User Tag List

Thanks useful information Thanks useful information:  5
Results 1 to 16 of 16

Thread: Help with exposure and histogram

  1. #1
    Member MadMax1412's Avatar
    Join Date
    24 Aug 2010
    Location
    Newcastle
    Posts
    52
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Help with exposure and histogram

    Hi guys,

    I shoot with a Nikon D90 in JPG + RAW. When I post process (I'm still learning, so be gentle) I find I sometimes can't decide between a few versions of the same photo. I would like to know - what makes a good photo? Probably a very generic question, like "what's the best length for a piece of string?" and I'm sure the answer's going to be "Whatever makes you like it".

    Here's an example of what I mean.

    Original JPG photo from camera - histogram is around middle of window.
    The JPG file created by RawThereapee when RAW file is opened and Auto-Levels is set - histogram is pushed to the left
    The JPG file created if I manually adjust RT's automatic adjustments and darken it a bit - histogram is pushed more to the left.

    Personally I like the slightly darker one (the third link), but I'm doubting myself and thinking it's probably the worst one because the histogram is more to the left then centred.

    PS - I wanted to use the "Manage Attachment" options, but the dialogue box says I've used "5.19 MB of 488.3 KB Used" and I didn't want to delete any photos from there in case it meant they deleted from previous threads where they're used in, plus I'm not sure how to even delete them if I wanted to.

    Thanks in advance.
    Last edited by MadMax1412; 14-05-2016 at 11:02pm.

  2. #2
    Ausphotography Regular Brian500au's Avatar
    Join Date
    03 May 2010
    Location
    Melbourne
    Posts
    1,057
    Mentioned
    2 Post(s)
    Tagged
    0 Thread(s)
    I wonder if you are getting a false reading from your monitor - I would be interested to know why you prefer the last edit?

    Are you using a calibrated monitor when editing?

    On my monitor the SOOC JPG is the correct exposure - and both the second and third are under exposed by a stop or two.

    I normally take notice of the histogram when shooting but I cannot say I use it much when editing. I suppose I might use it as confirmation of what I am looking at but I think I can see if an image is over or under exposed. Also you have to remember with a histogram - you need to interpret it correctly. It is not necessarily giving you the correct exposure of your subject - more the average exposure of the entire scene including the back ground behind your chosen subject.

    You might find in this case the auto setting is being influenced by the bright back ground in this shot and when calculating is leaving your subjects a stop or two underexposed.
    www.kjbphotography.com.au

    1Dx, 5DsR, 200-400 f4L Ext, 100-400 f4.5-5.6L II, 70-300 f4-5.6L IS, 70-200 f2.8L IS II, 24-70 f2.8L II, 16-35 f4 IS, 11-24 f4L, 85 f1.2L II, 500 f4L IS, 300 f2.8 IS, ∑50 f1.4 A


  3. #3
    Administrator ricktas's Avatar
    Join Date
    24 Jun 2007
    Location
    Hobart
    Posts
    15,129
    Mentioned
    8 Post(s)
    Tagged
    1 Thread(s)
    One of the best articles I have ever read regarding histograms, that makes it easy to understand them a lot more : https://luminous-landscape.com/under...ng-histograms/
    "It is one thing to make a picture of what a person looks like, it is another thing to make a portrait of who they are" - Paul Caponigro

    Constructive Critique of my photographs is always appreciated
    Nikon, etc!

    RICK
    My Photography

  4. #4
    A royal pain in the bum! arthurking83's Avatar
    Join Date
    04 Jun 2006
    Location
    the worst house, in the best street
    Posts
    7,701
    Mentioned
    2 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by MadMax1412 View Post
    ..... I would like to know - what makes a good photo? Probably a very generic question, like "what's the best length for a piece of string?" and I'm sure the answer's going to be "Whatever makes you like it".

    .....
    Where do you start!

    Even on the same photo just processed in a couple of different ways you'd have arguments for two sides of the fence as to whether it's good/not good.
    Same with exposure. You could use two extremes of exposure for that same image as well, and you'd have opposing views as to which one looks better!

    As always, shoot whatever way makes the image look good to you!

