Yahoo Groups archive

Digital BW, The Print

Index last updated: 2026-04-28 22:56 UTC

Thread

Digital/Film Dynamic Ranges (long)

Digital/Film Dynamic Ranges (long)

2005-12-12 by claudej1@aol.com

Ah yes, B&W in a color world! How shall we truly compare?

First of all, one needs to define "film." By way of color ouput 
capability on every desktop (monitor and printer) the digital capture 
world is processed to color (Bayer filtered monochrome capture except 
Foveon X3). So how do we compare that to "film?"

Do we compare to the dynamics of tri-filtered B&W film used in the Roy 
Rogers days with dye transfer prints? Since we view everything 
processed digitally, RAW or JPG, in POSITIVE color first, should we not 
compare it to the dynamics of TRANSPARENCY film (Velvia having the 
SHORTEST scale of all)? Or, should we compare to color negative film, 
since it outsold transparency or B&W film by several orders of 
magnitude? Or perhaps do we look only to INFRA RED film dynamic range 
(DR), since ALL solid state sensors are dismally bad at sensing UV and 
Blue and are WILDLY  sensitive (Quantum Effeciciency) to infra red, 
where the sensor DR is severy crippled by Cyan filtering in every 
digital camera in the world? How about duplicating tranparancy film 
with a Contrast index of 1, closely matching the linear gamma of 
sensors? Ot, as most people here, the VARIABLE dynamics of B&W film 
capture with PRE-PROCESS spectral response control via color filters 
and POST CAPTURE chemical/time controls?

Since the answer is, ALL of the above and NONE of the above, we can see 
that digital image capture can be CONTOLLED by the PHOTOGRPHER in such 
a way that you can serve all of those masters with a collection of 
different digital cameras that are optimized for different 
applications. We never owned just ONE format for film, and used 
different emulsions for different things.

So why the "double standard" for digital capture where only ONE camera 
is expected to do it all with a single click?

Film cameras were nothing more than display pieces when one stopped 
making the "virtual lease payment" called film. No film, no 
photography. Now with "re-usable film" we have reduced the "cost per 
click" down to just pennies here v.s dollars. Economics rule the day. 
It's in the history books.

So what's the real problem?? the PHOTOGRAPHER's unwillingness or 
incapability to make it work. Whether it be time, attitude, money, 
technical aptiude.

As to the D70 issue, my tests showed that the Sony sensors in the Nikon 
cameras like the D1X, D1H, etc. inherently blew out highlights worse 
than others, so that's clearly a worse case scenario comparison to 
digital capture. Color neg. film with a Contrast Index of less than 0.5 
(extreme luminance compression) can hold about 14 stops worth of linear 
information with a short toe and a shoulder that doesn't kick in until 
color crossover issues have reared their ugly heads. But linear 
spectral response doesn't matter to B&W shooter, or does it?

The DR argument is a pretty weak one since we are talking about seeing 
about 6 stops worth of range on PAPER under the best of condiditions. 
All we need is a good tripod, one stop bracketing, and multi-layers or 
High Dynamic Range files to do the necessary compression. Let's not 
forget dodging and burning. Or, for the lazy, processing multiple 
exposure shifting in RAW output to stretch the dynamics. Even after all 
that, we sit here and complain the 8-bits isn't smooth enough to contol 
gray ink levels averged out over 4-8 heads and an ABOLUTE VISUAL 
DISCRIMINATION QUANTIFICATION than can BARELY detect 100 gray tones 
under normal viewing conditions. The human eye is a wonderful 
comparator but a TERRIBLE quantifier. That's why we spectrophotometers 
and densitometers to sort it all out.

The only place left where digital capture ISN'T superior in every way, 
is in Large Format B&W with standard or specialty films. The emergence 
of a superior solution in that realm will occur in line with the 
economics of the demand.

As the character Morpheus in the movie "the Matrix" so aptly put it: "I 
believe it is not a matter of hope, it is simply a matter of time."

Claude Jodoin
Tech. Editor and Working Photographer
Rangefinder Mag.
Digital Out since 1986, Digital in since 1995.


<<The article in Luminous landscape (at a quick glance) seems to
highlight what in my opinion is the biggest drawback of digital
cameras -- ie. the limited dynamic range of the sensor. We can burn
down blown highlights (eg clouds) with negative film. In fact a C41
process b&w film gives us the ability to record detail in the
brightest highlights and deepest shadows. Blown highlights and
blocked shadows have no place in landscape photography and as such I
am not ready to make the digital leap.
I have 300 feet of Tech Pan in the freezer to use on lower contrast
subjects and teamed with my Zeiss lenses (35mm) and 5400dpi scanner I
have no worries about big enlargements. The only digicam files I have
handled came from the D70.>>

RE: [Digital BW] Digital/Film Dynamic Ranges

2005-12-12 by Paul Roark

> ...
> All we need is a good tripod, one stop bracketing, ...

