

Toshihisa Iida: "X-Trans is a 6x6 filter arrangement, not 4x4, it’s something like a 20-30% increase in processing requirement."

https://www.dpreview.com/interviews/6648162116/cp-2017-fujifilm-interview-we-hope-that-the-gfx-will-change-how-people-view-medium-format

Fuji have therefore prioritised offsetting the problem of the steeper angle of corner light incidence since the G mount produces a steeper angle of corner light incidence than the Fuji X mount.

What have Hasselblad done by comparison to offset steeper corner angle of light incidence? Nothing. There is not the slightest bit of evidence (e.g. patents, public statements) they had invested finances into R&D to mitigate this critical problem. But they do offer a drastically shorter flange distance, resulting in a markedly steeper corner light incident angle than the Fuji G mount.

Sadly, as with the lack of any mention of real-time exposure preview (as we will discuss later), when Hasselblad fail to do a huge "song and dance" highlighting the addition of a critical feature, instead conspicuously avoiding all mention of it whatsoever.

Redesigning Lenses to Offset the Steep Corner Angle of Light Incidence

There is a way of achieving this without major sensor modifications, and it is to make the lens design more telecentric. The steeper the angle of corner light incidence, the more telecentric the lens has to be. A couple of factors increase the corner angle of light incidence. Firstly, the wider the field of view, the steeper the angle of light incidence. Secondly, the wider the maximum aperture, the steeper the angle of light incidence at the corners. So that means, the wider the maximum aperture and the shorter the focal length, the more telecentric the lens has to be to compensate

Problem solved. It doesn't matter how short the flange distance, you can always partly offset the problem this causes by making the lens more telecentric in design.

But Houston we have a problem! We have replaced one problem with another one, because making lenses more telecentric makes them really big. It's all very well making the body smaller and more portable with a shorter flange distance, but if it makes the lens dimensions blow out in size, it just negates the size benefits from a smaller body (hmmm...maybe it might be better to compromise on the body size to reduce your lens size by increasing the flange distance). Never mind, you can always get your advertising department to just show how amazingly small the body is...you can then get buyers to overlook how elephantine the lenses are. Out of sight, out of mind. Fanboys will be eating out of your hand.

On the other hand, the fact that mirrorless ultra-wide lenses can omit retrofocal elements helps to offset the size increase. This means that lenses for ultra short flange distance mounts favour slower moderately wide angle lenses i.e. pancake or semi-pancake style lenses. This tendency will be more marked on the Hasselblad X mount than on the G mount, making the Hasselblad the ideal pancake or semi-pancake lens system for high-end walkabout casual shooting.

The Solution to Avoiding Huge Telecentric Lenses 

There is one, and it is to make the lenses slower! Problem partially solved. That is exactly what Hasselblad does...make slower lenses to avoid size blowouts. It means that the X system is ideal for walk around shooting with slower wider angle lenses e.g. for street and travel photography or even for putting into a backpack for landscape photography. The Hasselblad would be perfect for such purposes, and the shorter flange distance will mean that centre acutance will often be superior to the Fuji.

The reason for this is probably to avoid competition with the Hasselblad H series, which can take a 645 format digital back. If you want faster and more seriously professional Hasselblad lenses, the design intention here is that you should look at their H series of DSLR cameras. The Hasselblad X system is probably intended for more informal shooting. On the other hand, Fuji no longer have a medium format SLR or rangefinder series above the GFX, meaning that there will be no internal competition. That means the GFX has to be more of an all-round workhorse suitable for a wider range of professional applications.

In a sense, the X1D and GFX 50s are totally different cameras, and the reason for this lies in their very different flange distances.

But I Want Faster Aperture Lenses!

Then, between the two models on comparison here, the GFX 50s is the better option. The novel micro-lens array and silicone process on the G mount plus the longer flange distance means that this system can better support large aperture lenses for professional use without them becoming too absurdly elephantine and expensive. That's why the G mount lenses are consistently faster than the X mount lenses.

Still, given all the ignorance out there, expect Hasselblad fanboys to mindlessly ask why they can't make fast X mount lenses. This is based on the grossly premature assumption that an ultra short flange distance mount can support the same full range of lens development in a variety of focal lengths and aperture speeds just like on a traditional DSLR mount. If Hasselblad gives in to demands and makes fast X mount lenses, mark my words: expect to wet yourself laughing when you see how elephantine they are. Lenses like that, even if they are technically feasible, destroy the design philosophy of a more compact system.

