• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

News: Hasselblad H6D.

Jerome Marot

Well-known member
Hasselblad just announced their new H6D platform, available in 50 and 100 mpix cmos: http://www.hasselblad.com/h6d.

Also: 4K video in raw on internal CFast card(!), new central shutter up to 1/2000s, touch screen interface, USB-3 connection instead of firewire.
 

Jerome Marot

Well-known member
Makes me wonder who buys these types of cameras and for what purpose.

Photographers? And to take pictures? It is still a camera...

Or is the question "who needs the 100 mpix resolution?". Then, there are various uses:
-aerial photography: the pictures you enjoy in google maps come from somewhere and flight time is expensive.
-museums who need to digitise their collection and would rather have the highest possible resolution.
-some high end fashion and luxury photographers need the retouching capabilities made possible by the high resolution and want to print large billboards at maximum quality (there is a Vuitton shop where I live with much larger than life posters on the building at the moment).
-some artists want the effect of massive prints and find these cameras more convenient than a 8"x10" view camera (which some other artists still use)
-even some amateurs prefer the look or handling of MF cameras.

MF is not a very large market, but it appears sufficient to allow Hasselblad and Phase One and to a lesser extent Leica and Ricoh-Pentax to run a business.
 

Asher Kelman

OPF Owner/Editor-in-Chief
I wonder whether Ricoh Pentax is able to offer lower priced versions of the same MF 50MP Sony sensors in use by Phase one, Leica and Hasselblad because they have the technology paid for by their large experience in small format cameras. Or are they simply competing against such established digital MF brands that the only way they can grab market share is to charge less for about the same capabilities.

Note, however, that Pentax does not AFAIK, have any 100 MP camera on the near horizon, nor does it have a lens shutter capable of synching flash at 1/2000 second for those who want to overcome sunlight with flash.

Asher
 

Jerome Marot

Well-known member
I wonder whether Ricoh Pentax is able to offer lower priced versions of the same MF 50MP Sony sensors in use by Phase one, Leica and Hasselblad because they have the technology paid for by their large experience in small format cameras. Or are they simply competing against such established digital MF brands that the only way they can grab market share is to charge less for about the same capabilities.

Dealer margins appear to be quite different for PhaseOne/Hasselblad than for Pentax cameras. Why? Because Hasselblad regularly lowers their entry camera to about the same price as Pentax, so you know there is some margin built-in. That is part of the price difference. In fairness I should say that PhaseOne/Hasselblad dealers are expected to offer a much higher level of service.

Note, however, that Pentax does not AFAIK, have any 100 MP camera on the near horizon, nor does it have a lens shutter capable of synching flash at 1/2000 second for those who want to overcome sunlight with flash.

There are rumours of a 100 MP Pentax and their cameras can run lenses with central shutter. If they felt the need, they would just have to issue one or two lenses.
 

Michael Nagel

Well-known member
I wonder whether Ricoh Pentax is able to offer lower priced versions of the same MF 50MP Sony sensors in use by Phase one, Leica and Hasselblad because they have the technology paid for by their large experience in small format cameras. Or are they simply competing against such established digital MF brands that the only way they can grab market share is to charge less for about the same capabilities.
AFAIK, the Pentax 645Z shares some of the inner workings with the Pentax K-3, this helps to keep down the cost.

On the other hand it is pretty interesting to see that the H6D boasts features that were introduced in the MF market with the Pentax 645Z such as:
Video (H6D offers a step up with 4k RAW from the Pentax FullHD)
USB 3.0 (digital MF used primarily FireWire before)
Dual Card Slot (not 100% sure, but can't remember that feature on other digital MF cameras)

Concerning MF - are you aware of the Pentax 6x7/67/67II?
This camera was well-regarded among professional photographers (which apparently did sell well).
The 67II went out of production (together with the 645NII) back in 2009.

Note, however, that Pentax does not AFAIK, have any 100 MP camera on the near horizon, nor does it have a lens shutter capable of synching flash at 1/2000 second for those who want to overcome sunlight with flash.

The lens line-up for the Pentax 67 included leaf shutter lenses capable of 1/500s (first introduced back in 1971).
These lenses can be adapted to the 645Z.

I don't know if such leaf shutter lenses will be introduced for the 645Z and I doubt it as the focus of this camera is landscape.

The H6D features look interesting. The larger sensor of the 100Mpix variant might be the more popular version.

Best regards,
Michael
 

Asher Kelman

OPF Owner/Editor-in-Chief
Look and impact of a picture!

Jerome and Michael,

Of course, it's not just the MP count, but also the recipe that delivers the RAW file and the flavors that the brand decorates its output files. This results in different impacts for the same resolution. According to one's esthetic preferences then, a camera with a lesser MP count can contend, well above its prescribed MP "fighting weight" class!