    As for what Brian said: if you use exposure as a variable for referring to an image as either good/not good .. then be sure you are seeing(on your screen) what the histogram is showing you!
    The histogram doesn't lie .. even if two histograms show you different things for the same image(via two different programs)

    If you haven't yet installed or use Nikon's ViewNX2 or Capture NX-D, then I suggest you try it(or them).
    ViewNX is simpler, but sometimes too basic, and CNX-D is also similarly basic(as an editor) but has a few tools to help.

    But:
    If you open that raw file in VNX/CNXD, the one thing you will note is that both the jpg and NEF file will have close enough to the same rendering and hence histogram.
    There are very very slight differneces which you will see only with direct comparative views in Nikon's software.
    But that same NEF file in any other non Nikon software will be rendered differently(as you see in RT).
    So using RT, the rendering of the jpg will look different in RT compared to viewing the NEF file in RT.
    ie. the reasons for the difference between the camera's jpg and RT's default jpg rendering.

    The reasons for this is simple. Raw is not an image format, but a set of data that is manipulated by the raw program to produce an image.
    Each raw editor will define a different set of reference points for that data set .. so the result is that each raw program shows you different renderings for the same image.
    But it's the raw engine that's doing the work.
    That is, Adobe use their ACR as the raw engine, RT(and many other free raw programs) use DCRaw as the raw engine, Capture One uses it's own raw engine, Nikon's software uses it's own raw engine(in this case, it's the codec that Nikon issues for each camera).
    If all software makers used Nikon's codec, in theory all NEF images would be rendered the same way to begin with.

    The point is that if you use Nikon's software, then the raw file you see is the same(save for some very slight differences) as the jpgs out of the camera.

    Personally, the last darker image is way too dark on the skin tones, but the darkening has reduced the background brightness which is obviously over exposed(as shown in the histogram!)
    Remember the point that the histogram doesn't lie. And you can 'calibrate' your screen using a good histogram.
    The problem is that using a raw file and different editors, while the histogram isn't lying to you! .. they will be different simply because the translation from raw data to a usable image makes this a certainty! So the nett result for all that previous babbling is that you can't really use a histogram from a raw file to calibrate your monitor.
    When I say calibrate monitor here, obviously it's not a proper calibration .. just a rough estimate to help where a non calibrated monitor is almost certainly lying to you.

    From what I can gather, without being able to see the exif, I'd say you used some fill flash(in camera) in the jpg. It almost certainly looks a bit obvious judging by the exposure differences.
    The rendering in RT makes this slightly less obvious, and the darkened image in RT completely eliminates that look.
    The reason for using fill flash is to eliminate the exposure differences due to dynamic range issues .. and RT's dark image does this.
    The little girls face is too dark(for a little girl) and has just a bit too much of a shadowy rendering to it(on her LHS/ our right) compared to the left.

    So like Brian said, I think your screen may be not set to a proper brightness for you to see this.
    (ps. the womans face in the darkest image is wayy too dark .. much better in the brighter image)

    So! (if any of the above is anywhere close to correct) What I reckon you may want/need:

    If the background(and faces) is overly bright for 'ya then, if you haven't got VNX or CNX, I suggest you install it.
    I can't remember exactly how RT works, so I'll use VNX as the example coming .. and you can substitute CNX-D for VNX2 as well, and I'll highlight the differences needed.
    In VNX open the raw file of the jpg(that is the brighter image in this series).
    The faces are well exposed and show no blown highlights or lost shadows.
    In VNX with the image displayed, press the H key on the keyboard. Image will then look weird. What that does is to display lost highlights in the image. The black sections indicate no lost highlights, the bright coloured areas mean that of that colour(ie. red, green or blue) for that colour .. the channel has blown out .. value over 255.
    Note that in CNX-D that command will need to be Shift+H to do the same indicator thing!
    Going by the image of the bright jpg, if you do that in CNX-D or VNX2, almost certainly the sky will suddenly be rendered blue.
    Remember this is not what you've done to the image, it's just an easy way to show you blown out highlights!
    For shadows, you press S in VNX2 and Shift+S in CNX-D. To get rid of this display mode, just press that command again. and the image display will be restored normally.

