Reading mode: Light Dark

Marianne's data and the D4's low ISO behavior

Go to last post
R

rhlpetrus

Forum Pro
Messages
27,478
Solutions
3
Reaction score
5,418
Location
Campinas, BR
I think some may have overlooked what I find the most interesting info from Marianne's wonderful graphs posted here (which have reached 150 posts):

http://forums.dpreview.com/forums/read.asp?forum=1021&message=40756917

The ISO200 data indicates that the D4 behaves quite differently than the D3 and the D3s at base ISO:



Observe that the D4 curve, as the EV value goes down, starting at -2, stays flatter and actually more similar to the D800's than to the D3/D3s (use 12MP curves for reference). Eventually, the D800's curve stays flatter longer, from -4 and lower.

That means the shadows behavior of the D4 will be very close to the D800 down to -4EV, which should translate im much cleaner results, compared to D3/D3s, at low ISO. That will also translate in wider base ISO DR, something not many were expecting.

And we are looking at ISO200, the ISO 100 curves for D800 and D4 should be higher, pushing them even higher in terms of shadows/DR performance compared to D700/D3s.
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
Renato, Astute observation. I appreciate Marianne's efforts and generosity in sharing her preliminary data. If these hold true, then both the D4/D800 should have wider DR and detail recovery in the shadows at lower (base) ISO than the D3/D700. If I'm not mistaken, I may have seen a demonstration of this observation in one of your posts in the last several weeks.

Cheers.
 
reginalddwight said:
Renato, Astute observation. I appreciate Marianne's efforts and generosity in sharing her preliminary data. If these hold true, then both the D4/D800 should have wider DR and detail recovery in the shadows at lower (base) ISO than the D3/D700. If I'm not mistaken, I may have seen a demonstration of this observation in one of your posts in the last several weeks.
Click to expand...
The ISo 100 behavior should be about 3db higher for both along the whole range. The D3/D700 and D3s have base ISO at 200, so that would mean even better behavior than the ISO200 graph indicates for the new FFs.

:) !

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
rhlpetrus said:
I think some may have overlooked what I find the most interesting info from Marianne's wonderful graphs posted here (which have reached 150 posts):
Click to expand...
possibly the biggest improvement in the D4 sensor, along with video is the increase in dynamic range at base iso

The problem is that for the D800 there is no ISO200 or ISO100 NEF's in the wild

IF the D800 behaves like previous Nikon sony sensors (D3x, D7000), ISO100 for the D800 will be truly special meaning huge dynamic range

We really need to see that D800 ISO100 NEF
 
David314 said:
rhlpetrus said:
I think some may have overlooked what I find the most interesting info from Marianne's wonderful graphs posted here (which have reached 150 posts):
Click to expand...
possibly the biggest improvement in the D4 sensor, along with video is the increase in dynamic range at base iso

The problem is that for the D800 there is no ISO200 or ISO100 NEF's in the wild

IF the D800 behaves like previous Nikon sony sensors (D3x, D7000), ISO100 for the D800 will be truly special meaning huge dynamic range

We really need to see that D800 ISO100 NEF
Click to expand...
Guess, normalized DxO Mark 8MP: 15.3 EV.

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
rhlpetrus said:
I think some may have overlooked what I find the most interesting info from Marianne's wonderful graphs posted here (which have reached 150 posts):
Click to expand...
Thank you for the compliment, although to be fair, I should point out that this goes back to Bill Claff's earliest work assessing the D4 read noise. The same effect can be seen in his DR charts.

I must say, though, that the newest and most exciting revelation I have to share is regarding the D800:
http://forums.dpreview.com/forums/read.asp?forum=1021&message=40773567
 
Marianne Oelund said:
rhlpetrus said:
I think some may have overlooked what I find the most interesting info from Marianne's wonderful graphs posted here (which have reached 150 posts):
Click to expand...
Thank you for the compliment, although to be fair, I should point out that this goes back to Bill Claff's earliest work assessing the D4 read noise. The same effect can be seen in his DR charts.
Click to expand...
Yes, but I learned from my basic statistics course that the way one presents data is as important as the data's quality.
Marianne Oelund said:
I must say, though, that the newest and most exciting revelation I have to share is regarding the D800:
http://forums.dpreview.com/forums/read.asp?forum=1021&message=40773567
Click to expand...
Awesome!

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
rhlpetrus said:
That means the shadows behavior of the D4 will be very close to the D800 down to -4EV, which should translate im much cleaner results, compared to D3/D3s, at low ISO. That will also translate in wider base ISO DR, something not many were expecting.
Click to expand...
For that DR to be utilized naturally shadows need to be raised during PP and so it'll be important to see whether the D4 exhibits any pattern noises in the shadows like the D3/D700/D3s, as some of the extra DR would be wasted if corrupted with pattern noise. I looked at NEFs posted on imaging resource today and the shadows appear to be free of banding although the exposure and content make it difficult to fully assess (need uniform patches at varying EVs).
 
Horshack said:
rhlpetrus said:
That means the shadows behavior of the D4 will be very close to the D800 down to -4EV, which should translate im much cleaner results, compared to D3/D3s, at low ISO. That will also translate in wider base ISO DR, something not many were expecting.
Click to expand...
For that DR to be utilized naturally shadows need to be raised during PP and so it'll be important to see whether the D4 exhibits any pattern noises in the shadows like the D3/D700/D3s, as some of the extra DR would be wasted if corrupted with pattern noise. I looked at NEFs posted on imaging resource today and the shadows appear to be free of banding although the exposure and content make it difficult to fully assess (need uniform patches at varying EVs).
Click to expand...
If D7000's sensor is a good reference for D800's behavior, it'll be very free of any banding in shadows at base ISO. But that's another technology.

Which converter did you use for the D4's NEFs?
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
rhlpetrus said:
If D7000's sensor is a good reference for D800's behavior, it'll be very free of any banding in shadows at base ISO. But that's another technology.
Click to expand...
Right, D800 is almost a given, D4 is the question.
rhlpetrus said:
Which converter did you use for the D4's NEFs?
Click to expand...
I used LR 3. I hacked the files to look like D3 NEFs (Iliah's idea from way back).
 
I apologize if this was covered already.

So I am assuming the interpretation of these measurements is that at equivalent resolutions there is only about an approximate half stop difference between the D800 vs D4 in the dark regions of high ISO's. (is this the correct interpretation?)

Given this, why do the Nikon official "specs" for the D800 and D4 list the max ISO's as 26k vs 204k, a three stop difference?

--
Jeff at http://folkins.net
 
Jeff Folkins said:
I apologize if this was covered already.

So I am assuming the interpretation of these measurements is that at equivalent resolutions there is only about an approximate half stop difference between the D800 vs D4 in the dark regions of high ISO's. (is this the correct interpretation?)

Given this, why do the Nikon official "specs" for the D800 and D4 list the max ISO's as 26k vs 204k, a three stop difference?
Click to expand...
You need to normalize to same size to get that. At full res, D800 at 51K or 102K ISO would look very bad. The ISO ratings are a matter of market tolerance.

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
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