The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

D800 v D800E, chapter 7 (or is it 1007?)

H

HiroshiS

Guest
I know this has been discussed to death but I am still undecided on whether I should eventually “upgrade” from my D800 to the D800E - if the D800E ever becomes more widely available. Since Jack’s observations on the difference (or lack thereof) in image quality between the D800 and D800E have been posted all over the internet, I thought I should join this forum and post this question here to the real pros.
To help with my decision, I had previously downloaded Rob Galbraith’s images and processed them with two different raw processors (Aperture and Capture One). This basically confirmed Jack’s observation that with a little extra sharpening the D800 file becomes virtually indistinguishable from the D800E version.
However, today I downloaded the trial version of DxO Pro and there the difference between the two cameras was much more striking. I processed both files with DxO Pro using their default settings with their modules and didn’t apply any extra processing. Whereas the D800 file is very much comparable to the D800/E versions from Aperture or Capture One, the D800E image looks much crisper and three-dimensional. What do the experts think? Does DxO Pro reveal a real difference or can this again all be equalized by different amounts of sharpening or post-processing? I tried to play around with the DxO-processed images in Photoshop but I wasn’t able to bring the quality of the D800 file up to the D800E level but then I am not a sharpening expert. Since there are subtle color differences between the different processors I only attach the DxO comparison but again I would say that the D800 file processed with DxO is from a sharpness-point-of-view very much like the D800/D800E versions from the other two processors.
I am looking forward to reading all your comments…
 

tjv

Active member
Going from your examples, differences seem minimal to say the least. In fact, other than Nikon's official sample comparisons and perhaps those posted on LL, the decision seems to almost come down to the price difference alone. Good technique (when taking the shot) and processing (which is a subjective, creative set of choices) will most likely level any playing field, if any leveling even needs to take place...
 
H

HiroshiS

Guest
Actually, I just noticed that the attached jpg picture was down sampled, which makes it very hard to see any differences. I have now tried to upload tiff or pdf files, but they "exceed my quota"... How can I display this file in its original size?
 

Bob

Administrator
Staff member
My opinion is that more information is better than less.
First the anti-aliasing filter of the D800 is refreshingly weak in comparison to many others I have seen in the past, but it is still causing blur of about a pixel's width or so in two dimensions. The D800e has a theoretical resolution benefit that no amount of sharpening can erase. OTOH, processed aggressively with micro-contrast enhancement, the D800 files can give the appearance of images that appear just about as sharp as the D800e BUT that is APPEARANCE only since sharpening adds no information in fact it reduces it to some degree. The difference observed would be greater if the lenses were a bit better, since high frequency roll-off is part of the sharpness issue.
Now here is the interesting part. Since we render images to screen or to print, there are other factors involved in the image delivered to our eyes. In particular, printing produces a major perturbation since the pixels are re-sampled and transformed and dithered to deliver the image on-paper from a smaller repertoire of colored inks. The printing process itself will cause some of these differences to disappear in most cases. Even displaying a 1:1 image in an lcd screen reduces micro contrast due to bleed and flare within the lcd itself.
So when might the difference be noticeable? Mainly that might occur if the image needed to be up-rezed for printing. I expect that the d800e files might generate a sharper looking print than the d800, but only if up-rezed by a factor of 2 or more. I do not think that there will be any appreciable difference on a finished print unless those were the conditions. Since full frame images, so up-rexed will produce larger prints which to do them justice need to be viewed from a greater distance, that small advantage evaporates UNLESS, the image is heavily cropped and up-rezed where the difference may become more visible.
A key part of the imaging chain is our eyes. We accomodate to shifts in white balance and we perceive resolution in a way such that sharpening and micro-contrast enhancement very effectively fools us into "seeing" more resolution than actually exists once edges have been enhanced.
In the end, does the d800e file contain a bit more information? Yes it does if you use the right lenses. Will it be visible in print or on screen? Usually not for an appropriately processed file, but there are a few scenarios when it might.
So the comparison becomes an issue more of ultimate sharpness and information content vs the nuisance of dealing with moire more often. At least Nikon has given us the choice.
-bob
 