Hasselblad are better off sticking to moderately wide angle lenses for the X mount designed with slow aperture speeds for walkabout shooting, and reserving faster professional lenses for their H series. This also prevents internal competition between your own product lines of the sort you see between the Sony E vs. A mounts.

Central Shutter vs Focal Plane Shutter

The Fuji also has the advantage of a maximum mechanical focal plane shutter speed of 1/4000s compared to only 1/2000s on the central shutter on the Hasselblad. When the electronic shutter is enabled, the Fuji can shoot at up to 1/1600s. The range of options include electronic shutter, electronic first curtain shutter, mechanical + electronic shutter, electronic first curtain shutter + electronic shutter. The all-electronic shutter mode without second curtain shutter will make it whisper quiet. The Hasselblad altogether lacks any option of an electronic shutter whatsoever.

The electronic shutter of the Fuji further reduces shutter shock, decreases shutter lag, and is whisper quiet. However, at over 1/1000s it increases vignetting, can introduce motion artefacts, and bokeh may be adversely affected.

One of the major selling points of the Hasselblad is that it offers central ("leaf") shutters. This allows you to shoot with strobes at shutter speeds up to 1/2000 to overpower sunlight or stop action. By comparison, the Fuji is limited to a meagre 1/125 flash sync speed. On the other hand, a central shutter does mean that you end up buying a new shutter with every lens, although it does make the body even smaller. Central shutters are made by hand in small volumes thus disproportionately increasing the price of an overall system.

Conversely, central shutters may make the lens a touch larger, or else limit the maximum aperture. For example, Leica officially states of their 100mm f/2.0 Summicron Leica S lens that "due to its high light sensitivity, the Summicron requires lens elements with a larger diameter, so that a central shutter could not be accommodated". Fuji may not have been able to offer the upcoming 110mm f/2.0 lens at all if the GFX 50s had relied solely on central shutters.

The key advantage of a focal plane shutter is that you only have to buy one shutter, rather than paying for a new one with every lens. However, the real reason for Fuji omitting central shutters from their design is that in a few years time, the advent of global shutters will render central shutters technologically obsolete. If you buy into the Hasselblad system for the central shutters, in a few years time you could rapidly find yourself stuck with an obsolete system. A smart buyer might prefer to future proof purchases.

http://www.fujirumors.com/panasonic-develops-organic-sensor-with-global-shutter-impressive-123db-dynamic-range-made-with-fujifilm/

Global shutters also improve other parameters of performance rather than just permitting higher sync speeds, including reduced motion artefacts, with newer design iterations now permitting improvements in dynamic range.

In the meanwhile, as a stop-gap until global shutters become more commonplace, Elinchrom and Profoto are expected to release high speed synch options for the Fuji system in the near distant future.

Lens Development

Fuji have several years of experience in designing and manufacturing mirrorless lenses. They have proven that they possess the optical R&D infrastructure to bring out a remarkably complete lens microenvironment in a surprising short period of time. There will be three lenses available at launch (63mm f/2.8, 32-64mm f/4, 120mm f/4 OIS macro) with a further three more coming shortly thereafter (110mm f/2.0, 23mm f/4, 45mm f/2.8). There has also been mention of plans for other focal lens depending on customer feedback, with focal lengths of 18mm, an ultra-wide zoom, a telephoto zoom, a tilt-shift lens, and a prime lens around 200mm being discussed. Fuji lenses are both designed and manufactured in house.

Fuji have stated that lenses designs are being future-proofed to perform at 100MP resolution, which is reflected in the more complex optical formulae with an increased number of lens elements compared to rival medium format lenses. For example the Phase One-Schneider 55mm f/2.8 has eight elements, while the Pentax 55mm f/2.8 and the Hasselblad XCD 3,5/45mm both have nine elements. By comparison, the Fujifilm 63mm f/2.8 has ten elements. Medium format prime lenses within the 45-65mm focal length generally have similar numbers of elements (just as the Sigma 35mm and 50mm f/1.4 Art lenses both have 13 elements).

Unusually for a medium format maker, Fuji explicitly declare the build of the aperture control of the lenses, which are constructed with nine rounded aperture blades to produce more rounded bokeh. It remains a pity, however, that this is still less than the 11 blade apertures increasingly found on recent Sony lenses.

