Film versus DSLR dynamic range

Started Jul 24, 2007 | Discussions
The Davinator
The Davinator Forum Pro • Posts: 22,383
Re: Naw....

dt8155 wrote:

Dave Luttmann wrote:

....useless is claiming digital's superiority over film....when you...

...are dealing with fanboys who think photography is a game and their
team needs to win.

...are dealing with a web experts who doesn't take pictures.....just visits forums pronouncing everyone who disagrees with him a fanboy. Found a picture for us dt1855, or are you just following people to different threads? I at least will hear out Pixsurgeon because it's obvious he has done some testing. You on the otherhand are simply nothing more than a web expert troll.

 The Davinator's gear list:The Davinator's gear list
Canon EOS D30 Canon EOS 10D Nikon D2X Fujifilm X-Pro1 Fujifilm X-T1 +17 more
Benedict Slotte Senior Member • Posts: 1,165
Re: LoOoOoL!!!

Harry Lime wrote:

Weren't you banned from this forum at one point for being an annoyance?

That's also what I remember a couple of people saying about him in the past. But I guess some people just never, never learn. Or learn maybe, but can never change their style the slightest bit. Or maybe the real Ferenc is a nice guy (at least I want to believe that!) but this PixSurgeon character that he hides behind is the opposite. Or maybe it's all a bad joke. I don't really know...

-- hide signature --

B. Slotte
Turku, Finland
http://bslotte.smugmug.com
See profile for equipment

smhagger Senior Member • Posts: 1,418
Question Roman: Calibrating

Roman,

As you know, I have the same Sekonic meter as you do. I haven't calibrated the meter / camera relationship yet but need to do so (to my D2Hs, D200 and S5 Pro). How did you go about doing this? Are there instructions in the Sekonic manual that I need to review for doing this or can you point me in the right direction via weblink?

What did you use for a test target, etc.

I would also like to know how my suspicions of DR ranking between the three cameras stack up.

Sean

PIXSurgeon Veteran Member • Posts: 4,009
Oh yes!...

Harry Lime wrote:

Weren't you banned from this forum at one point for being an annoyance?

...Of course: annoyance to non-sensers, to blah-blahers, and to folks like have NOTHING valulable to add in a discussion.... like you, maybe?

8-))

-- hide signature --

TIP: If you do not like this post, simply press the 'COMPLAINT' button. Mommy/Daddy are just one click away.

Truman Prevatt
Truman Prevatt Veteran Member • Posts: 9,156
Re: ..... and when I think about it....

The dye transfer process of color film, slide or negative prohibits one from play with the exposure and/or development much. However, with B&W film there are many tools one can use to capture scenes with high dynamic range. Film is one. Developer is the second. Developer dilution is another. Developer agitation is another. Then there are two part developments, development using a water bath (put the film in the developer for 15 seconds and in water for 30 seconds. Repeat so many times.

Then there is you enlarging equipment. A cold light head will do much better on details in the highlights than a condenser head.

All will impact the dynamic range of the film and tonal range. One of the most difficult scenes is snow in the sunshine. I'm not even sure I would try it with a digital - maybe the S5.

But with good old film it can be done.

I even used a red filter on this shot to drop the tone of the trees.

Kodak PanX 120 film. Development in Rodinal dilution one part Rodinal to 100 parts of water. Agitation - 10 seconds every minute.

With B&W film the zone system is based on the concept of "expose for the shadow detail" and "develop for the highlight detail." With digital it's gets down to expose for the highlights and pray for the shadow detail.

olstrup wrote:

.... I scanned some old Tri-X negatives and found that I could reveal
details in the sky that was all "blown out" when the negative was
printed in the enlarger, so the may be even more range hidden in
those films than I ever knew. And yes, I know, those datil might also
have been brought out if I had taken the trouble to "burn in" the sky
while copying.