Last edited:

Bob

Administrator
Staff member
Actually, I just noticed that the attached jpg picture was down sampled, which makes it very hard to see any differences. I have now tried to upload tiff or pdf files, but they "exceed my quota"... How can I display this file in its original size?
If you can host the images somewhere where they are accessible by url, then use the little picture tool when posting and paste the url in the dialog box when it appears.
-bob
 

Tektrader

Member
I couldnt see much of a difference at all. There is a small difference in exposure. The 800e is slightly brighter.

Based on Jacks examination of the pair. I changed my order back to a D800 from an 800e
 
D

davexl

Guest
The good news is that you are already playing with the best publicly available RAW files I have found to compare.

I think the hassle of selling and switching, and the likely loss exceeds the benefits. That money would be far better spent on upgrading to better glass.

Also, if your current D800 has no AF issues, HANG ONTO IT ;-)

In my experience the biggest challenge with the D800E is getting the focus perfect - which is usually a much bigger impact than D800 vs D800E.

(I say that as a happy D800E owner)
 

Guy Mancuso

Administrator, Instructor
It might look sharper, but not nicer.
And that's the issue, you can easily start killing the tonal range of a image by keep punching up the sharpness beyond what is useful. Things start looking crunchy and digital looking. One of the biggest issues many folks do is go for ultimate sharpness by over doing the sharpening.
 

Steen

Senior Subscriber Member

By now we know there is an apparent difference between the two sensors.
The blurring filtration on one of them, and the lack of blurring filtration on the other, is clearly visible on screen in the untouched capture.

We also know that with some skilled post processing you can compensate for the blur to such an extent that the difference is nearly no longer visible.

To find out whether the difference between the two sensors was sufficiently important for me I simply bought both camera versions and shot them side by side, sometimes in a controlled manner i.e. with the same settings and the same subjects.
(I then just sold off the regular model later on even without any substantial loss since there had been a price increase in the meantime).

First I compared the captures without any post processing at all.
I chose to also shoot some direct jpegs to avoid running the captures through any kind of software.

After that I experimented with the RAW files trying to process them to their best rendering, and I do agree that you can make them come very, very close.
So close that you might not be able to identify them correctly in a blind test (though personally I'm not that good at post processing).

Like Bob points out the discussion, and the conclusions, and the choice is even further complicated because it all depends on whether you look at the result as a full frame fit to a screen, a 100 % crop on a screen, a print to a specific print size, or a full size print at a specific resolution like 240, 300 or 360 dpi.

All in all I think you have to try both versions yourself either by borrowing the E from a Nikon colleague in your region or by trying it out in a store side by side to your own camera and by simultaneously shooting jpeg as well as RAW.

In that way you will rather quickly be able to judge what is due to the sensor and what is due to your best post processing and what it all means to your final results.

I very quickly came to the conclusion that I wanted to keep the E version.
Because when I can clearly perceive the blurring effects from a blurring filter I simply find it annoying that it is there, and I spend far too much time looking for the effects even though I have been taught the filter is there for a good reason in the first place.
So I guess you could say it is to some extent also a psychological thing.
 

Steen

Senior Subscriber Member

And that's the issue, you can easily start killing the tonal range of a image by keep punching up the sharpness beyond what is useful. Things start looking crunchy and digital looking. One of the biggest issues many folks do is go for ultimate sharpness by over doing the sharpening.

With the E version I have for now dialed the in-camera capture sharpening down from 3 to 2 in the Standard Picture Control mode and with this modest setting I even do not need to apply any output sharpening or clarity at all when using Capture NX2.

So if I get the exposure and composition right I don't need to tweak the files at all, I can just convert them to Tiff or jpeg according to the purpose (of course resizing if necessary). This is really new to me, such a luxury.
 

