Lets talk about dynamic range.
Jan 2, 2018 04:27:50 #
THE HUMAN EYE
The human eye can actually perceive a greater dynamic range than is ordinarily possible with a camera. If we were to consider situations where our pupil opens and closes for varying light, our eyes can see over a range of nearly 24 f-stops.
On the other hand, for accurate comparisons with a single photo (at constant aperture, shutter and ISO), we can only consider the instantaneous dynamic range (where our pupil opening is unchanged). This would be similar to looking at one region within a scene, letting our eyes adjust, and not looking anywhere else. For this scenario there is much disagreement, because our eye's sensitivity and dynamic range actually change depending on brightness and contrast. Most estimate anywhere from 10-14 f-stops.
The problem with these numbers is that our eyes are extremely adaptable. For situations of extreme low-light star viewing (where our eyes have adjusted to use rod cells for night vision), our eyes approach even higher instantaneous dynamic ranges.
Paper Prints
Kodak, in its technical literature, still refers to photos as "reflection prints."
The paper reflects some light, even from its darkest portions; and it doesn't reflect all the light that falls on it, even from its whitest portions.
Let's plug in some numbers and see what this means. Even if the paper reflected just 5% of the light in the darkest areas and fully 80% of the light in its lightest areas, that would be a range of 4 stops. (80=5*16...and a 16x difference in light is four stops) an estimate of 2.5 stops is probably closer to reality in most cases.
You can also achieve a greater dynamic range with transmitted light against a highly reflective surface in a darkened room, i.e., by projecting a slide.
Monitors seem to vary greatly in what they can show, but it's probably fair to say it's more than a print.
Film v Digital
Kodak states Vision3 has a maximum dynamic range of 13 stops. perhaps a little more.
Digital varies a lot, the maximum value is at base ISO and decreases steadily as ISO increases. In another thread a figure of 6.5 stops was given as what was needed to get a clean exposure below that and noise becomes apparent.
DXO has tested many camera's and from my own limited experience the 6.5 stop figure seems fair. The DXO test data puts 6.5 stops at iso 800 on my K200D around 1600 on my K5 and i'd have to say going to the next stop noise is apparent on both of these bodies.
You can look up the data for your own camera's but as a trend it seems the crop sensor seems to hit this value around 1600 ISO Full frame seems to be a stop higher at 3200. The Nikon D5 is remarkable reaching ~6400 for the 6.5 value, The Pentax 645Z seems to be about the same as the D5 from iso 1600 upwards (it's better than the D5 below iso 1600).
I've seem people say that 12800 is still good on the D5 but the DX0 chart has the D5 and 645Z around 5.59 Stops (which the renowned "expert" claims is too low).
Whats that mean for us?
I have to be honest and say that 800 and 1600 seem to be a fair limit on ISO on my own camera's, If I go higher I still get a photo but the IQ seems to clearly have gone down. I'm clearly seeing noise above these thresholds on my monitor.
With the dynamic range of a paper print being 4 or perhaps less according to the sources i've found maybe you can push the iso a couple of stops further, but projected as a slide or on a screen, i think you will notice the difference.
For low light photography the exposure is a combination of ISO , Shutter speed and Aperture.
For my k5 if I limit ISO to 1600 a Shutter speed of 1/60th and f2 aperture I end up with EV8
"Subjects by candle light, flood lit buildings and structures, subject under bright street lighting"
So a 50mm lens at f2 and iso 1600 should look good. Bring out the tripod and I can lower the shutter speed.
I'm not going to get a kit lens to open so wide and f2 is probably not going to look very sharp either.