-- hide signature --
 Truman Prevatt's gear list:Truman Prevatt's gear list
Fujifilm X-Pro2 Fujifilm X-H1 Fujifilm XF 18mm F2 R Fujifilm XF 35mm F1.4 R Fujifilm XF 55-200mm F3.5-4.8 R LM OIS +9 more
uk102 Senior Member • Posts: 1,286
Re: Question Roman: Calibrating

The meter will not produce a 1 to 1 reationship as DSLR bias the meter readings. Plus the fact the are meter direct verse reflected light. I sekonic always seams to be correct!!

smhagger wrote:

Roman,

As you know, I have the same Sekonic meter as you do. I haven't
calibrated the meter / camera relationship yet but need to do so (to
my D2Hs, D200 and S5 Pro).

RomanJohnston
RomanJohnston Forum Pro • Posts: 18,827
Huh?!?

The process is to allow the meter to know the DR of your camera and to let you know when a range of readings exceed your cameras DR.

Your cameras meter doesnt even come into play.

Roman
--

The Law Of Attraction is ALWAYS working. Your only choice is whether you drive 'it'...or 'it' drives you.
-Me
http://www.pbase.com/romansphotos/

Steve Bingham
Steve Bingham Forum Pro • Posts: 26,469
Re: ..... and when I think about it....

Truman, having spent a lifetime playing with b&w film and development I certainly agree that there was a lot of latitude available. However, Pan-X was noted for it's lack of it (as opposed to Tri-X) - even with water bath development.

Digital "development" is just as much an art form. By using a simple double raw conversion, quick luminance masking (2 seconds), and some judicious partial mask erasing, level adjustment of both forms as well as the mask, it is possible to get 10-11 stops with my D200 - in less than a minute or two! It's technique, just like film selection and development methods. With 1 stop available for highlights and 3 stops available in the shadows, it is quite easy to dig down for shadow detail if your conversions are 16 bit. Most digital users, have no idea what I am talking about, but some fine art pros practice techniques like this every day. It is vital to what they do.

Even more important is getting all this information on print. Without an extensive knowledge of levels, curves, dodging, burning, masking, etc the average printer is lucky to see 5 stops on his print. Those who send their film into the corner drug store would be in this group. If we are talking about slide film (chromes) than the DR ranged from 7-8 stops. In any case the average amateur never saw more - as is the case with digital today.

It is amazing that this thread should generate so much debate from so many who know so little. And I am not talking about you, Truman.

By the way, the dye transfer process was great but INCREDIBLY expensive and time consuming. Even Cibachrome prints required contrast masking which doubled the time and cost. God I am glad those days are gone! Both techniques were noted for their archival properties - and both are now easily exceeded in that category by both HP and Epson.

Truman Prevatt wrote:
The dye transfer process of color film, slide or negative prohibits
one from play with the exposure and/or development much. However,
with B&W film there are many tools one can use to capture scenes with
high dynamic range. Film is one. Developer is the second. Developer
dilution is another. Developer agitation is another. Then there are
two part developments, development using a water bath (put the film
in the developer for 15 seconds and in water for 30 seconds. Repeat
so many times.

Then there is you enlarging equipment. A cold light head will do much
better on details in the highlights than a condenser head.

All will impact the dynamic range of the film and tonal range. One of
the most difficult scenes is snow in the sunshine. I'm not even sure
I would try it with a digital - maybe the S5.

But with good old film it can be done.

I even used a red filter on this shot to drop the tone of the trees.

Kodak PanX 120 film. Development in Rodinal dilution one part Rodinal
to 100 parts of water. Agitation - 10 seconds every minute.

With B&W film the zone system is based on the concept of "expose for
the shadow detail" and "develop for the highlight detail." With
digital it's gets down to expose for the highlights and pray for the
shadow detail.

olstrup wrote:

.... I scanned some old Tri-X negatives and found that I could reveal
details in the sky that was all "blown out" when the negative was
printed in the enlarger, so the may be even more range hidden in
those films than I ever knew. And yes, I know, those datil might also
have been brought out if I had taken the trouble to "burn in" the sky
while copying.