    back to VNX2(but I'll recommend you to use CNX-D on this specific image!)
    With the raw file still displaying the blown highlights, go to the exposure slider and reduce the exposure by a very small amount. Don't over do it tho(more to come).
    All you want to try to do is to minimise the impact of the distracting background. Maybe up to -0.5Ev drop.
    Then look for the Picture Control tool. Choose Portrait style. This lowers contrast a little and softens colour just a small amount. it should help to reduce the look of the fill flash a little too(you can just make out a shadow on the young girls chin/neck) .. this should help to remove that.
    If the faces have become too bright(I doubt it with only -0.5Ev tho!!) .. you can always recover a little using the shadow recovery slider. But I doubt you'll need it.
    The reason I recommend you to use CNX-D instead of VNX2 is this next step:
    Only in CNX-D! .. In the tool pane, look for the Lens correction tool. It's a small tool icon next to the noise reduction tool marked NR! Open the Lens correction tool and scroll down to find the vignette tool within that area. It's could be unticked, but some Nikon cameras have vignette control enabled by default, so it may be already ticked with a setting of 80.
    What you want to do is to make sure it's ticked, but then use the slider to set it to -100.
    This vignette will help to reduce the brightness of the background(sky) a bit.
    Another reason to use CNX-D over VNX2 is that VNX doesn't have any levels tool, where CNX-D has. You could use the levels tool to drop the histogram's max value from 255, say to 253-250 which will help reduce the brightness of that background too.
    Nikon's implementation of the levels curves too is not the best in CNX-D as you can't directly input a value. and it's finicky to mess about with it using the mouse if you want exactness.
    Don't over do it with the levels tool lowering too .. it reduces contrast in the entire image.
    All you really want to try to do here is reduce the visual impact of the distracting highlights. They take away from the subject matter, which is why they're distracting(ie. the eye).

    My suspicion is that your screen may be set to render overly bright .. if it's not calibrated, so what you see in the darkest image is not what we see on our calibrated screens.
    if you cant' calibrate, I'd recommend that you get those three images printed ... even if only small sized at 6x4 or whatever.
    The important point is that a print is a calibration point! If the print looks good, then you can adjust your screen to mimic that print .. afterall that's the entire point of a calibrated screen!

    Anyhow .. sorry for the long reply. Probably misses the mark .. but hopefully it gives you an indication of what you think could be improved when you say "I'm doubting myself".
    The point is that you are your best judge as to what you like and don't like .. but! .. just be sure you're seeing your preferred choice accurately!
    Nikon D800E, D300, D70s
    {Nikon} -> 50/1.2 : 500/8(CPU'd) : 105/2.8VR Micro : 180/2.8ais : 105mm f/1.8ais : 24mm/2ais
    {Sigma}; ->10-20/4-5.6 : 50/1.4 : 12-24/4.5-5.6II : 150-600mm|S
    {Tamron}; -> 17-50/2.8 : 28-75/2.8 : 70-200/2.8 : 300/2.8 SP MF : 24-70/2.8VC


  5. #5
    Administrator ricktas's Avatar
    Join Date
    24 Jun 2007
    Location
    Hobart
    Posts
    15,129
    Mentioned
    8 Post(s)
    Tagged
    1 Thread(s)
    Quote Originally Posted by MadMax1412 View Post
    I would like to know - what makes a good photo? Probably a very generic question, like "what's the best length for a piece of string?" and I'm sure the answer's going to be "Whatever makes you like it".
    Outside the technical aspects of your question, this bit is important.

    A photo does not have to be technically well taken to be a good photo. Some of the best photos ever taken are not technically perfect. It is the connection with the subject. You could show me a photo of your daughter (if you have one), and I am likely to glance at the photo of the girl, and then notice things like lack of sharpness, etc. Show that same photo to the grandmother and she will love it. It is the connection that is most important. What that connection is between the subject and the viewer, tends to change the way we see a photo. I enjoy taking photos of fungi, and enjoy seeing others photos of them, but if you cannot see the point in photographing fungi and have no interest in it, no matter how technically good my photo is, you will still find it boring.

    Have an understanding of the technical of photography, but do not let that over-ride your creativity.

  6. #6
    Administrator ricktas's Avatar
    Join Date
    24 Jun 2007
    Location
    Hobart
    Posts
    15,129
    Mentioned
    8 Post(s)
    Tagged
    1 Thread(s)
    Quote Originally Posted by MadMax1412 View Post
    PS - I wanted to use the "Manage Attachment" options, but the dialogue box says I've used "5.19 MB of 488.3 KB Used" and I didn't want to delete any photos from there in case it meant they deleted from previous threads where they're used in, plus I'm not sure how to even delete them if I wanted to.