With Hasselblad there are also three lenses available at launch (45mm f/3.5, 90mm f/3.2, 30mm f/3.5). Manufacture of the lenses has been outsourced to Japanese firm, Nittoh. Outsourced fabrication invariably results in higher prices due to a middleman markup. Four more lenses have subsequently been announced as of March 2017: a 22mm, 65mm, a 120mm f/3.5 macro, and a 35-75mm lens. Note that although the Hasselblad macro may be 1/3 stop faster, it lacks image stabilisation because Hasselblad do not own patents for their own proprietary image stabilisation system again reflecting their more impoverished R&D budget. The 1/3 stop increase in maximum aperture has also necessated an increate in the filter size from 72mm on the Fuji, to 77mm on the Hasselblad. Still, it is reassuring that Hasselblad are going to be highly competitive with Fujifilm when it comes to quickly developing a wide range of lenses suitable for different purposes.

Like Phase One, Hasselblad remain suspiciously silent about their aperture control. The Hasselblad X series lenses produce hexagonal shaped bokeh, which suggests that the reason for their conspicuous evasiveness is that their lenses are only made with six aperture blades. Worse still, the hexagonal shaped bokeh persists even when shooting at maximum aperture. It is likely that having a central shutter on both Hasselblad and Phase One lenses might has something to do with the 5-6 bladed aperture blades often found on medium format lenses. Even then it is disappointing in that not even the plastic $125 Canon 50mm f/1.8 STM has been upgraded to a seven blade aperture design, raising questions as to why one would put up with less from lenses costing several thousand dollars.

Support for Adapted Lenses

This is the worst possible reason to choose a mirrorless system. There is growing evidence that adapting lenses designed for a DSLR mount to a mirrorless mount degrades performance. Although misalignment from the addition of an adapter to the optical pathway has been suggested as one of the mechanisms it is likely that degradation of performance may be more complex than assumed.

Mirrorless lenses need to be designed to be as telecentric as possible to work their best, something which is lacking in DSLR mount lenses. Mirrorless lenses also have a special element designed to increase the distance of the rear exit pupil from the sensor. This puts lie to the idea that mirrorless lenses are just DSLR lenses with a shorter flange distance. Whereas modern DSLR lenses are modernisations of older designs, mirrorless lenses have to almost be designed anew from scratch. Engineers would hardly bother to do all of that extra work if all they had to do was adapt DSLR lens designs to a mirrorless mount just by shortening the flange distance. The assumption simply fails to hold that all you have to do is correct the flange distance with an adapter and all will be well.

Due to the longer flange distance of the G mount over the X mount, it might seem plausible that G mount lenses could be adapted to the X1D in the same way that Nikon F mount lenses can be adapted to the Canon EOS mount. The problem is that G mount lenses lack a central shutter making this unfeasible.

ISO Range

On paper it would appear that these two cameras have the same base ISO of 100. That is not entirely correct because Fuji and Hasselblad probably use two different ISO standards. Whereas both Fujifilm and Pentax use the SOS standard, Hasselblad more likely use the more common REI standard. That means that the Fujifilm base ISO is actually 64 when translated to the more common REI standard used by most firms. The following is the conversion scale for converting Fuji ISOs to those found from Canon, Sony, Nikon, Sekonic—and probably also Hasselblad.

It was said, when comparing the Nikon D810 against the Pentax 645Z, that because the Nikon had an unusually low base REI ISO of 64, the D810 could produce pictures at base ISO almost as noise-free and clean as those of the Pentax 645Z at its higher base ISO of 100. This partially reduced any advantage conferred by a larger format sensor when shooting under ideal light conditions such as in studio, the medium format camera's native habitat.

Although it has largely gone unnoticed, one of the strongest selling points of the Fujifilm GFX is that it has the advantage of having the same sensor as the 645Z while having the same ultra low base ISO of 64 (in the REI system) as the Nikon D810. The X1D may suffer from the same problem as the 645Z of having a base ISO of only 100. Given the fact that there are no reports that Hasselblad ISOs have to be "converted" like Fuji ISOs, it is likely that Hasselblad use the SOS system. This might be reason to choose the GFX over the X1D for studio use because of an actual lower base ISO

Like the Pentax 645Z using the same sensor, the X1D has an ISO range of 100-25600, and although neither Hasselblad nor Pentax make it clear if this is a native ISO range or an extended ISO range, it is likely that this represents a native ISO range.

