Ugly Hedgehog - Photography Forum
Home Active Topics Newest Pictures Search Login Register
Main Photography Discussion
Dynamic Range
Page <prev 2 of 3 next>
Jan 26, 2016 18:00:09   #
camerapapi Loc: Miami, Fl.
 
You nailed it right. The farther away from the computer the wider the dynamic range.

Reply
Jan 26, 2016 18:20:18   #
Apaflo Loc: Anchorage, Alaska
 
Bull-Dozer wrote:
Thinking about the dynamic range of modern sensors I am concerned that as the size of the "quantum wells" (active piece of silicon on the sensor for one pixel) continue to get smaller and smaller. As the wells get smaller and smaller there is literately not enough physical space for the electrons that determine brightness of the pixel, with out "saturating", which will reduce DR going forward as resolutions continue to rise. This is mostly for 35 mm DSLR cameras as medium format camera "backs" still have fat little "quantum wells" that can hold more than enough electrons.
For example a 14EV would require 2 raised to the 14th power electrons or 16,384 to be contained within the quantum well without overfilling it.. 15 EV would require 2 raised to the 15th power or 32,768 electrons.
Thinking about the dynamic range of modern sensors... (show quote)

So lets consider some examples for a few cameras.

Camera Max. Saturation Capacity Electrons
Canon 1D MIV 47756
Canon 1DX 90101
Canon 5D MIII 70635
Canon 7D MII 29544
Nikon D3300 35989
Nikon D3s 84203
Nikon D4 118339
Nikon D4s 128489
Nikon D610 74971
Nikon D750 81608
Nikon D800 48818
Nikon D810 78083


Hi quality sensors with a full well capacity greater than 32,768 electrons are not unusual.

Reply
Jan 26, 2016 18:37:16   #
John_F Loc: Minneapolis, MN
 
As I understand 'dynamic range' it is the ability of the sensor to record the presence of light that exceeds the nominally human visible wave lengths. Generally, wave lengths in the 400 to 700 nanometer range. It is reported that some people claim to see down to 380 and somewhat less and up to 720 and somewhat more. Is this wrong :?:

Reply
 
 
Jan 26, 2016 19:12:26   #
SharpShooter Loc: NorCal
 
Mark7829 wrote:
Bracketing is HDR....


Mark, with all due respect, bracketing, in and of itself, is not HDR. The do HDR, yes you definately have to bracket.
I have bracketed for years and not done HDR. 90% of my bracketing today, is NOT intended to ever be HDR. It could be but its not.
When I hear someone talking about bracketing, to me, HDR does not even come to mind, but maybe because that's not what I call it or have ever heard it called. maybe just me but I certainly see them as completely separate and individual process with separate names.
Not saying you're wrong, it could just be me!! ;-)
SS

Reply
Jan 26, 2016 19:44:20   #
dhellner Loc: milwaukee wi.
 
kymarto wrote:
Rockwell is dead wrong, in my experience. That's like saying that an old VW beetle is as good as a Ferrari Testrosa, because both can cruise at 60 mph on a straightaway. Let the road be a bit more challenging, and you will immediately feel the difference.

In my case I moved from a Nikon D300 to a D800E. I also use a Canon 5D III professionally at work. The D300 has a dynamic range of 12 EV. The Canon is even less, at 11.7 EV. The D800 comes in at 14.3 EV. The differences are noteworthy in my photographs.

First, let's get something straight: if the scene you are recording has less dynamic range than your camera sensor, it makes no difference at all. It is when the dynamic range of the scene itself gets high that you can tell the men from the boys.

Rockwell says that a good photographer will adjust his camera to the dynamic range of the scene. That's total BS. You can only make a compromise--if the dynamic range of the scene exceeds that of your sensor, you have to decide what to sacrifice--you can either crush the shadows or blow the highlights. Period.

If all you are doing is taking whatever jpeg the camera spits out, then you aren't really going to appreciate higher dynamic range: it is when you get into a good graphics program like PS that you start to appreciate what a couple of extra EV of dynamic range really give you.

I'm going to post some examples. First is a pic I shot a month ago of a stream in the mountains. To retain highlight detail I had to underexpose the rocks. That's the first pic. Then in PS I lifted the shadows. There was plenty of detail in the shadows, and no noise to speak of. That shot would not have worked with either the D300 or the 5D. The tonal values in the shadows would have been blocked up in the D300, and extremely noisy in the 5D.

Just how noisy are Canon shadows compared to Nikon. That's the next frame. This is a direct comparison between the 5D3 and the D800E. I shot identical frames with the two cameras, underexposed 3 EV, and then raised the exposure 3 stops to get a normal exposure--at both ISO 100 and ISO 800. The dynamic range of this particular photo is quite low, so of course I could have exposed normally, but had there been highlights in this picture, or perhaps a bright sky, the ability to expose for the highlights and then raise the shadow values would be invaluable.