The announcement of another 100 MP camera in the new Hasselblad offering is welcome news to me. I do like the idea of more resolution when folk poke their noses, closely inspecting my prints in an exhibition from just 15-25 cm. I have even seen folk close enough for their breath to condense of the glass! Anyway, I do not like my prints to seem to break down at close inspection. Still I am able to make 0.5 meter sized prints that pass muster with my Canon 6D and easily deliver beautiful 1.5 meter prints with my A7R I.

Just the same, I would like to not have to work so hard, as I do, at the limits of the cameras capability. So I'd love to have much more resolution......... and. out of the camera, better dynamic range. Then I will be able to open shadows without having to now remove the resultant noise. Both the A7R and Nikon 810 would do that at about $3,000. However, the hurdles in selling my prints do not come from any lack of impressiveness. So as much as I would like a better camera, it won't necessarily help me sell my pictures! If the picture looks obviously rendered better by a more refined sensor and camera, I would go for it. In fact, I am thinking of returning to film. I would shoot with digital and then when I have the form I like, repeat it immediately with a few shots of 8x10 film. That I must try.

Back to the nature of the output of the digital camera: what's so interesting and challenging to me are the differences in processing of the data by the camera, delivered as the "RAW" image. My belief was that if I was a super-competent to the nth, I should get the same "look" and impact from any camera with a similar sensor. Well, my findings, using Adobe Camera Raw and Capture One, shows more impact from my Sony A7R at 30 something MP than an 80MP Phase One!....... (although the latter has definitely more resolution of hair.)

I have gotten so far behind. I have not yet tested the 50MP 5DS with is now just over $2,000. Perhaps that's no big loss. From a DPreview lab test, it appears that the 50MP badge of the 5DS does not deliver the relatively noise-free images comparable to its Nikon 810 and Sony A7 Mark II 50 MP peers, as it has much more read noise. The differences are obvious.

I know the Pentax 645D and 50 MP 645Z from Nicolas Claris' work. No complaints, AFAIK about noise. So far, I have not come across any comparison of the read noise in the shadows of the MF high resolution cameras and that is important. For sure, they do show exceptional rendering and presence. These latter terms are so indefinite and unscientific. Together with processing difference, choosing a camera above the level of the A7R I, is ,to me at least, a daunting challenge.

Not to be overlooked by our interest in more MP for detail, are sophisticated built-in camera options for customizing JPEGs out of the camera. With current presets, post-processing might not even be needed. Are we just stubbornly working hard ourselves with our RAW files? With these new cameras, are there many pro photographers delivering their images as JPEGS right out of their modern high resolution MF cameras, I wonder? I know several wedding photographers who do just that with their 35mm gear.

Asher
 

Jerome Marot

Well-known member
Why would they spend time on developing 4K video?

Now, that is a good question: is video useful to MF shooters? Who is going to use that?

I don't know what market they have in mind. Maybe they don't know themselves, but want to try new things and 4K video is very much "in". Also: keep in mind that they partnered with Arri for the Alexa 65.
 

Asher Kelman

OPF Owner/Editor-in-Chief
Why would anyone use 70mm film for a movie?

Michael,

It Needs to be shown 2.3:1 rather than 2.2:1 bu still has more detail and allows film editor more leeway in editing. 70mm film will be shown as projected video any way. But you know that!

I can imagine getting extra "coverage" footage of an "expensive to get to" or "to build" location of set. That way the film editor as a particularly rich source for mining other shots to make the edited movie richer. That's what I would use it for - coverage!


Asher
 

Michael Nagel

Well-known member
What is your point Michael?
The point is the same why a photographer would use MF instead of Kleinbild or APS-C: Image rendering.

I can imagine getting extra "coverage" footage of an "expensive to get to" or "to build" location of set. That way the film editor as a particularly rich source for mining other shots to make the edited movie richer. That's what I would use it for - coverage!
The point for using a larger sensor with the same 4k resolution is rendering. Mentioning 70mm film was just an analogy (pun intended).

The Arri Alexa 65 mentioned by Jerome was already a hint.

With the Phantom 65 by Vision Research being discontinued (Ametek apparently has other plans with their Material Analysis Division, Vision Research being part of it) and the upcoming IMAX 2D (derived from the Arri Alexa 65), the Hasselblad H6D is the only MF camera with 4k video you can actually buy (Arri has the the Alexa 65 only for rental).

The Pentax 645Z was the first step into MF video and is was more of an add-on.
Hasselblad seems to be more serious about it.

Best regards,
Michael
 

Jerome Marot

Well-known member
the Hasselblad H6D is the only MF camera with 4k video you can actually buy

Not quite. The Leica S (Typ 007) is also able to take 4K videos (but only in crop mode, non-crop mode is limited to HD).

