Reading mode: Light Dark

F-Stop Range of Digital Photography?

Go to last post
T

T Blair

Well-known member
Messages
212
Reaction score
0
Location
LA, US
Another in a series of technical questions:

I'm not sure if I read or heard that film has about an 11 stop range; meaning that it can differentiate contrast accross that range. Not being sure if this is even a correct statement; do you know what the range is for digital photography? My practical use is for spot metering in long-exposure photography.

--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
 
I've never done any tests on it, but it's on par with the dynamic range of slide film --- very limited.
T Blair said:
Another in a series of technical questions:

I'm not sure if I read or heard that film has about an 11 stop
range; meaning that it can differentiate contrast accross that
range. Not being sure if this is even a correct statement; do you
know what the range is for digital photography? My practical use
is for spot metering in long-exposure photography.

--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
Click to expand...
 
A quick look at the LCD display shows 5 stops. Add 2 to that for the Raw exposure compensation capability and you have about 7. That's probably about it.

Rich
Ben Horne said:
T Blair said:
Another in a series of technical questions:

I'm not sure if I read or heard that film has about an 11 stop
range; meaning that it can differentiate contrast accross that
range. Not being sure if this is even a correct statement; do you
know what the range is for digital photography? My practical use
is for spot metering in long-exposure photography.

--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
Click to expand...
Click to expand...
 
T Blair said:
Another in a series of technical questions:

I'm not sure if I read or heard that film has about an 11 stop
range; meaning that it can differentiate contrast accross that
range. Not being sure if this is even a correct statement; do you
know what the range is for digital photography? My practical use
is for spot metering in long-exposure photography.

--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
Click to expand...
--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
 
That's the great thing about digital, just keep adding more bits and you've added dynamic range. A 16 bit/channel Tif has a whole lot of dynamic range. A 64 bit/color tif would have even more (assuming a constant color space)

Now, you probably actually want the dynamic range of a sensor of some particular camera, or perhaps of some format (eg. 8 bit jpeg). Which sensor/camera are you asking about?

Jason
T Blair said:
Another in a series of technical questions:

I'm not sure if I read or heard that film has about an 11 stop
range; meaning that it can differentiate contrast accross that
range. Not being sure if this is even a correct statement; do you
know what the range is for digital photography? My practical use
is for spot metering in long-exposure photography.

--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
Click to expand...
 
As long as you know what you're doing in photoshop and you had the ability to take several identical pictures at different apertures (not shooting race cars or something) you can go nuts.
No need for ND filters here.
 
T Blair said:
Another in a series of technical questions:

I'm not sure if I read or heard that film has about an 11 stop
range; meaning that it can differentiate contrast accross that....
Click to expand...
It's important to understand that each stop means a doubling or halving of light.

Film has about 7 stops, not 11, though print film has some further exposure latitude available at print time that slide film doesn't.

7 stops would therefore be variance of 2^7, or 128 times brightest areas compared to darkest areas. My math may be off by a digit here or there, I'm just trying to lay out the principle. If someone wants to suplly the exact right numbers, that's fine.

re other post: The stops shown in the viewfinder have nothing to do with digital's range, they merely indicate how far above or below ideal exposure one is.

re other other post: Adding more bit depth doesn't add dynamic range, it adds steps within the range available from the CCD or CMOS.

I've been told that digital has less dynamic range than film, though you couldn't prove that by me, it seems like more.

The CCD or CMOS and their attendant circuitry are the limiting factor in digital's dynamic range.

What I do know is that digital maintains vastly (IMO) more shadow detail than film, allowing wide ranging adjustments in Photoshop that just aren't there with scans from film.

The real advantage, if you're going to be working with an image in your computer, is the lack of scanning step (it's usually the weak link), as well as the lack of film grain.
 
Bob Graham rvg said:
re other other post: Adding more bit depth doesn't add dynamic
range, it adds steps within the range available from the CCD or
CMOS.
Click to expand...
Actually it does.

Say you have an 8 bit system, and you have an object which is illuminated with an average lighting of 1 bit above black (eg. 1) Say it's a grey object, then the representation would be 1 Red, 1 Blue, 1 Green. Now, say you want to take a little blue out of it, to say adjust the white balance. But wait, you don't have the dynamic range to do that, because the only graduation below 1 is 0, and now you've got no blue.

However, if you had 8 more bits, you'd have 256 levels between 0 and 1, and now you can take just a bit of blue out. Said in a more on-topic fashion, you can now resolve an object in a wider range of brightnesses (darker). You can "pull more detail" out of the shadows as they say.

Think of the 1 dimensional equivalent to 2d images, sound. Add more bits, you add more dynamic range.