Fuji states that the GFX 50s has a native ISO range of ISO 100-12800. This is more like 64-8000 in the REI system. This is extendable to 50-102400 (around 32-64000 in the equivalent REI system). Reducing the base ISO to 100 may have reduced the upper native ISO by a stop, but this is compensated for by a higher extended ISO range. This adds to the impression that the GFX is optimised primarily for professional studio use and only secondarily for field work, whereas the X1D is optimised for carrying into the field, leaving the H system for studio work.

Focus Points

The Sony 4433 sensor has no phase detection autofocus (PDAF) points, and both the Fuji and Hasselblad cameras rely solely on contrast detection autofocus (CDAF).

The Fuji allows you to move your focus point over the majority of the total area of the viewfinder. There are a total of 425 (17 x 25) CDAF points selectable, which are selectable using the touchscreen. The viewfinder can be magnified 16.7x during composition by pinching to zoom on the touchscreen to ensure that the focus point is exactly over the point desired. This will allow you to ensure that, for example, the eyelashes are perfectly in focus during a portraiture session. It helps that the Fuji also has eye detection autofocus.

On prototypes, the Hasselblad had only one focus point stuck in the dead centre forcing you to always focus and recompose, but this has been fixed on production models so that it now has 35 autofocus points. There are less than a tenth of the number of autofocus points compared to the Fuji. Each autofocus "point" consists of a rather large square, which will make it difficult when you want the focus to be on an exact point such as the eyelashes, especially when eye detection autofocus is lacking. You may find that the focus ends up on the eyebrow or nose rather than the eyes. The Hasselblad autofocus points cover a slightly smaller portion of the viewfinder compared to the Fuji, and these cannot be selected using the touch screen—instead you have to hold down the AF/MF button to bring up the focus points to select one by using the front and the rear dials.

Electronic Viewfinder (EVF)

The Fuji sports a 0.85x class leading EVF magnification. Hasselblad have not released information about the EVF magnification, which is usually a sign that they prefer that we did not know.

The GFX has a 3.69M-dot OLED EVF. By comparison the X1D features a 2.36M-dot XGA electronic viewfinder. An OLED display is of higher quality than XGA.

The Fuji also offers a high resolution EVF with a frame rate of 60 fps to reduce display lag. By comparison, the frame rate on the X1D offers only 30 fps, which will produce viewfinder lag. By the time the viewfinder refreshes, a fast moving subject may have already moved out of view.

Fuji also offer the option of a detachable variangle EVF Tilting Adapter (EVF-TL1). This can be moved in wide variety of positions, even upwards by 90 degrees.

LCD Viewfinder Display

The Fuji features a 3.2 inch 2.36M-dot resolution touchscreen LCD display. The X1D by comparison has a 3 inch, 24-bit colour, 920K-dot resolution touchscreen LCD display. That means that the Fuji has an LCD screen not only 5mm larger but with around 2.5x higher resolution.

Touchscreens are becoming increasingly more common on high end models. Together Fujifilm and Hasselblad are helping to lead the way with both featuring touchscreens on their respective models. On the Fuji, this allows the user to select focus points using the touch screen, although a joystick ("Focus Lever") on the Fuji can also be use to select focus points. The X1D lacks a joystick in keeping with its minimalist design philosophy.

However, in addition to a touch screen, the Fuji further adds the practicality of an articulating variangle LCD monitor. This means you can shoot from more creative angles, thus increasing the versatility of the camera over a wider range of scenarios. Having an articulating screen invariably adds to bulk, but, if size is of prime concern, you are better off choosing a smaller format camera.

The Fuji further has an 1.28-inch 128x128-dot high resolution monochrome LCD sub-monitor at camera top with back illumination for use in the dark. This feature is absent on the X1D.

Live Exposure Preview

One of the principle attractions of mirrorless cameras is that the electronic viewfinder (EVF) allows a live preview of the exposure that simulates the exposure of the final photograph. By comparison, on a DSLR you can only tell if the exposure was correct after taking the shot. You may find out all too late, only after the crucial moment has passed, that the shot had been incorrectly exposed.

