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!

Nikon D800 and LR Color Calibration?

glenerrolrd

Workshop Member
OK keep in mind I use LR almost exclusively and what I know about NX2 I learned yesterday on the Lynda tapes . I have no reference point to C1 other than they seem to produce really good profiles that create a finished product look .

To understand color bias you need a scene that includes RGB and varies in tones . It also helps if you have skin tones as they pick up even a sight variation in tint . To understand the color bias WB needs to be set for the light you are viewing ....not to an arbitrary Kevin (thats fine for a starting point in color balancing but not as a reference point). Lloyd showed in his tests that cameras vary by more than 10 percent in reported kelvin . (he matched the files and compared the reported kelvin).

This is why contrary to conventional wisdom ..you can learn from the AWB . (You can always go back and set the temp and tint to daylight or take a reading off a grey card in post ). AWB on my D800E typically gives me a reading of 4600-4800 and tint of 15-20magenta . This reduces the yellow red bias and increases the saturation of the blues .

If I run a raw NEF file into NX2 with a full spectrum of color and use AWB ......I get almost perfect color with no adjustments . I am using the standard settings in NX2 and I believe the full RGB color space .

If I take the same file into LR4 and use the adobe std profile .. I have the over saturated yellow red bias . The custom profile based on a color checker looks almost the same . Skin tones are not awful as a slightly warm bias is ok if the exposure on the face is right . (taste dictates). I am sure there is a technical reason for this . There are plenty of ways to work around this but they all seem to compromise the overall accuracy of the reference color chart . To use presets I favor getting a neutral color chart first which is what the std camera calibration is supposed to do. For now we can t fix this we can only adjust for the bias based on individual color perspectives .

LR combination of Clarity (for mid tone contrast) and sharpening (deconvolution ) is as good as I need . Add in the ability to use local area adjustment brushes to apply the fine tuning .. This is pretty great . I use LLyod s settings to start 50.0.7,70,20 but found with the S2 that having different presets by ISO is better ..you can hammer a ISO200 file but not so much one at ISO1600.

NX2 has significant complaints about its sharpening algorithms being last generation . They aren t as effective and require a lot more skill to optimize. But NX2 really does produce a more detailed rendering with better tone separation than LR ...I see more depth in the images immediately . In fact a D800E file processed in NX2 is stunning . The question here on sharpening is do you forgo sharpening in Nx2 and only sharpen in LR or PS . (we could do an entire topic on sharpening as we are now dealing with vey large files that are often resampled down for output ..this seems to be best accomplished in stages ..but the LR team says they have this covered in their process).

We are back to using a “best of breed” approach . Pick the raw conversion software of choice . Process to a TIFF . Then finish with PS,LR . This leads to a whole rehashing of optimum workflow .
 

ustein

Contributing Editor
>If I take the same file into LR4 and use the adobe std profile

Did you try the Camera Standard profile. It actually is not only colors also the contrast is different in Adobe Standard.
 

kuau

Workshop Member
And like I mentioned before I believe that NX2 by default does not apply any tone curve, but if you have D-Lighting turned on in the camera this will affect the tone curve. I confirm this later by reprocessing my image I posted earlier and turn of D-Lighing in NX2 and compare again against my Lr4 version which again by default applies a medium contrast tone curve, hence the reason why I leave it on linear. There is no one tone curve that works fort all images. So I find starting with a linear curve a better place to start with.

Om regards to a workflow that goes like NX2, sharpening 0 then bring in the 16bit tiff into LR4 or CS5/6 I am not sure this is optimal. I have read in a few different places the advantage of doing capture sharpening in the raw conversion process as opposed to doing it after the fact. Again we could be splitting hairs here and in the final print you would probably never see the difference.

My journey is to really establish whether NX2 does in fact do a noticeable better job converting D800/E files then does LR4/ACR. For me I cannot make that statement yet.
Personally I would like to stay with LR4 because I like the print module and soft proofing now. Adobe's version of "Camera Standard" I think is very close in terms of color to NX2, again we are talking about pleasing color and not exact color match.

Steven
 

kuau

Workshop Member
OK so here is what I was talking about.
I turned off Active D-Lighting in NX2, and here is the result compares to LR4,
To me they look very similar now again except the sharpening which the LR4 adds some contrast whereas the NX2 looks a tad flatter

LR4



New NX2


Steven
 

kuau

Workshop Member
So my initial analysis is the following,
In terms of color, Adobe did in fact do a pretty good job in "emulating" Nikons "Camera Standard" at least with this image which I realize is mostly all GREEN lol..
With Active D light turned off, tonal separation to me is a wash. Again this all has to do with what tone curve you choose, or in my case linear = no tone curve applied.
I can also use the shadow tool in LR4 to mimic what Active D-Lighting does in NX2.