LIGHTING SITUATIONS EXPOSURE VALUES
Night, away from city lights, subject under starlight only. -6
Night, away from city lights, subject under crescent moon. -5
Night, away from city lights, subject under half moon. Meteors (during showers, with time exposure). -4
Night, away from city lights, subject under full moon. -3
Night, away from city lights, snowscape under full moon. -2
Subjects lit by dim ambient artificial light. -1
Subjects lit by dim ambient artificial light. 0
Distant view of lighted skyline. 1
Lightning (with time exposure). Total eclipse of moon. 2
Fireworks (with time exposure). 3
Candle lit close-ups. Christmas lights, floodlit buildings, fountains, and monuments. Subjects under bright street lamps. 4
Night home interiors, average light. School or church auditoriums. Subjects lit by campfires or bonfires. 5
Brightly lit home interiors at night. Fairs, amusement parks. 6
Bottom of rainforest canopy. Brightly lighted nighttime streets. Indoor sports. Stage shows, circuses. 7
Las Vegas or Times Square at night. Store windows. Campfires, bonfires, burning buildings. Ice shows, football, baseball etc. at night. Interiors with bright florescent lights. 8
Landscapes, city skylines 10 minutes after sunset. Neon lights, spotlighted subjects. 9
Landscapes and skylines immediately after sunset. Crescent moon (long lens). 10
Sunsets. Subjects in deep shade. 11
Half moon (long lens). Subject in open shade or heavy overcast. 12
Gibbous moon (long lens). Subjects in cloudy-bright light (no shadows). 13
Full moon (long lens). Subjects in weak, hazy sun. 14
Subjects in bright or hazy sun (Sunny f/16 rule). 15
Subjects in bright daylight on sand or snow. 16
Rarely encountered in nature. Some man made lighting. 17-21
Extremely bright. Rarely encountered in nature. 22-23
The table above gives lighting situations and the sort of EV value your going to have.
My kit lens runs from around f3.5 at 18mm to f5.6 at 55mm which limits me to around 10 Ev at 18mm and 11 EV at 55mm handheld (keeping shutter speed at 1/60th)
So that seems to be where I am limited of course if I use a tripod and increase my exposure time I can photograph darker scenes.
But other than that a typical full frame will gain me a stop and the d5 2 stops
It's not totally straight forward with sensors I have a panasonic G5 which is about 1 stop worse than my k5 but a panasonic gh4 is around the same as the k5 or a typical crop sensor dslr.
Hopefully this makes sense, and you can use this in planning your own photos. At least it shows there is an advantage to printing your photos. Sorry for not including references B&H has a good article Dynamic range explained by Bjorn Peterson.
Of course I don't expect any agreement on this and expect to get roasted :) When conditions are good you can get by with fairly modest equipment. Hopefully you can see from this it is pretty easy to get beyond your camera and lenses limits.
The human eye can actually perceive a greater dynamic range than is ordinarily possible with a camera. If we were to consider situations where our pupil opens and closes for varying light, our eyes can see over a range of nearly 24 f-stops.
On the other hand, for accurate comparisons with a single photo (at constant aperture, shutter and ISO), we can only consider the instantaneous dynamic range (where our pupil opening is unchanged). This would be similar to looking at one region within a scene, letting our eyes adjust, and not looking anywhere else. For this scenario there is much disagreement, because our eye's sensitivity and dynamic range actually change depending on brightness and contrast. Most estimate anywhere from 10-14 f-stops.
The problem with these numbers is that our eyes are extremely adaptable. For situations of extreme low-light star viewing (where our eyes have adjusted to use rod cells for night vision), our eyes approach even higher instantaneous dynamic ranges.
Paper Prints
Kodak, in its technical literature, still refers to photos as "reflection prints."
The paper reflects some light, even from its darkest portions; and it doesn't reflect all the light that falls on it, even from its whitest portions.
Let's plug in some numbers and see what this means. Even if the paper reflected just 5% of the light in the darkest areas and fully 80% of the light in its lightest areas, that would be a range of 4 stops. (80=5*16...and a 16x difference in light is four stops) an estimate of 2.5 stops is probably closer to reality in most cases.
You can also achieve a greater dynamic range with transmitted light against a highly reflective surface in a darkened room, i.e., by projecting a slide.
Monitors seem to vary greatly in what they can show, but it's probably fair to say it's more than a print.
Film v Digital
Kodak states Vision3 has a maximum dynamic range of 13 stops. perhaps a little more.
Digital varies a lot, the maximum value is at base ISO and decreases steadily as ISO increases. In another thread a figure of 6.5 stops was given as what was needed to get a clean exposure below that and noise becomes apparent.
DXO has tested many camera's and from my own limited experience the 6.5 stop figure seems fair. The DXO test data puts 6.5 stops at iso 800 on my K200D around 1600 on my K5 and i'd have to say going to the next stop noise is apparent on both of these bodies.
You can look up the data for your own camera's but as a trend it seems the crop sensor seems to hit this value around 1600 ISO Full frame seems to be a stop higher at 3200. The Nikon D5 is remarkable reaching ~6400 for the 6.5 value, The Pentax 645Z seems to be about the same as the D5 from iso 1600 upwards (it's better than the D5 below iso 1600).
I've seem people say that 12800 is still good on the D5 but the DX0 chart has the D5 and 645Z around 5.59 Stops (which the renowned "expert" claims is too low).