On all of Fuji's mirrorless models live exposure preview works extremely well. This is a standard feature of all mirrorless cameras from Olympus, Panasonic, Sony, Canon, Samsung etc. Live exposure preview is altogether lacking on the Hasselblad. The fact that Hasselblad even condescended to release a mirrorless camera lacking this quinessential feature beggars belief. This highlights Hasselblad's severe lack of experience in making mirrorless cameras. This one shortcoming is reason to instantly disqualify the more expensive X1D from consideration, unless you are only ever going to shoot with it in studio with strobes (on the Fuji, you can disable exposure preview when shooting with strobes)

Start Up Lags and Blackout Time

The X1D has a notoriously long start up time of around five seconds with another 2-3 second lag for the menu to come up. It also has another lag introduced by having to half-press to release the shutter, followed by a 2-3 second delay to initiate live view. Next there is a prolonged blackout time of around 2-3 seconds after taking a shot during which the camera loses all responsiveness. Playback is also slow, taking another 2-3 seconds before an image can be reviewed. The X1D also runs hots after startup.

The GFX 50s is almost instantaneous in starting up, runs cool, and there are no such lag times to engaging the camera, starting up menus, blackout times, or delays in playing back images.
Memory Card Slots

Both the Fuji and the Hasselblad have dual card SD card slots. The Hasselblad only takes slower UHS-I memory cards, whereas the Fuji takes much faster UHS-II memory cards. Even if the maximum frame rates on medium format cameras may be low, it does not change the fact that the larger file sizes mean that memory card write speed may still impose a significant technical limitation, especially when shooting video.

The Fuji also offer lossless compression on RAW files to reduce files size without degradation of image quality. This feature is absent on the X1D.

Maximum Frame Rate

X1D: 1.7 - 2.3 fps.
GFX: 3fps. Buffer allows up to 13 consecutive RAW shots to be taken.

These figures suggest that both models have only one processor, unlike the 50MP Canon 5Ds with dual processors and able to shoot at 5fps using a dedicated off-focal plane phase detection autofocus (PDAF) system. The Sony 4433 sensor has no on-focal (on image sensor) plane PDAF capacity (on-sensor PDAF). That means the GFX and X1D both rely on contrast detection autofocus (CDAF), which is inherently slower.

There is a noticeable buffering limitation on the Pentax 645Z, especially since it only takes UHS-I cards, and it remains to be seen if there are any buffering issues with either the Fuji or the Hasselblad cameras.

Video

Both models can shoot HD 1080 video but not 4K video. It is unclear whether the GFX will have either internal or external F log. The Fujifilm X-T2 only offers external F log. No log gamma options appear to exist for any Hasselblad systems.

Battery Life

The Fuji is rated for 400 shots (with Auto Power Save ON) from its large capacity battery, which also contributes to the large size of the body. In keeping with it being a workhorse camera for professional users, a battery grip (VG-GFX1) is also available. The Fuji battery has a capacity of 1230 mAh. Focus Numérique rate the number of shots per charge at around 300.

The X1D takes a 3200 mAh battery permitting approximately 300 shots per charge. A Hasselblad technical expert answering a buyer question on B&H Photo stated that:

"On average you can expect approximately 400 images. If Live View is used extensively, then you can expect that count to be lower."

Independent reviewers have estimated the number of shots per charge with average typical usage to be between 250-300.

No battery grip is currently available for the X1D, which may reflect the minimalist design philosophy of the Hasselblad, being more of a high-end portable consumer model than a professional's industrial workhorse. A battery grip would have ruined the svelte aesthetics

Colour Depth

The Hasselblad offers 16 bit colour definition, whereas the Fuji only offers 14 bit colour depth. Whether the Sony 4433 CMOS sensor is even capable of outputting true 16 bit colour depth data in the manner of CCD sensors remains somewhat doubtful, in which case this may be little more than hype. A now withdrawn DXO Mark score for the Pentax 645Z (14 bit colour depth) using the same 4433 CMOS sensor showed a colour depth score measured at 26 bits, which compares favourably with the 26.5 bit colour depth score from the Phase One IQ180 Digital Back (CCD sensor).

Extras

Both the Fuji and the Hasselblad offer focus peaking, but not zebras.

The Fuji permits the recording of voice memos and has Wi-Fi capabilities. The Hasselblad has both Wi-Fi and built in GPS.

Both the Fuji and the Hasselblad offer support for tethered shooting via a USB 3.0 connection.

The Fuji offers a novel Super Fine JPEG format with minimal compression, resulting in files so large that it takes up to 10 seconds to preview them in Lightroom. There is also the capacity to covert files in camera to TIFF format. The Fuji also offers a range of excellent film simulations.

The Fuji also offers the option of a novel RGB histogram in addition to the brightness histogram also found in the Hasselblad. The Fuji can also give highlights warning with the histogram.

The Fuji offers eye-detect autofocus. This has previously not be as precise as that found on the Sony bodies, but if Fuji have improved on earlier versions, it may start to become an indispensable feature during portraiture.