 Steve Bingham's gear list:Steve Bingham's gear list
Nikon D810 Nikon D7200 Nikon AF Nikkor 85mm f/1.8D Fujifilm XF 14mm F2.8 R Nikon AF-S DX Nikkor 18-140mm F3.5-5.6G ED VR +24 more
Jay Turberville Forum Pro • Posts: 12,917
Re: Bit depth and dynamic range...

Sure. It is possible. But show me one manufacturer who is doing
that when they record their raw data.

all CMOS sensors are able to do that during pre-amplification before
AD conversion at the price of loosing some resolution (-> an
intelligent form of "binning"). I know that recent Canon CMOS sensors
have this property, and Kodak has recently filled a patent with a
similar idea:
http://www.wipo.int/pctdb/en/wo.jsp?IA=WO2006130518&wo=2006130518&DISPLAY=DESC

There is no doubt about the possibility. The question is the existence of sensors that actually do it.

There is some slight non-linearity in real life in the darkest
shadows which is probably due to lens flare. That probably explains
why we can actually reach and even exceed a tiny bit the theoretical
DR limits. You can see that slight non-linearity in this Imatest
graph. It is interesting on its own merits, but I don't think it has
much relevance to practical photography with most cameras. Note the
otherwise linear response.

You suggest that one cannot exceed 12 stops because of the linear
response. This is false.

Where the heck do you get that from what I wrote? Just look at my first sentences that you even quoted in this response (at the top of this thread). I was very clear and speciific multiple times that you could get more than 12 stops with 12 bits. Please re-read my post and pay attention to what I actually wrote this time.

Mario is right on this point.

No he isn't, because the 12 levels quite typically do map very directly to 11 stops of DR because most sensors (at least every one I've ever tested) are nearly perfectly linear in their responses. The correct assessment is that the 11 mathematical levels may or may not map directly to dynamic range. It is possible to put more than 12 stops of DR into 12-bits. But Luttman is correct in his analyis because nobody seems to actually be doing that. When they actually map the tonal response non-linearly, then the analyis won't hold.

You can have
an output very close to linear but some non-linearities applied
before AD conversion.

Sure. You could impose a complex "characterisic curve" if you'd like. But please just show me the examples of some actual sensors with these non-linearities. All the tests I've done show very linear responses with some very slight non-linearities in the darker tones (that I believe are simply the results of very low levels of lens flare.) These slight non-linearities occur in tones where it isn't really helping anything practical.

Read the graph of the 1DMkII: the maximum value
is not achieved at the lowest (basal) ISO value but one-two stops
higher.

Show me a response curve from a 1DMII that has any significant non-linearities is all I'm asking.

Please note that nobody is claiming more than 12 stops for the 1DMII even though Clark says that the sensor itself can record well more than 12 stops of DR. So I doubt that there are any nonlinearites that are significant enough to increase DR much. Canon probably isn't concerned since the noise at those limits is very high, just as it is very high in sensors with smaller photodiodes at 11-12 stops of DR.

Furthermore, his graph of the 1DMII's response at ISO 50 shows that it actually is very linear.

http://www.clarkvision.com/imagedetail/evaluation-1d2/index.html

So, like I said, while it is certainly possible to map the tonal values in a non-linear fashion in order to store more than 12 stops of DR in a 12-bit file (the same way we store more DR than 8 stops in an 8-bit JPEG), I'm just not seeing anybody doing it. While that doesn't mean that nobody is doing it, it does mean that it isn't common practice. We have tests of Canon, Kodak, Sony and Matsushita sensors and they all show nearly perfectly linear responses.

What is possible and what is actually done are not the same thing.

-- hide signature --
Chris101 Veteran Member • Posts: 4,945
Apples are superior to Oranges! ;)

this is #5 in the series, right? Just trying to keep up! Thanks!

-- hide signature --
mcd3 Regular Member • Posts: 225
Re: I am as far from film bashing as it gets

Go shoot some color negative film in the bright sun and compare it to a digital image.