So for me, at least for now, maybe I will stick with LR4, and leave NX2 in the box.

Steven
 

glenerrolrd

Workshop Member
The color issue will work itself out overtime and the tradeoffs in approach have been covered . Can t argue that LR s advantages look to win out over Nx2 ..which is awful to someone that has to learn it coming from LR.

Not sure why this is occurring but on my low resolution MacBook Pro 1440x 900 ...the two images look close enough and really neither looks that sharp . However on my new iPad 2088x ... they look much better and I can see differences in the detail rendering without looking hard at all. The NX2 shows much more detail everywhere ?

I realize you have to match the output file size to the intended viewing screen size or the system throws away detail (Correct?) .
 

danielmoore

New member
I think it's even simpler than that. My laptop is a 1920x1200 pixel screen, my desktop is the same. The laptop is 17", the desktop 24". Kuau's two above images look very similar on the desktop and show a dramatic difference on the laptop. Nothing being thrown away, just incapable of being resolved.
 

danielmoore

New member
I've had a look around the net for a sharpening plugin and none of the usual suspects will play nice with NX2. Does anyone have any ideas on that? Without going too far off topic, I find the default sharpening in C1 to be so subtle and effective and the product so damned good looking that it's going to mean NX2 output tif into C1 for now.
 

Guy Mancuso

Administrator, Instructor
Guy I will send it in a few hours
Take your time.

Reading about monitors here. I use a NEC wide gamut monitor which i highly recommend they have them in various size but very much like the Eizo brands. Check B&H on them. Happy to help if folks need it on which one.

Roger check your PM. Like to give you some money. ROTFLMAO
 

glenerrolrd

Workshop Member
I think it's even simpler than that. My laptop is a 1920x1200 pixel screen, my desktop is the same. The laptop is 17", the desktop 24". Kuau's two above images look very similar on the desktop and show a dramatic difference on the laptop. Nothing being thrown away, just incapable of being resolved.
Not sure I follow your logic . On my iPad which resolves more detail than my macbook ...I can see a clear difference between the two images . The NX2 file shows more detail than the LR file . On the MB they appear the same .

The discussion is about whether Nx2 actually produces a better (more detailed ) file than LR . What did your test show?
 

kuau

Workshop Member
I think Daniel is right, when I open both of the images on my Desktop system, the 16bit tiffs out of both LR4 and NX2, to my eyes the LR4 is sharper because of the decon. sharpening I use in LR4, NX2 only offers basic USM, which I am still waiting to here back from Jason O'Dell the NX2 "expert" on what he is doing for sharpening in NX2 and D800/E files....
 

danielmoore

New member
Roger, I was commenting on the fact that they appear the same since one screen is simply incapable of showing detail beyond it's pixel frequency. Seems I missed your point.
The NX2 product did at first seem to have more detail but it's the quality of that detail I have issue with, at the pixel level it's grotesque compared to all other programs output. It has a mindless mottled look that breaks up smoothness in tones. In print the NX2 file may not suck so bad, I don't know, all of my investigation has been screen only at this point. I read that deconvolution sharpening in Lightroom only kicks in after setting the detail slider to 100%, resorting to USM below that. Can someone confirm this?
 

glenerrolrd

Workshop Member
Daniel That was my point ....on my macbook pro ..the level of detail provided exceeds the screens ability to show . But on the IPad with a higher pixel count ...I could easily see the NX2 file had more detail . Steve indicates he sees the opposite on his screen at 100% .



At this point this discussion has out lasted my technical knowledge . There are many factors at play in completing a test that can be considered valid . It becomes an endless looping of what s a fair comparison ..screen,print ..processed to final ,original conversion etc. Probably best for me to quit until I have time to really work on my own testing .

As Jack showed with the D800 vs the D800E ..post processing knowledge and skills can minimize the differences between two cameras with very different “original output “ . Its about getting satisfactory (for the photographers use) image quality with a reasonable level of effort . This is where I believe LR is letting us down because it should be able to render a D800E raw file to the color chart . Today to my eye it doesn t .
 
Top