D&A

Well-known member
And that's the issue, you can easily start killing the tonal range of a image by keep punching up the sharpness beyond what is useful. Things start looking crunchy and digital looking. One of the biggest issues many folks do is go for ultimate sharpness by over doing the sharpening.
+1 Exactly!....and this is not new to just the D800/D800E cameras and their files, but almost from the earliest days of digital, ever since people had the ability to keep adding sharpening to their images. It often is bad enough on web viewed images, but in my opinion, often destroying the rendering of fine detail and as Guy pointed out the tonal range, in those files selected to be printed.

It's almost akin to the philosophy of "if a little bit is good, a lot must be better" as opposed to being judicious in the amount of seasoning you use and thereby "a little goes a long way" :)

Dave (D&A)
 

D&A

Well-known member
My opinion is that more information is better than less.
First the anti-aliasing filter of the D800 is refreshingly weak in comparison to many others I have seen in the past, but it is still causing blur of about a pixel's width or so in two dimensions. The D800e has a theoretical resolution benefit that no amount of sharpening can erase. OTOH, processed aggressively with micro-contrast enhancement, the D800 files can give the appearance of images that appear just about as sharp as the D800e BUT that is APPEARANCE only since sharpening adds no information in fact it reduces it to some degree. The difference observed would be greater if the lenses were a bit better, since high frequency roll-off is part of the sharpness issue.
Now here is the interesting part. Since we render images to screen or to print, there are other factors involved in the image delivered to our eyes. In particular, printing produces a major perturbation since the pixels are re-sampled and transformed and dithered to deliver the image on-paper from a smaller repertoire of colored inks. The printing process itself will cause some of these differences to disappear in most cases. Even displaying a 1:1 image in an lcd screen reduces micro contrast due to bleed and flare within the lcd itself.
So when might the difference be noticeable? Mainly that might occur if the image needed to be up-rezed for printing. I expect that the d800e files might generate a sharper looking print than the d800, but only if up-rezed by a factor of 2 or more. I do not think that there will be any appreciable difference on a finished print unless those were the conditions. Since full frame images, so up-rexed will produce larger prints which to do them justice need to be viewed from a greater distance, that small advantage evaporates UNLESS, the image is heavily cropped and up-rezed where the difference may become more visible.
A key part of the imaging chain is our eyes. We accomodate to shifts in white balance and we perceive resolution in a way such that sharpening and micro-contrast enhancement very effectively fools us into "seeing" more resolution than actually exists once edges have been enhanced.
In the end, does the d800e file contain a bit more information? Yes it does if you use the right lenses. Will it be visible in print or on screen? Usually not for an appropriately processed file, but there are a few scenarios when it might.
So the comparison becomes an issue more of ultimate sharpness and information content vs the nuisance of dealing with moire more often. At least Nikon has given us the choice.
-bob
Bob, extremely well said (and explained). I've often have observed much the same in a wide variety of comparisons of similar or near identical files from the D800 and D800E. Both output and how the files are subsequently handled, has much to do with what differences, perceived or real, are observed between images from both these cameras.

Dave (D&A)
 
H

HiroshiS

Guest
Thanks for all your comments so far!
Here's a link to my little comparison (full size). Please note that you most likely need to download it to see the jpeg in full size:

https://www.dropbox.com/s/w7qkuiboq49dlk2/D800_E_Comp.jpg

To me the difference between the two is pretty striking, especially if you look at them side by side. I will now also try to print them to see whether the difference is visible on paper.
 

Jan Brittenson

Senior Subscriber Member
That test scene is very high contrast, midday sun. In that scenario with a really good lens there is not much difference.

However, go shoot in flat evening light and look at texture in low-contrast dark areas... The D800E is in a completely different league.

Or use a lens with a really nice rendering, but perhaps not super detail. It will deconvolve better because the D800E captures the lens blur more accurately. The D800 doesn't. As a result you can improve sharpness from some lenses very nicely in post.

Finally, for long exposures and high ISO work it requires less sharpening to begin with, which reduces noise levels.
 
Top