D3 versus D2Xs in studio (comparo shots)

[Ottrott's Human]

[DISCLAIMER] I'll premise this by saying that I Do Not have a scientific mind. Also I tend to frequently overlook the glaringly obvious!!

These types of tests are very difficult for me and so there may be glaring errors that would invalidate any results.

Please feel free to point out ways in which I can make this a more accurate comparison.

I do not have an L Bracket for my D3 yet so I didn't use a tripod which made getting the framing identical tough. I got them as close as I could without taking all day to get it done.

So I set up Woody's old friend Bridget under a single Norman ML600 which was shot through a 36" Octabox with a PVC/White Felt reflector opposite.

Light was metered with a Sekonic and WB was set in NX via a Getrag Macbeth Gray Card. I see that the D3 images appear warmer, but i don't know what to attribute this too since they were adjusted via the same method of WB in post. Any ideas?

Both camera's were set to the D2Xs color mode II.

Files were opened in NX, handed off to CS3 where the following steps were performed:
1. Resized to 900 pixels on the long side
2. Minor levels adjustments on the highlight side
3. USM at 10% and 50 pixels (a defogging action)
4. Smart Sharpen at 35% and .3 pixels (twice)
5. Converted to sRGB and Saved As jpeg

I shot the D2Xs set to the exposure as measured by the Sekonic which resulted in exposure that required a minor levels tweak to brighten the image on the highlights side.

The D3 images taken with the camera set as metered by the Sekonic (f.16) were slightly blown on the highlights side. Fortunately I caught this on the LCD (though the camera's histo read correctly) and stopped the D3 down to f/20 for the shots you see here.

This on-the-fly-adjustment resulted in a D3 histo, when viewed in CS3, that was fairly close to the histo of the D2Xs and like the D2Xs images required a slight levels adjustment in order to make contact with the highlight data.

D2Xs with 28-70 2.8 (1/125s f/11.0 ISO100 50mm):

Subscribe to see EXIF info for this image (if available)

D3 with 28-70 2.8 (1/125s f/20.0 ISO200 70mm):

Subscribe to see EXIF info for this image (if available)

D2Xs with 24-70 2.8 (1/125s f/11.0 ISO100 48mm):

Subscribe to see EXIF info for this image (if available)

D3 with 24-70 2.8 (1/125s f/20.0 ISO200 70mm)

Subscribe to see EXIF info for this image (if available)

 

[SMH77]

Unfortunately, stopping down this far on the D3 shots will be putting the D2Xs at an advantage to begin with since f/20 is in a worse aperture range for either lens than f/11 is. I think a better approach would have been to alter the shutter speed to correct for exposure.

Also, what 'picture style' are you using on the D3? If you're using 'standard' it seems to automatically do some form of levels and saturation adjustment from my experience--which would probably explain the 'warmer' colors that you're seeing. You may want to compare the 'neutral' picture style on the D3 and tweak a little from there using levels in post to get a better comparison. Just my thoughts...

Sean

 

[Czechman01]

Unfortunately, stopping down this far on the D3 shots will be putting the D2Xs at an advantage to begin with since f/20 is in a worse aperture range for either lens than f/11 is. I think a better approach would have been to alter the shutter speed to correct for exposure.

I agree... I'd try to get something around f8 which is generally the sweet spot for most lenses and definitely closer to something reasonable rather than f20. Try natural light and a tripod. Unless the shutter speed winds up over 20 seconds or something equally ridiculous it's irrelavent.

However, I'd be amazed if there was THAT much difference.

 

[Beezle]

Don't you need real human skin to evaluate?

Seems to me the only thing you can get from a mannequin is how shadows fall.

 

[Ottrott's Human]

Unfortunately, stopping down this far on the D3 shots will be putting the D2Xs at an advantage to begin with since f/20 is in a worse aperture range for either lens than f/11 is. I think a better approach would have been to alter the shutter speed to correct for exposure.

Also, what 'picture style' are you using on the D3? If you're using 'standard' it seems to automatically do some form of levels and saturation adjustment from my experience--which would probably explain the 'warmer' colors that you're seeing. You may want to compare the 'neutral' picture style on the D3 and tweak a little from there using levels in post to get a better comparison. Just my thoughts...

Sean

Good point Sean. I'll reshoot at the same aperture. See......glaringly obvious.

My OP does state that the color profiles were D2Xs Mode II for both cameras.

I agree... I'd try to get something around f8 which is generally the sweet spot for most lenses and definitely closer to something reasonable rather than f20. Try natural light and a tripod. Unless the shutter speed winds up over 20 seconds or something equally ridiculous it's irrelavent.

However, I'd be amazed if there was THAT much difference.

As stated in the OP I don't have my RRS plate for the D3 yet. I'll see if I have another plate that will work for the time being.

Don't you need real human skin to evaluate?

Seems to me the only thing you can get from a mannequin is how shadows fall.

I think we need this kind of study as well as a real skin study. My only available model today is 7 and doesn't hold still long enough for this kind of test.:smile:

I believe this test is useful for gauging differences in DR among other things though.

 

[Czechman01]

Don't you need real human skin to evaluate?Seems to me the only thing you can get from a mannequin is how shadows fall.

The ballyhoo over this camera is more about DR and detail, not skin tone which is a subject for discussion at some other time.

One thing at a time, please.

 

[SMH77]

My OP does state that the color profiles were D2Xs Mode II for both cameras.

Oops, I missed that one. Sorry. I haven't checked out the "D2XModes" on the D3 yet. Do you find them to be similar to what you actually get from the D2Xs? In this test, it doesn't look that way. Try using 'neutral'--just to see what it yields, if nothing else.

The D3 is definitely a different animal than previous Nikon bodies--it's gonna take some learning.

I can sympathize with trying to work with the 'other models'--mine don't like to cooperate either. :biggrin:

It'll be interesting to see how this thread develops...


Sean

 

[Beezle]

The ballyhoo over this camera is more about DR and detail, not skin tone which is a subject for discussion at some other time.

One thing at a time, please.

Okay. I thought the differences being noted were more about color.