Jason
 
Jason Hutchinson said:
Bob Graham rvg said:
re other other post: Adding more bit depth doesn't add dynamic
range, it adds steps within the range available from the CCD or
CMOS.
Click to expand...
Actually it does.
Click to expand...
No that’s not correct. Bit depth only defines the size of the steps (how many) between totally black and totally white – if you will the shades of gray. The sensor will have a Dmin and a Dmax value that ends up setting the total dynamic range in absolute terms. Any light value less than Dmin will be registered as black by the sensor and any value greater than Dmax will be registered as white.

Think of it this way. Ground floor is black and the balcony is white. The height of the balcony above the ground floor represents the dynamic range,, the steps leading from the ground floor up to the balcony represent bit depth – the greater the bit depth the more steps there are.

Cheers,
Phred
 
Some B&W negative films may have the ability to dig out 11 stops but that really is not very practical.

Digital is a lot like slide film, like you say. A good image will have good tonality, perhaps at the expense of a broad dynamic range.

Here's the characteristic curve for the D30. D60 is most likely very close.
http://home.attbi.com/~ajpacheco/D30_Normal_Contrast.html

The linear portion of the curve contains the most tonality and contrast. Top and bottom demonstrate where Canon's software reduces contrast.
RichW said:
Rich
Ben Horne said:
T Blair said:
Another in a series of technical questions:

I'm not sure if I read or heard that film has about an 11 stop
range; meaning that it can differentiate contrast accross that
range. Not being sure if this is even a correct statement; do you
know what the range is for digital photography? My practical use
is for spot metering in long-exposure photography.

--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
Click to expand...
Click to expand...
Click to expand...
 
Phred said:
No that’s not correct. Bit depth only defines the size of
the steps (how many) between totally black and totally white
– if you will the shades of gray.
Click to expand...
Unfortunately, I don't have time to explain this fully. There are a lot of confusing things on this issue I am skipping over. You guys are talking about an input function to a camera, and I'm talking about the digital domain, the results (data). Basically we're talking about two different things.

Let me say first, that dynamic range has EVERYTHING to do with the number of valid bits. In fact, you can specifiy the dynamic range of a digital system in bits, as done here:

http://www.dpreview.com/reviews/canong2/page17.asp

Basically, 1 bit = 6dB of dynamic range

What you guys are talking about, is where the white and black points are mapped to by a camera systems input, which is VERY relavant to digital imaging, I agree. You could call it the "input dynamic rnage" I guess.

What I am talking about is the dynamic range of a digital image. Say an 8 bit tif file. This file has a limited dynamic range, and the ONLY way to add dynamic range capability to the format is to add more bits. That's it, because:

0,0,0 is defined as infinitely black
256,256,256 is defined as infinitely bright (again 8 bit system)

Dynamic range is defined as the ratio of the brightest to the darkest "signal" you can display.

Think of it this way. Say you've got a new camera that outputs only 3 bit tif files. The catch is, 0EV is mapped to 0,0,0 and 22EV is mapped to 4,4,4. So, do I have an image with 22 stops of dynamic range right? Well, not really, you've got an image with 4 bits of dynamic range, or 24dB. As you can imagine, a 4 bit image looks like junk. You captured data across 22 stops, but try pulling out shadow detail of a 4 bit image, which is what dynamic range is all about.

So, like I said, we're sort of talking about different things. But, once again my point is that dynamic range in a digital system has everything to do with bits.

jason
Member said:
Dmin and a Dmax value that ends up setting the total dynamic range
in absolute terms. Any light value less than Dmin will be
registered as black by the sensor and any value greater than Dmax
will be registered as white.

Think of it this way. Ground floor is black and the balcony is
white. The height of the balcony above the ground floor represents
the dynamic range,, the steps leading from the ground floor up to
the balcony represent bit depth – the greater the bit depth
the more steps there are.
Click to expand...
0,0,0 represents infinitely black
256,256,256 represents infinitely white (in an 8 bit system)
 
I appreciate all this information, some of it I get, some I don't. Practical application is what I am after. I was shooting in the LA Train Station yesterday and I found myself wondering if I had to choose to wash out the light fixtures to get the detail in the tiled walls. I estimated over a 7 stop difference and I knew that that couldn't be had on slide film. I remember the owner of Camera West in Monterey Ca telling me that some films have an 11 stop range. I had little time last night to experiement, but it seems like it may be worth the time to take notes on exposure range as I take pictures, then post some and see if we agree on the effective range of the D60.

As an aside, I find these technical discussions much more intereting than the "When is Canon coming out with an EOS3D camera?" ones.