If you notice, the Canon is basically noisier at ISO 100 than the Nikon at ISO 800. The Nikon is extremely clean in the shadows at ISO 100--almost as clean as it is exposed normally. Now if I had exposed both normally, you wouldn't see a difference, but with the Canon I would have had no chance to underexpose to preserve highlights and still had clean shadows to work with.

Finally is a pic of a house abandoned five years ago in Fukushima. Again, I was able to expose to preserve highlights, and still keep good shadow tonality and detail.

So that's why I say Rockwell is an amateur if he says that DR makes no difference.
Rockwell is dead wrong, in my experience. That's l... (show quote)


:thumbup: :thumbup: :thumbup:

Reply
Jan 26, 2016 20:06:28   #
selmslie Loc: Fernandina Beach, FL, USA
 
SharpShooter wrote:
... bracketing, in and of itself, is not HDR. ... Not saying you're wrong, it could just be me!! ...

No, you are correct. Bracketing predates HR and digital imaging by decades. I would venture to say that most of us that bracket have not intention to use HDR.

When we resort to bracketing we are more likely are doing it because we recognize that we cannot always trust our meters or even our own judgement. They are not infallible. It is a prudent measure to insure that at least one of the images is closest to the optimal exposure.

Reply
Jan 26, 2016 20:10:42   #
BHC Loc: Strawberry Valley, JF, USA
 
kymarto wrote:
Rockwell is dead wrong, in my experience. That's like saying that an old VW beetle is as good as a Ferrari Testrosa, because both can cruise at 60 mph on a straightaway. Let the road be a bit more challenging, and you will immediately feel the difference.

.../...

Rockwell says that a good photographer will adjust his camera to the dynamic range of the scene.

I said the guy was useful, not right. Maybe he copied his definition from Wiki; that was the only part I was referring to. What else do you expect from someone who says that "P" mode means Professional? You're right; I should have quoted his definition. I would not be troubled by putting a note under the remainder of the quotation to the effect that the remainder of the article was BS.

Reply
 
 
Jan 26, 2016 20:17:30   #
kymarto Loc: Portland OR and Milan Italy
 
John_F wrote:
As I understand 'dynamic range' it is the ability of the sensor to record the presence of light that exceeds the nominally human visible wave lengths. Generally, wave lengths in the 400 to 700 nanometer range. It is reported that some people claim to see down to 380 and somewhat less and up to 720 and somewhat more. Is this wrong :?:


Actually DR refers to the range of light intensities that the sensor can record without introducing a certain level of noise. Low light intensities tend to be compromised because the sensor itself introduces various types of noise, and as the signal gets weaker (less light) at a certain point the inherent noise in the sensor overwhelms it.

I think what you are referring to is "color depth"--not what exceeds human vision, but what chrominance values can be recorded, again as compared to how much noise is present at those values.

Reply
Jan 26, 2016 20:30:15   #
kymarto Loc: Portland OR and Milan Italy
 
[quote=SharpShooter]
Mark7829 wrote:
"you have to decide what to sacrifice--you can either crush the shadows or blow the highlights. Period. []

You can also...., gasp..., use the other arrow in the quiver labeled HDR!!
No matter how much DR any camera has, you will eventually run out of DR if you have deep enough shadows and the light source is strong enough.
I could really care less about DR, because my camera can actually do HDR. You can make HDR look as good or as bad as you want. Used properly it's a much more powerful tool than any high DR camera! ;-)
SS
"you have to decide what to sacrifice--you ca... (show quote)


Yes and no. You are preaching to the choir about HDR--it has been my main focus in photography for a number of years and I think I've gotten pretty good at it. And indeed when you run out of DR, HDR is the only way to go. But there is a fly in the ointment, which is movement. Even with Photomatix's "selective deghosting", which has saved many a shot for me, subject movement tends to limit what you can do with HDR.

I have been amazed shooting with the D800 at how many shots I can cook up well from a single RAW frame that formerly would have required HDR. Of course there is always going to be the shot that won't work from a single frame, but having a couple of extra stops of DR has dramatically extended what I can do with action shots.

This formerly would have required HDR
This formerly would have required HDR...
(Download)

Of course no choice here but HDR
Of course no choice here but HDR...
(Download)

An abandoned game parlor in Fukushima. Even with the D800 I needed +-7 EV brackets
An abandoned game parlor in Fukushima. Even with t...
(Download)

This would not have worked as HDR, but DR saved the day :)
This would not have worked as HDR, but DR saved th...
(Download)

Reply
Jan 26, 2016 22:50:56   #
Trabor
 
Apaflo wrote:
So lets consider some examples for a few cameras.