Whats that mean for us?
I have to be honest and say that 800 and 1600 seem to be a fair limit on ISO on my own camera's, If I go higher I still get a photo but the IQ seems to clearly have gone down. I'm clearly seeing noise above these thresholds on my monitor.
With the dynamic range of a paper print being 4 or perhaps less according to the sources i've found maybe you can push the iso a couple of stops further, but projected as a slide or on a screen, i think you will notice the difference.
For low light photography the exposure is a combination of ISO , Shutter speed and Aperture.
For my k5 if I limit ISO to 1600 a Shutter speed of 1/60th and f2 aperture I end up with EV8
"Subjects by candle light, flood lit buildings and structures, subject under bright street lighting"
So a 50mm lens at f2 and iso 1600 should look good. Bring out the tripod and I can lower the shutter speed.
I'm not going to get a kit lens to open so wide and f2 is probably not going to look very sharp either.
LIGHTING SITUATIONS EXPOSURE VALUES
Night, away from city lights, subject under starlight only. -6
Night, away from city lights, subject under crescent moon. -5
Night, away from city lights, subject under half moon. Meteors (during showers, with time exposure). -4
Night, away from city lights, subject under full moon. -3
Night, away from city lights, snowscape under full moon. -2
Subjects lit by dim ambient artificial light. -1
Subjects lit by dim ambient artificial light. 0
Distant view of lighted skyline. 1
Lightning (with time exposure). Total eclipse of moon. 2
Fireworks (with time exposure). 3
Candle lit close-ups. Christmas lights, floodlit buildings, fountains, and monuments. Subjects under bright street lamps. 4
Night home interiors, average light. School or church auditoriums. Subjects lit by campfires or bonfires. 5
Brightly lit home interiors at night. Fairs, amusement parks. 6
Bottom of rainforest canopy. Brightly lighted nighttime streets. Indoor sports. Stage shows, circuses. 7
Las Vegas or Times Square at night. Store windows. Campfires, bonfires, burning buildings. Ice shows, football, baseball etc. at night. Interiors with bright florescent lights. 8
Landscapes, city skylines 10 minutes after sunset. Neon lights, spotlighted subjects. 9
Landscapes and skylines immediately after sunset. Crescent moon (long lens). 10
Sunsets. Subjects in deep shade. 11
Half moon (long lens). Subject in open shade or heavy overcast. 12
Gibbous moon (long lens). Subjects in cloudy-bright light (no shadows). 13
Full moon (long lens). Subjects in weak, hazy sun. 14
Subjects in bright or hazy sun (Sunny f/16 rule). 15
Subjects in bright daylight on sand or snow. 16
Rarely encountered in nature. Some man made lighting. 17-21
Extremely bright. Rarely encountered in nature. 22-23
The table above gives lighting situations and the sort of EV value your going to have.
My kit lens runs from around f3.5 at 18mm to f5.6 at 55mm which limits me to around 10 Ev at 18mm and 11 EV at 55mm handheld (keeping shutter speed at 1/60th)
So that seems to be where I am limited of course if I use a tripod and increase my exposure time I can photograph darker scenes.
But other than that a typical full frame will gain me a stop and the d5 2 stops
It's not totally straight forward with sensors I have a panasonic G5 which is about 1 stop worse than my k5 but a panasonic gh4 is around the same as the k5 or a typical crop sensor dslr.
Hopefully this makes sense, and you can use this in planning your own photos. At least it shows there is an advantage to printing your photos. Sorry for not including references B&H has a good article Dynamic range explained by Bjorn Peterson.
Of course I don't expect any agreement on this and expect to get roasted :) When conditions are good you can get by with fairly modest equipment. Hopefully you can see from this it is pretty easy to get beyond your camera and lenses limits.
Jan 2, 2018 04:54:34 #
Good read and thank you! It will take a few passes and some thought, but nice to learn something we often hear about.
Jan 2, 2018 06:27:02 #
DaveO wrote:
Good read and thank you! It will take a few passes and some thought, but nice to learn something we often hear about.
I'm probably wrong on a few points but a few references
https://www.cambridgeincolour.com/tutorials/dynamic-range.htm
https://www.bhphotovideo.com/explora/photography/tips-and-solutions/dynamic-range-explained
https://www.photo.net/discuss/threads/why-dynamic-range-limit-exists-in-photo-paper-like-fuji-froniter.286356/
https://wolfcrow.com/blog/where-cameras-stand-in-dynamic-range-film-vs-digital/
This can show your camera dynamic range v ISO
http://www.photonstophotos.net/Charts/DXOPDR.htm
Jan 2, 2018 07:21:36 #
It's nice to see things like this on the forum that are informational and quite possibly a stimulus that will further enhance hobby enjoyment for some.