If you are still convinced that film has less range, you don't know what your looking at. Digital latitude is a joke in comparison. In fact, go shoot some 5 stop brackets with both and then tell me which has more range.

PIXSurgeon Veteran Member • Posts: 4,009
You are correct...

uk102 wrote:

These are very pretty graphs that show DSLR’s as having huge DR.

...They have LARGE DR ( IN F-COLOR , btw), and the better and cleaner the cam, the more of it is USABLE. On EOS 1D-class cams, 5D, 30D (at lower ISOs) and S3/S5, there is plenty of it, being the EOS-1D and 5D series the most efficient-per-stop (that is, the cleanest and most "pushable", stop-by-stop) and the S3/S5 as the most "elastic" of the bunch (incredible latitude at lower ISOs).

What
is puzzling me is the following.

On my film (not slide) camera I can take a photo and the sky looks
lovely and I can have sky in my all photo’s without thinking. I meter
for the foreground and the foreground looks lovely so does the sky.

On my DSLR doing the same with .jpg and yes the sky nearly always
blows out if the foreground is correctly exposed. Therefore I use RAW
and thing are better but not perfect.

...That is because your equipment is trailing, and/or your skills, as well.

Watch the skies "fall down" 8-):

(non optimized) http://www.pbase.com/feharmat/image/50086720/original
(optimized) http://www.pbase.com/feharmat/image/50090435/original
(optmized) http://www.pbase.com/feharmat/image/50090436/original

And this was EONs ago. Today, I can pull out better contrast, more crispness, and overall better tonality.

From what I have found slide film is about 5 stop,
colour film 11 stop (depend on brand) and some B&W up to 19 stops.

...Just a quick question... HOW did you measure this?

8-)

-- hide signature --

TIP: If you do not like this post, simply press the 'COMPLAINT' button. Mommy/Daddy are just one click away.

xrdbear Veteran Member • Posts: 3,938
Re: And...

uk102 wrote:

On my DSLR doing the same with .jpg and yes the sky nearly always
blows out if the foreground is correctly exposed.

I simply don't understand this. I have no trouble getting sky and foreground correctly exposed together except on featureless overcast days. Do you have different sky where you are?

-- hide signature --

Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/
Pbase gallery Pictures from Isle of Skye
http://www.pbase.com/xrdbear

mcd3 Regular Member • Posts: 225
Re: And...

xrdbear wrote:

uk102 wrote:

On my DSLR doing the same with .jpg and yes the sky nearly always
blows out if the foreground is correctly exposed.

I simply don't understand this. I have no trouble getting sky and
foreground correctly exposed together except on featureless overcast
days. Do you have different sky where you are?

You have trouble with exposure on overcast days?

sergey1968 Senior Member • Posts: 1,068
I did!

160S - in bright midday Colorado sun.

And I was easily able to expose 5D for highlights and then push shadows to get all the details film had - and more. Sure my film skills are rudimentary (not that digital are great but I feel I've got it right. The most challenging thing for digital is something like bright neon lights - you need to expose digital FEW stops lower than negative film to get color right - but clean shadows allow to get all details and textures back. It may be tedious and requires practice to do it right - but it works.
--
Sergey
http://www.pbase.com/sergeyushakov/
http://www.photo.net/photos/SergeyUshakov

PIXSurgeon Veteran Member • Posts: 4,009
You should know by now (SAMPLES)...

mcd3 wrote:

xrdbear wrote:

I simply don't understand this. I have no trouble getting sky and
foreground correctly exposed together except on featureless overcast
days. Do you have different sky where you are?

You have trouble with exposure on overcast days?

...Or aren't you sufficiently experienced, already, to anticipate such potential troubles and their real root?

Since folks love to write , I will choose (once again), to ILLUSTRATE, in DETAIL:

http://www.pbase.com/feharmat/image/82888247/original
http://www.pbase.com/feharmat/image/82888248/original

That's overcast, mixed with sunlight, then shadows, and you name it. A real headache, from any angle... yet perfectly manageable with digital (accross 10-11 f/stops, approx).

