[Leica] Interesting digital DMR info not seen elsewhere

[imagist at imagist.cnc.net (George Lottermoser)]

I've included a couple of posts from the "ProRental" email list (a list
devoted to the hi-end medium format backs). The discussion began around
the announcement of 30+ megapixel backs. To see the mention of a Leica
on that list is rather surprising. But even more surprising is the
mention of micro lens technology incorporated into the DMR as well as
the mention of the ROM information relating to the micro lenses. It
would seem that this technology would up the ante and image quality.

------------------------------------
Message: 1 Date: Thu, 29 Sep 2005 16:22:28 -0400 From: "John O'Donnell"
<john@prographics.com> Subject: [Prorental] The new 31 and 39 mp
sensors. To: <prorental@prorental.com> Message-ID:
<BF61C544.4241%john@prographics.com> Content-Type: text/plain;
charset="ISO-8859-1"

To Mark and all the list members.

I9d like to shed some light on the subject of the latest new 31 and 39
megapixel sensor arrays. There are some technical obstacles specific to
the new technology I would like to make you all aware of. I hope to
accomplish this before your eyes glaze over. Although I am a long time
Hasselblad user and a Hasselblad full-line dealer, what follows is my
personal, researched opinion and not that of Hasselblad or any other
company. I9m just a digital photographer today, speaking out.

Recently PhaseOne announced upcoming products using these new sensors.
In fact, they offered a timetable for the introduction of these new
products. Unquestionably, every digital back manufacturer is developing
products around these new sensors. My objection is that announcing new
products and the timetables for their introduction at this infant stage
of the new sensor development is inherently guesswork. The pressure to
engineer and introduce products early to meet the optimistic promises of
the marketing department has given us some recent disappointments like
the Kodak DCS 14n. Even when a company does the right thing and holds
back pre-announced products until they are ready, their image suffers
from the delays. I9m not sure exactly how long we9ve been waiting for
the Leaf/Mamiya integrated solution. I9ve even forgotten the name of it.
To Leaf9s and Mamiya9s credit, however, they didn9t release a turkey.

While high dynamic range is mandatory for the highest quality digital
images, resolution is, to varying extents, secondary. An analogy to this
has always existed in the film shooters world. Medium format images had
a broader dynamic range than 35mm. Even when it9s larger image was
unnecessary, the snap and clarity of medium format was always welcome.

Further, neither of these sensors are a full 6x4.5 cm in size. It9s
unlikely we9ll see one any time soon. The current limiting factors on
sensor size are:

a. The size of the individual sensor chips which must be stitched
together to make a full sensor array. Increasing the number of these to
make a larger sensor would raise the price to unmarketable levels.

b. The difficulty in mounting and silicon coating larger assemblies at a
consistent enough rate to make them feasible at all.

c. The performance of lenses near the edges and corners of larger chips.
The larger the physical dimensions of the chip the worse the image
quality at the edges, all other things being equal.

It9s far more likely that we will see newly designed lenses optimized
for digital sensors and higher magnification viewfinders, which will
normalize the viewing and coverage issues of the sensors which do exist.

Instead, the resolution increase is accomplished by making the pixel
size smaller. The smaller the pixel, the less light it can hold. Less
light means lower dynamic range. 22 million 9 micron pixels holds more
image data than 31 or 35 million 6.8 micron pixels of the same type.
Imagine the capacity of 22 million quart cans versus 35 million pint
cans. The difference in the volume of the cans is not proportional to
the difference in their diameter, but the square of the difference. The
larger cans are deeper as well. Resolution is the diameter, bit depth is
the volume.

The two new sensors have some new technology and some new problems,
which are yet to be solved. The 31mp sensor employs microlens
technology. A tiny light gathering lens designed to increase the
sensitivity of the pixel by funneling light into it, enabling higher ISO
ratings, caps each pixel. The microlens exaggerates the usual falloff
and color shift between the center and edges of the chip. This is worse
on wide-angle lenses and with view camera movements. To date, only
Hasselblad has successfully managed the microlens sensor in a
professional level product, the digital back they make for the Leica R8.
While the specification for the chip stated that this behavior would
occur with wide-angle lenses, only, Hasselblad discovered that it
occurred with all lenses. To correct these aberrations, Hasselblad`
mapped out the characteristics of each lens in the Leica R8 system at
every aperture and light level and included this information in the
firmware of the back. The camera would send the lens and exposure data
to the back, which would in turn correct the image In other words, the
corrections had to be integrated at the system level. Hasselblad has
already added the capability for this type of lens to camera to back
communication in its latest digital platform, the H2. Previous
strategies of mechanically adapting backs to fit different cameras may
no longer provide an optimal solution with these higher technology
sensors.

The 39mp sensor has no microlenses but is designed with dual data
outputs. This provides a higher data transfer rate for faster recycling.
In testing however, any minute difference between the two output busses
results in a visible difference between the upper and lower half of the
image. This must be controlled continually with extreme accuracy and is
likely to be a result of a hardware/software negotiation within the
system. Once more, tighter integration is the key to making the new
technology operate at the highest level of image quality. Here too,
Hasselblad alone is addressing this in its H2 platform.

The current 22 mp sensor will continue to be the quality benchmark until
all of the open issues with the new sensors can be ironed out and this
will take some brand new solutions. It remains to be seen how or even
whether these technological hurdles can be overcome. Simply outputting
12 bits (for instance) of data into a 16bit space does not increase the
amount of image data, only the file size. This fuzzy math works in the
DSLR and prosumer products but will not wash in the high-end digital
imaging market.

My point here is that, while all of the digital back manufacturers are
hard at work testing with prototype sensors and designing new backs,
announcing a product release before these fundamental technological
problems have been solved is disingenuous at best. And marketing
future-ware under the premise of service to the user base is, any way
you slice it, baloney. Only when the production sensors can be tested
will anyone know whether the technology is capable of consistently high
image quality we demand as commercial photographers. And clearly, the
new sensors will require a tighter integration between the lens, camera,
back and software to live up to its design parameters. The 3one back
fits all2 solutions will be unable to deliver the results that the
integrated system will.

I think as photographers we should all be from Missouri when it comes to
this level of investment in our equipment. The only defense we have
against being unwitting beta testers of buggy new technology is to test
shoot equipment before we buy it. Further, by demanding to be treated
with respect by companies wishing to win our business, the business gets
better for us all.

Thank you all for reading this and I look forward to your considered
responses.

John O'Donnell Technical Sales & Training - Digital Imaging Systems
Professional Graphics Systems & Services, Inc. 3 West Main Street
Elmsford, NY 10701 (914) 345-3033 x34 cel (914) 224-8642
john@prographics.com
----------------------------

On Sep 30, 2005, at 8:22 AM, John O'Donnell wrote: > > I?d like to shed
some light on the subject of the latest new 31 and 39 > megapixel sensor
arrays.

Hello all I wanted to add to John's excellent post on the new chips.

It's worth noting that the H2 also passes focus data to the back because
in order to know how to correct CA fall off etc you need to know focal
length (ie on the zoom) apeture AND focus distance. I think you are
right in saying that Phase will have a lot of problems introducing the
31MP due to the microlenses and they do not have access to the lens data
as Hasselblad backs do. Also as you rightly point out the time spent
with the Lieca has helped them problem solve.

The H2 launch has been very quiet and I think many people have been a
bit underwhelmed by the changes but as the new chips come on line they
will really see the benefits of the H2 (getting the lens data) and the
new DSP in the backs. Regards

Nick Tresidder Photography http://www.nick-t.com 649 379 3778 6421 972
111
-------------------------------

regards, George