Another feature adding bulk on the Fuji are the multiple dials for exposure compensation, ISO and shutter speed. Many will like the palpability of having all elements of exposure at the user's fingertips, including having a manual aperture dial on the lens. Even though the Hasselblad lacks such dials in keeping with its minimalist design philosophy, the improvement in ergonomics from the touchscreen may make up for this.

Construction

Both models are made with an magnesium alloy body, and come weather sealed.

L-Plate Compatibility

The Fuji has an unusually large battery that is inserted sideways from the left. Unless someone invents a clever solution, an L-plate will obstruct the opening of the battery door. If using an L-plate with a battery rather than an AC adapter in studio is crucial this could pose a problem, as you would have to remove the L-plate whenever you replaced the battery. The X1D does not suffer this design problem, and should be readily usable with L-plates.

Size Comparison

See Fuji Rumours for images:

http://www.fujirumors.com/fujifilm-gfx-exact-fujifilm-gfx-50s-vs-hasselblad-x1d-pentax-645z-size-comparison-gfx-price-yen-leaked/

GFX weight: 825g (825 + 405 = 1230g with 63mm f/2.8 lens, plus EVF = 1325g)
X1D weight: 725g (725 + 417 = 1142g with 45mm f/3.5 lens)

Note that the variangle detachable EVF adds 95g to the total weight. Camera weights are inclusive of the battery

Notice the way there is a trend towards the disparity being reduced once total weight including the lens is considered. Always look at the total package including the lens, and not just at the body. The fact is that you cannot shoot without a lens attached, which, although seemingly quite blindingly obvious, is almost universally forgotten

Industrial Design

Problems with optical engineering aside, for some the decisive factor that makes them drool not over the Fujifilm GFX, but the Hasselblad X1D instead, is the sleek simplicity of the Swedish industrial design. The X1D is slimmer due to the shorter flange distance, and lack of an in-body shutter system. This reflects the fact that the Hasselblad CEO at the time, Perry Oosting, had a background in fashion working with Bulgari, Prada and Gucci. It is likely that this relates to the fact that Hasselblad are severely cashed strapped, and given that engineering R&D is forbiddingly expensive, they have wisely chosen to invest what little cash they have in industrial design and marketing.

The Fuji GFX design has by contrast been driven by optical engineers rather than fashion designers, and, as a result, form follows function. There are chunky industrial features on the Fuji like the optional detachable variangle EVF and an articulating LCD viewfinder. What a shame the Fuji engineering team had to add such practical working features for the professional photographer when it destroys the minimalist design aesthetic. We can't have that, can we

Now let's be frank. The only reason to choose the X1D based entirely on minimalist industrial design alone is so that you, the nouveau riche amateur can show off your Hasselblad badge as an arriviste status symbol like a Prada handbag. The X1D isn't designed as a work tool for the pro to earn money from, but as a fashion accessory to be shown off on your yacht. That said, be a critical consumer, if compactness is a priority, look at your lens plus body combination as a package and then decide if that is decisive for you, assuming that getting a smaller format system doesn't actually make more sense if size is driving your buying process.

Never mind the optical engineering compromises that go into achieving the sleek industrial design either. Always choose the pretty Swedish blonde bimbo over the sophisticated yet geeky Fuji brunette. Never mind the core of the X1D is a Japanese Sony sensor, the lenses are made by Japanese optics firm Nittoh, and that Hasselblad is owned by the Chinese, the corporate image of Aryan superiority is all that matters. Luckily, the X1D is much more than just a pretty face, but the point here is that looks are the worst possible reason to choose it over other rivals.

Conclusions

Both of these models are attractive offerings, yet in many ways they are more dissimilar than similar in terms of the guiding design philosophy based on intended usage. The greatly superior practical package for the working pro for use primarily in studio, and only secondarily in the field, is the GFX 50s, though the Hasselblad should still have much wider appeal to well-heeled fashion conscious consumers using this as their sleek walk-about camera for everyday use, leaving the Hasselblad H system intact as the studio system intended for serious working professionals. It is horses for course, as there is a place for both under the sun, and it is always good to see healthy competition.

Comments

Popular posts from this blog

PROOF that the Larger the Camera Format the Shallower the Depth of Field!

Follow Up: Effect of Format Size on DOF

PREDICTION: The Canon 1DX II Successor Will be a Pellicle Mirror Global Shutter Camera