Enjoy!

-- hide signature --

TIP: If you do not like this post, simply press the 'COMPLAINT' button. Mommy/Daddy are just one click away.

xrdbear Veteran Member • Posts: 3,938
Re: And...

mcd3 wrote:

xrdbear wrote:

uk102 wrote:

On my DSLR doing the same with .jpg and yes the sky nearly always
blows out if the foreground is correctly exposed.

I simply don't understand this. I have no trouble getting sky and
foreground correctly exposed together except on featureless overcast
days. Do you have different sky where you are?

You have trouble with exposure on overcast days?

Only to the extent of getting detail in the sky while the foreground looks right without doing any PP. The detail is still there and can be successfully printed with post processing. No more than you would have to do with film to get it onto paper.

-- hide signature --

Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/
Pbase gallery Pictures from Isle of Skye
http://www.pbase.com/xrdbear

xrdbear Veteran Member • Posts: 3,938
Re: You should know by now (SAMPLES)...

PIXSurgeon wrote:

mcd3 wrote:

xrdbear wrote:

I simply don't understand this. I have no trouble getting sky and
foreground correctly exposed together except on featureless overcast
days. Do you have different sky where you are?

You have trouble with exposure on overcast days?

...Or aren't you sufficiently experienced, already, to anticipate
such potential troubles and their real root?

Was that directed at me? I can successfully post process all my overcast shots to produce satisfactory results. I was suggesting that there are some occasions when both sky and foreground can't be accommodated without PP or grads. Not meaning to be contentious but the two examples you just offered were a breeze with the foreground bathed in sunlight. We are not often that fortunate here in the highlands.

-- hide signature --

Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/
Pbase gallery Pictures from Isle of Skye
http://www.pbase.com/xrdbear

The Davinator
The Davinator Forum Pro • Posts: 22,383
Re: I did!

sergey1968 wrote:

160S - in bright midday Colorado sun.
And I was easily able to expose 5D for highlights and then push
shadows to get all the details film had - and more. Sure my film
skills are rudimentary (not that digital are great but I feel I've
got it right. The most challenging thing for digital is something
like bright neon lights - you need to expose digital FEW stops lower
than negative film to get color right - but clean shadows allow to
get all details and textures back. It may be tedious and requires
practice to do it right - but it works.
--
Sergey
http://www.pbase.com/sergeyushakov/
http://www.photo.net/photos/SergeyUshakov

Odd.....when I expose film for the highlights, I can dig deeper than my 1Ds Mk2 can pull from the shadows.....but a couple of stops.

What scanner were you using?

 The Davinator's gear list:The Davinator's gear list
Canon EOS D30 Canon EOS 10D Nikon D2X Fujifilm X-Pro1 Fujifilm X-T1 +17 more
Mario Senior Member • Posts: 2,648
All I was trying to say...

is that 12 bits is not what is limiting the dynamic range. As a matter of fact all images we view online are 8-bit per channel images, even the HDR merged ones that cram more than 12 stops of DR into 8-bit (showing detail in both shadows and highlights) at the expense of mid tones.

So once more, bit depth is not a limiting factor to DR. If you have a scene that has shadow detail and highlight detail 15 stops apart, it is possible to create JPEG image of the scene that displays both (at the expense of low contrast in the mid tones) perfectly fine, and JPEG is only 8 bits.

I was not even talking about the acquisition or sensors. Of course there may be limitations there. At he physical level, sensor is indeed a liner counting device that counts how many photos it caught, so to speak. The higher the number of photons caught, higher electrical current generated by the photosite. And yes, if twice as many are collected by the site, current will be twice as strong (i.e. relationship is liner)

But, the next step is digitalization of that current to create a RAW data. And this is where some non linearity can be applied. I don't know if anyone does it or not, but it is certainly possible.

-- hide signature --

Mario

Keyboard shortcuts:
FForum MMy threads