Camera Max. Saturation Capacity Electrons
Canon 1D MIV 47756
Canon 1DX 90101
Canon 5D MIII 70635
Canon 7D MII 29544
Nikon D3300 35989
Nikon D3s 84203
Nikon D4 118339
Nikon D4s 128489
Nikon D610 74971
Nikon D750 81608
Nikon D800 48818
Nikon D810 78083


Hi quality sensors with a full well capacity greater than 32,768 electrons are not unusual.
So lets consider some examples for a few cameras. ... (show quote)


The numbers you refer to represent one end of the dynamic range, the high end or saturation , the low end is related to the lowest usable value which is related to the noise level. For example when the RMS noise is equal to the average signal. Or perhaps when SNR=3 dB. The ratio of saturation to lowest usable is the dynamic range. Either number by itself is only part of the story

Reply
Jan 27, 2016 00:42:39   #
Apaflo Loc: Anchorage, Alaska
 
Trabor wrote:
The numbers you refer to represent one end of the dynamic range, the high end or saturation , the low end is related to the lowest usable value which is related to the noise level. For example when the RMS noise is equal to the average signal. Or perhaps when SNR=3 dB. The ratio of saturation to lowest usable is the dynamic range. Either number by itself is only part of the story

But what I was specifically responding to was a statement about the required full well capacity of a sensor to get a 15 EV range, with the implication that 32,768 was somehow an astoundingly high number. It obviously is not.

As for the low end, here's more data, all of which comes from

http://www.sensorgen.info/

Camera Max. Saturation Capacity Electrons Min. Read Noise Electrons
Canon 1D MIV 47756 1.4
Canon 1DX 90101 1.2
Canon 5D MIII 70635 2.4
Canon 7D MII 29544 2.0
Nikon D3300 35989 2.3
Nikon D3s 84203 2.8
Nikon D4 118339 1.8
Nikon D4s 128489 1.3
Nikon D610 74971 2.9
Nikon D750 81608 2.1
Nikon D800 48818 2.6
Nikon D810 78083 1.3

Just be aware the these numbers provide an "Engineering" dynamic range, which does not actually indicate the practical or useful dynamic range of the camera.

Reply
 
 
Jan 27, 2016 00:48:30   #
BHC Loc: Strawberry Valley, JF, USA
 
Y'all lost me with "Max. Saturation Capacity Electrons"!

Reply
Jan 27, 2016 02:02:31   #
Apaflo Loc: Anchorage, Alaska
 
Mogul wrote:
Y'all lost me with "Max. Saturation Capacity Electrons"!

That is the number of electron charges held by a sensor well when it reaches maximum saturation. It is not possible to capture more photons. Minimum read noise is the charge that will be read if no light is allowed to reach the sensor. At least that many photons would need to be captured to begin recording light at a single sensor site.

One is the maximum signal, the other is the minimum signal. The ratio between the two is "dyanmic range".

Reply
Jan 27, 2016 03:48:58   #
BHC Loc: Strawberry Valley, JF, USA
 
Apaflo wrote:
That is the number of electron charges held by a sensor well when it reaches maximum saturation. It is not possible to capture more photons. Minimum read noise is the charge that will be read if no light is allowed to reach the sensor. At least that many photons would need to be captured to begin recording light at a single sensor site.

One is the maximum signal, the other is the minimum signal. The ratio between the two is "dyanmic range".

Sorry, that's way beyond me, far too technical! Now, if you would care to discuss the psychological motivations for introducing data that is undoubtedly correct but has no meaning for most photographers and which seems to far exceed the needs of the OP, I would be glad to oblige you. Or we can debate the number of electrons an undetermined number of angels can hold while perched on the head of a pen, I might have some points to make. But, as things stand, you've lost me.

Reply
Jan 27, 2016 04:00:43   #
Apaflo Loc: Anchorage, Alaska
 
Mogul wrote:
Sorry, that's way beyond me, far too technical! Now, if you would care to discuss the psychological motivations for introducing data that is undoubtedly correct but has no meaning for most photographers and which seems to far exceed the needs of the OP, I would be glad to oblige you. Or we can debate the number of electrons an undetermined number of angels can hold while perched on the head of a pen, I might have some points to make. But, as things stand, you've lost me.

If we dumb everything down to the least common denominator nobody would learn anything. I am sorry you are lost, but it isn't my job to find you.

Reply
Page <prev 2 of 3 next>
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Main Photography Discussion
UglyHedgehog.com - Forum
Copyright 2011-2024 Ugly Hedgehog, Inc.