Thanx for the additional resources!
Thanx for the additional resources!
Jan 2, 2018 09:31:51 #
This is the single best post I've seen on the web regarding DR...thank you for taking the time and effort to share with us!! 
Jan 2, 2018 10:37:45 #
bclaff
Loc: Sherborn, MA (18mi SW of Boston)
blackest wrote:
... In another thread a figure of 6.5 stops was given as what was needed to get a clean exposure below that and noise becomes apparent.
DXO has tested many camera's and from my own limited experience the 6.5 stop figure seems fair. ...
DXO has tested many camera's and from my own limited experience the 6.5 stop figure seems fair. ...
6.5 is the figure used at PhotonsToPhotos to determine Low Light ISO.
Here's a link to the table with relevant values: www.photonstophotos.net/Charts/PDR.htm#resizableT
It's important to realize that such sensor testing reports dynamic range in linear space.
But the final image, and human vision, operate in a more logarithmic fashion.
So any comparisons would be "apples to oranges".
Jan 2, 2018 11:08:24 #
bclaff wrote:
6.5 is the figure used at PhotonsToPhotos to determine Low Light ISO.
Here's a link to the table with relevant values: www.photonstophotos.net/Charts/PDR.htm#resizableT
It's important to realize that such sensor testing reports dynamic range in linear space.
But the final image, and human vision, operate in a more logarithmic fashion.
So any comparisons would be "apples to oranges".
Here's a link to the table with relevant values: www.photonstophotos.net/Charts/PDR.htm#resizableT
It's important to realize that such sensor testing reports dynamic range in linear space.
But the final image, and human vision, operate in a more logarithmic fashion.
So any comparisons would be "apples to oranges".
isn't that the purpose of the F stop ? each stop is double the value of the one before. so they are log values.
Maybe what I have is an over simplification but it seems fair.
I prefer this chart
http://www.photonstophotos.net/Charts/DXOPDR.htm
It may not be as accurate but really who needs that :) just a fair idea of what iso setting is usable to get good results. It's a lot less disappointing than trying to use iso's which are too noisy to produce anything to be proud off.
Jan 2, 2018 11:24:32 #
bclaff
Loc: Sherborn, MA (18mi SW of Boston)
bclaff wrote:
It's important to realize that such sensor testing reports dynamic range in linear space.
But the final image, and human vision, operate in a more logarithmic fashion.
So any comparisons would be "apples to oranges".
But the final image, and human vision, operate in a more logarithmic fashion.
So any comparisons would be "apples to oranges".
blackest wrote:
isn't that the purpose of the F stop ? each stop is double the value of the one before. so they are log values.
Maybe what I have is an over simplification but it seems fair.
Maybe what I have is an over simplification but it seems fair.
Your response doesn't have to do with my warning.
Dynamic range reported by PhotonsToPhotos, DxOMark, etc. is linear dynamic range.
Dynamic range that you see is logarithmic.
The raw linear data goes through a power function and tone curve to make an image for you to view.
So linear dynamic range cannot be compared to human vision dynamic range in a meaningful way.
Jan 2, 2018 13:36:40 #
bclaff wrote:
6.5 is the figure used at PhotonsToPhotos to determine Low Light ISO.
Here's a link to the table with relevant values: www.photonstophotos.net/Charts/PDR.htm#resizableT
It's important to realize that such sensor testing reports dynamic range in linear space.
But the final image, and human vision, operate in a more logarithmic fashion.
So any comparisons would be "apples to oranges".
Here's a link to the table with relevant values: www.photonstophotos.net/Charts/PDR.htm#resizableT
It's important to realize that such sensor testing reports dynamic range in linear space.
But the final image, and human vision, operate in a more logarithmic fashion.
So any comparisons would be "apples to oranges".
Your correct it is comparing apples to oranges. The human eye is not manufactured under strict tolerances and with the limitations of sensor development. At present to have a camera processor that could compare to the human brain to interpret this data would be akin to lugging around the old Kodak DSLR with its back pack.
I felt this was an excellent post to get us to realize the limitations between camera and the human perspective. To me, it also pointed out why different people view photo's differently, the human eye is NOT manufactured.