Best to all,
--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
 
I think the relevant point here is that a CCD has two limits: low-light sensitivity (noise) and saturation limit. These two comprise the total range. The number of bits define the tonal range, as Jason says.
 
Remember after you shoot you have to pull the detail out of the shadows with curves or some other method if you expose for a bright source. The "fill flash" in P.S. Elements works well on some subjects.

A lot of people complain that they expose a shot with something super bright in it, and everything else is really dark. Well, the camera is rendering the scene accurately. For a pleasing image, you need to use curves to bring the detail out of the dark areas. I have a great example of this at home, which I'll try to post later. There are samples here too.

Jason
T Blair said:
I appreciate all this information, some of it I get, some I don't.
Practical application is what I am after. I was shooting in the LA
Train Station yesterday and I found myself wondering if I had to
choose to wash out the light fixtures to get the detail in the
tiled walls. I estimated over a 7 stop difference and I knew that
that couldn't be had on slide film. I remember the owner of Camera
West in Monterey Ca telling me that some films have an 11 stop
range. I had little time last night to experiement, but it seems
like it may be worth the time to take notes on exposure range as I
take pictures, then post some and see if we agree on the effective
range of the D60.

As an aside, I find these technical discussions much more
intereting than the "When is Canon coming out with an EOS3D
camera?" ones.

Best to all,
--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
Click to expand...
 
See http://www.luminous-landscape.com/discus/messages/8/14987.html
(4-5 posts down)

quote:

If you refer to the characteristic curves for various film types you’ll see that density is measured on the vertical axis and exposure on the horizontal (both Log). It should also be pretty obvious that the slopes of the curves are significantly different. The slope for a negative type film is shallow and that of a chrome very steep.

For negatives the typical exposure range (from the curve) is in the order of 8 fstops (maybe more), but the density range is approximately 2.4

For chromes the typical exposure range (from the curve) is in the order of 5 fstops, but the density range is close to 3.6

Density range and dynamic range is the same thing, so you will see that a negative type film compresses the 8 fstops of exposure into a density range of 2.4, whist a chrome type film only compresses 5 stops into a density range of 3.6

The slope of a negative film being quite shallow means that an exposure error of say 1 stop will have considerably less impact on film density than the same exposure error on a chrome, where the slope is very steep. This why we say that a negative type film has greater exposure latitude than chromes.

I suggest that those really interested in the subject get their hands on the data sheets for the films they use. Fuji supplies an excellent handbook called the “Professional Data Guide”. Once you would get to see the curves the mist will lift. Trying to match the slope (I think some of you are calling this film gamma) with the gamma of the scanner is very unlikely to prove anything.

--
Photos, tips and tests at:
http://www.geocities.com/glowluzid
 
glowluzid said:
See http://www.luminous-landscape.com/discus/messages/8/14987.html
(4-5 posts down)
Click to expand...
From the link above, there was this:

"I think we need to remember that dynamic range and exposure range is different. I thought this should have been clear from above. Dynamic range relates to density.

The negative will have a much smaller dynamic range and as such will be overall less dense than a chrome. However, the negative does have a much broader exposure range than chromes. That is to say that negatives will better handle situations were light to dark is extreme. "

I wish I could have said it that simply.

Jason
 
You'll get a better understanding of how your camera responds if you plot out the curve. Shoot in raw and expose a grey card over a broad range...at least 8-9 stops. This would represent the average brightness range of a normal outdoor scene. You'll see where the camera (actually the software) distorts the data, usually for good reason, and what your latitude is. The D30 has, in my opinion, about 5 good stops of information. And with virtually no latitude whatsoever. The D60 might be different.

In any event, expose for the lamps in the station and you'll maybe lose some shadow detail on the wall. But perhaps you can dig it out in Photoshop? Seven stops is borderline.
T Blair said:
I appreciate all this information, some of it I get, some I don't.
Practical application is what I am after. I was shooting in the LA
Train Station yesterday and I found myself wondering if I had to
choose to wash out the light fixtures to get the detail in the
tiled walls. I estimated over a 7 stop difference and I knew that
that couldn't be had on slide film. I remember the owner of Camera
West in Monterey Ca telling me that some films have an 11 stop
range. I had little time last night to experiement, but it seems
like it may be worth the time to take notes on exposure range as I
take pictures, then post some and see if we agree on the effective
range of the D60.

As an aside, I find these technical discussions much more
intereting than the "When is Canon coming out with an EOS3D
camera?" ones.

Best to all,
--
T Blair
Gear: D60, EOS3, G1, 16-35, 28-70,
70-200 2.8, 85 1.8, 550EX, 420EX
Click to expand...
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top