In the near future (i.e. within a year or so), I would expect all new semi-pro and pro cameras to film in 4K. Whether it makes sense for MF is another question, of course.
 

Doug Kerr

Well-known member
Hi, Jerome,

Not quite. The Leica S (Typ 007) is also able to take 4K videos (but only in crop mode, non-crop mode is limited to HD).

I don't know anything about the details of that camera. But I assume that "crop mode" is a mode in which only part of the sensor is used.

Proceeding under that assumption, it is to me curious that using part of the sensor the camera can take video with a higher frame pixel resolution that in the mode using all of the sensor.

What am I missing here?

Thanks.

Best regards,

Doug
 

Jerome Marot

Well-known member
I don't know anything about the details of that camera. But I assume that "crop mode" is a mode in which only part of the sensor is used.

Indeed it is.

Proceeding under that assumption, it is to me curious that using part of the sensor the camera can take video with a higher frame pixel resolution that in the mode using all of the sensor.

What am I missing here?

Video is much lower resolution than photography. HD is about 2 mpix (1920x1080), "4K" is about 8 mpix (4096x2160). The Leica S is about 37.5 mpix (7500x5000).

The question becomes: "how does one go from 7500x5000 to 1920x1080 (or 4096x2160)?". The correct way would be to compute an average on neighbouring values, but doing so on such massive amounts of data 30 or 60 times per second needs a large bandwidth either to read the sensor or to compute the final data.

So the engineers choose something simpler. They can chose to either read along the whole surface of the sensor, but only read every so many lines or so (say, every fourth line) effectively implementing a cheap "nearest neighbour" algorithm. Or they can choose to read so many consecutive lines at the center of the sensor.

I think that the processor in Leica S (Typ 007) is only capable to compute 1920x1080 pixels in real time, and resorts to reading the center of the sensor to output 4096x2160. I am not so sure, tests are not available.

I also think that one of the first camera to be able to compute an average on its complete sensor for 4K images was the Sony RX10/II (and RX100/IV) and Sony manufactured a special stacked sensor for doing so, with the processors directly mounted on the back of the sensor to be able to have distributed connections and thus sufficient bandwidth for transmitting the signal.

The process by which lines are skipped, although widely popular, produces very annoying artefacts on moving images.
 

Jerome Marot

Well-known member
I'll take the example of the H6D, since this is the camera subject of the thread.

The H6D native resolution is 11600 × 8700, which is a 4:3 aspect ratio common in MF cameras.

The video aspect ratio is 16:9 and from published material, we understand that the video reads the full width of the sensor, skipping the top and bottom part for aspect ratio conversion. That would give a readout of 11600 columns x 6525 lines over the whole sensor.

Everything being easier when divisions are limited to integer numbers with no remainders and 3x2160 (the number of lines for 4K) being 6480 lines, I suppose that Hasselblad actually reads 6480 lines, cropping minimally left and right. This would imply that, on these 6480 lines, either they read them all and average the signal by group of 3 lines (hard, huge bandwidth) or they skip 2 lines out of 3 (easier, but with artefacts).
 

Doug Kerr

Well-known member
Hi, Jerome,

Thanks for all that scoop and insight.

Still, I am puzzled by why the Leica S (Typ 007) is able to develop video at a higher resolution when running from only part of the sensor than when running from the whole sensor (understanding that in the latter case there must be some "cropping" to account for the video aspect ratio).

Is is a matter of processor load for doing the operations you discuss?

Thanks.

Best regards,

Doug
 
I'll take the example of the H6D, since this is the camera subject of the thread.

The H6D native resolution is 11600 × 8700, which is a 4:3 aspect ratio common in MF cameras.

The video aspect ratio is 16:9 and from published material, we understand that the video reads the full width of the sensor, skipping the top and bottom part for aspect ratio conversion. That would give a readout of 11600 columns x 6525 lines over the whole sensor.

Everything being easier when divisions are limited to integer numbers with no remainders and 3x2160 (the number of lines for 4K) being 6480 lines, I suppose that Hasselblad actually reads 6480 lines, cropping minimally left and right. This would imply that, on these 6480 lines, either they read them all and average the signal by group of 3 lines (hard, huge bandwidth) or they skip 2 lines out of 3 (easier, but with artefacts).

Actually it's much simpler than that... There are two kinds of 4k video, there is cinema 4k (4092 columns of pixels) or (TV) HD4K (3840 columns). H6D-100 uses 3x3840=11520 pixels by combining three columns to make one... (nine pixels used as one). There is no mathematical correction applied as to insert or ignore columns.

Leica S uses the same pixel size sensor as the SL (6μm pixel pitch) it then uses the same area as the 24x36mm sensor and does 4K video as if it was a 24mp 24x36mm sensor DSLR...
 
Top