Ron
Jan 2, 2018 15:44:04 #
blackest wrote:
THE HUMAN EYE br The human eye can actually percei... (show quote)
Thank you for your informative article. Question: Does it follow that, if a DSLR is capable of producing a photo with a dynamic range of up to 6.5 stops, if you bracket a shot +/- 2 stops, you potentially can produce a photo with a dynamic range of up to 10.5 stops using an HDR program like Photomatix?
Jan 2, 2018 17:19:37 #
This is a bunch of pseudo-scientific bafflegab which obfuscates what any photographer can figure out within one hour of getting a different camera with just a little experimentation.
Jan 3, 2018 06:57:51 #
mcveed wrote:
This is a bunch of pseudo-scientific bafflegab which obfuscates what any photographer can figure out within one hour of getting a different camera with just a little experimentation.
There is no original research here, it's all on the internet. The only thing i wasn't so sure about is the 6.5 stop range needed for a good capture but from comparing my own cameras with the dxo charts it seems a fair guideline of the limits for a reasonable photo.
Check some of the links , see if i'm misinterpreting what they say. I'm happy to be wrong if the right becomes clear.
What I'm saying really there is a high iso beyond which results become disappointing this varies by camera. Paper has a limited dynamic range which you need to fit your image into...
Not great revelations true enough but gives you an idea of the practical limitations of your equipment.
Oh and the D5 really does well in low light.
Jan 3, 2018 07:33:19 #
billnikon
Loc: Pennsylvania/Ohio/Florida/Maui/Oregon/Vermont
blackest wrote:
THE HUMAN EYE br The human eye can actually percei... (show quote)
On the other hand, if I am shooting green grass with green trees in the back ground on a bright overcast day, I have no worries.
Jan 3, 2018 15:15:04 #
blackest wrote:
THE HUMAN EYE br The human eye can actually percei... (show quote)
I can confirm that my GH4 is in the same ballpark as many APS-C and DX crop sensor dSLRs and mirrorless cameras... in good light. It's definitely better than all the midrange Canons up through the EOS 7D II. I was shocked to see that in the dxomark ratings, and to actually test it. However, there is a slight low light penalty for using Micro 4/3. It seems to be about 1/3 to 2/3 stop, depending on real world comparisons.
Your analysis is pretty sound. However, I'm continually amazed at the emphasis folks on this site place on noise, dynamic range, and other technical nuances. In actual, real-world usage, there is seldom much difference to notice unless you are a highly critical, seasoned photographer. The rest of the world (about 7 billion of us) doesn't know or care.
They DO care about what they are SEEING in our photographs.
Unless your technical expertise is really terrible, what they're seeing isn't noise, dynamic range, or incredible shadow and highlight detail. It's probably not even perfect color balance. All those things help, but only so much.
There are plenty of really good cameras on the market now, and in the right hands, just about any of them can make a great image. In the wrong hands, just about any of them can make a horrible snapshot.
I may get a lot of flak for this, but the Pareto Principle (that roughly 80% of the effects come from 20% of the causes) applies to photography as well as any other situation. In the case of photography, the important causes are mostly behind the camera (knowledge, training, experience, vision, passion, mental focus...).
That doesn't mean we shouldn't strive for everything we can get from our gear! It just means that there are other, more important factors in play most of the time.
Jan 4, 2018 06:17:04 #
burkphoto wrote:
I can confirm that my GH4 is in the same ballpark ... (show quote)
Thank you, I'm so glad the GH4 does appear to match up in the real world with the Dx0 data.
Cameras have their limits and knowing the limits helps. If you are shooting within the limits of the camera then technically there isn't a problem. You can just concentrate on the photo you want to capture, the story you want to tell.
Take my k5 for instance I now have a practical hard limit on ISO of 1600 so giving an iso range of 80 - 1600 thats useful to know because sometimes that isn't enough my fastest lens if f1.7 50mm I think with image stabilisation that 1/60th is a reasonable minimum shutter speed
https://www.scantips.com/lights/exposurecalc.html comes up with 3.5 EV
I havent got enough of a lens in my 18-55mm kit lens to get that low minimum at 18mm would be around 6 EV.
At least with the understanding of what light i need, I can make better choices of which lens to use.
The rest of it is up to me, but at least i'm not trying to put a mini in an f1 race and expecting to get a result :)
Your right it is a little thing really, but those little things make a difference at times. An F2.8 lens and a full frame camera will get results at evening sports events where an aps-c camera with an f4 is at a 2 stop disadvantage.
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.