Is there any digital camera that can automatically bracket exposure in a
burst mode that is so fast it could be done without a tripod?  The SLRs with
their flopping mirrors are probably not going to be likely candidates here,
but I notice some digital cameras can shot at least low res frames very
quickly for video.  It seems like a hand-held, fast-burst, auto-bracket mode
would have some real value given the limited dynamic range of the cameras.

Paul
www.PaulRoark.com

Re: [Digital BW] Digital/Film Dynamic Ranges

2005-12-12 by Ernst Dinkla

Paul Roark wrote:
> 
>> ...
>> All we need is a good tripod, one stop bracketing, ...
> 
> Is there any digital camera that can automatically bracket exposure in a
> burst mode that is so fast it could be done without a tripod?  The SLRs with
> their flopping mirrors are probably not going to be likely candidates here,
> but I notice some digital cameras can shot at least low res frames very
> quickly for video.  It seems like a hand-held, fast-burst, auto-bracket mode
> would have some real value given the limited dynamic range of the cameras.
> 
> Paul
> www.PaulRoark.com 

The Sony Cybershot DSC-R1 should be the best option but 
doesn't get above 3 frames per second, but no raw mode then 
and the camera has no anti-shake. I think you will have to 
wait a bit longer till the Sony KM marriage is there.

Better dynamic range in sensors is more promising and may not 
be that long away:

http://www.rochester.edu/news/show.php?id=2367

                    --
           Ernst Dinkla


www.pigment-print.com
(         unvollendet         )

Re: [Digital BW] Digital/Film Dynamic Ranges

2005-12-12 by Mark Savoia

Did you download the high res image? Is that a real sample of what  
the chip can do or is it just the storytellers shot? It is very noisy.
Mark

On Dec 12, 2005, at 11:35 AM, Ernst Dinkla wrote:

> Better dynamic range in sensors is more promising and may not
> be that long away:
>
> http://www.rochester.edu/news/show.php?id=2367
>
>                     --
>            Ernst Dinkla



[Non-text portions of this message have been removed]

Re: [Digital BW] Digital/Film Dynamic Ranges

2005-12-12 by Ernst Dinkla

Mark Savoia wrote:
> Did you download the high res image? Is that a real sample of what  
> the chip can do or is it just the storytellers shot? It is very noisy.
> Mark
> 
> On Dec 12, 2005, at 11:35 AM, Ernst Dinkla wrote:
> 
>> Better dynamic range in sensors is more promising and may not
>> be that long away:
>>
>> http://www.rochester.edu/news/show.php?id=2367
>>
>>                     --
>>            Ernst Dinkla

Mark,

Don't know. Still downloading that image :-) Doubt they put a 
file like that on the page without further information how it 
was made.

There's an interesting thread on the news forum of dpreview on 
this subject (but sadly enough a non topic thread running 
through it). An excerpt of that thread:

Just Looking wrote:
 > Pixim has been doing in-pixel A-to-D conversion for years 
using a
 > Delta-Sigma scheme. How is this different?

Different company, different packaging. Looks like Pixim is 
actually further along, because they already have commercial 
products.

Other than that - very similar, except for this modified 
spacing of pixels that's supposed to favor jpeg conversion (I 
wonder if it affects the cost of the Bayer and AA filters for 
this sensor), so good find there.

Pixim's products advertise "512X Dynamic Range" - that's 9 
stops, the sensor is made for video cameras:
http://www.prosecuritys.com/gsdct-ps615xt.html

Actually, they recently issued a press release about a new 
chipset a few months ago - but again, this is a video chip, 
doesn't target still photography:
http://www.pixim.com/html/press_release_050927.htm

For those who (like me) needed more information about 
Delta-Sigma conversion, here's a Wikipedia article (it gets 
pretty involved though):
http://en.wikipedia.org/wiki/Delta-sigma_modulation

"The principle of the sigma-delta architecture is to make 
rough evaluations of the signal, to measure the error, 
integrate it and then compensate for that error."
It also has a link to java-based simulation:

http://www.analog.com/... 
.../interactiveTools/sdtutorial/sdtutorial.html

End of quote.

So there has been experience with a similar system already.

Ernst

-- 

                    --
           Ernst Dinkla


www.pigment-print.com
(         unvollendet         )

Re: [Digital BW] Digital/Film Dynamic Ranges

2005-12-12 by Ernst Dinkla

Mark Savoia wrote:
> Did you download the high res image? Is that a real sample of what  
> the chip can do or is it just the storytellers shot? It is very noisy.
> Mark

Size is 2x3. Must be an APS or FF sensor take or scanned 35 
mm, no compact has that size. Doubt they use that size for a 
prototype, Kodak would favor a 4/3 format from RIT isn't it 
:-)  More likely that there first prototype chip isn't more 
than 1 MP.

Storytellers shot made with a Sony R1 at 1600 ASA, size fits, 
noise too.
By accident there's a loop in my messages :-)


Ernst

                    --
           Ernst Dinkla


www.pigment-print.com
(         unvollendet         )

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.