    Thanks in advance.
    Re this, I am not sure why you go that, there are no limits set on how many photos members can upload to the site.

    and back to the discssion. When you have over-exposed areas of a photo that you cannot recover (ie the sky in your photos), consider making the images mono (black and white).
    Last edited by ricktas; 15-05-2016 at 9:01am.

  7. #7
    Member
    Threadstarter
    MadMax1412's Avatar
    Join Date
    24 Aug 2010
    Location
    Newcastle
    Posts
    52
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Thanks for the replies.

    I didn't realise that different programs treated RAW files differently, so I'll install either VNX or CNXD or both.

    The photos were done in the shade of a tree at the top of a mountain. The sun was behind us, late in the day, so in other photos where I'm aiming downward, everything in the background is lit up and not in shadow. I did use a fill in flash.

    Note that in CNX-D that command will need to be Shift+H to do the same indicator thing!
    Going by the image of the bright jpg, if you do that in CNX-D or VNX2, almost certainly the sky will suddenly be rendered blue.
    Here's a screenshot of what it looked like when I went Shift+H in CNX-D

    On my monitor, it didn't appear blue. As you said, perhaps my monitor needs calibrating.


    For shadows, you press S in VNX2 and Shift+S in CNX-D
    - When I went Shift-S, the whole picture became white.

    Only in CNX-D! .. In the tool pane, look for the Lens correction tool. It's a small tool icon next to the noise reduction tool marked NR! Open the Lens correction tool and scroll down to find the vignette tool within that area. It's could be unticked, but some Nikon cameras have vignette control enabled by default, so it may be already ticked with a setting of 80.
    What you want to do is to make sure it's ticked, but then use the slider to set it to -100.
    This vignette will help to reduce the brightness of the background(sky) a bit.
    When I went into this, the bar was ranging from -100 to 200 and the arrow was 1/3rd from the left with the box displaying "0". I ticked the option and set it to -100.

    One weird thing with Capture NX-D. I navigated to my D Drive where my documents are, went to the directory and loaded the RAW file. When I went to "Convert File", it said I didn't have permission to save in this location and see the Administrator. I'm using Windows 10 and I'm the only user, so not sure what needs to be done. I ended up saving it to the desktop and then manually moving the file.


    As for the "Manage Attachments" part, I thought I could use that to show the photos in-line in the thread rather than have links to external sites such as Photobucket. What do most members prefer to see when reading threads?

    Anyway, thanks for all the replies. I'll stop using Raw Thereapee and start using the Nikon software. I think my biggest problem was knowing how to correct a photo where most is exposed correctly, but some is either under or over exposed.
    Last edited by MadMax1412; 15-05-2016 at 5:50pm.

  8. #8
    Member
    Threadstarter
    MadMax1412's Avatar
    Join Date
    24 Aug 2010
    Location
    Newcastle
    Posts
    52
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Sorry guys,

    I put the wrong link in the previous post and I can't edit the post.

    What I meant to say was:

    Here's a screenshot of what it looked like when I went Shift+H in CNX-D


    This is the photo after doing the suggested changes

  9. #9
    Loves The Wildlife. Mary Anne's Avatar
    Join Date
    19 Dec 2009
    Location
    Brisbane Southside.
    Posts
    32,844
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by MadMax1412 View Post
    As for the "Manage Attachments" part, I thought I could use that to show the photos in-line in the thread rather than have links to external sites such as Photobucket. What do most members prefer to see when reading threads?
    Some Members do not like clicking on links and prefer looking at the images on other Members posts.
    And there is no reason to use links that way when downloading from Photobucket, up the top of the Page here click on Library> *How do I * to find the correct way.

    This is what you do, on the right side of your photo on the Photobucket page under Share Photo left click on the last one the IMG TAG you will see it go yellow copied, then paste on here so easy
    That way there is no restrictions to size or kbs. Do press the enter key a couple of times between images so they do not look like they are joined together so you can add your text.
    I use flickr now, though If you look at this old Thread of mine from Photobucket you will see what I mean
    My 52/2011 Challenge

    I shoot with Canon And Olympus Cameras



  10. #10
    Member
    Threadstarter
    MadMax1412's Avatar
    Join Date
    24 Aug 2010
    Location
    Newcastle
    Posts
    52
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    This is what you do, on the right side of your photo on the Photobucket page under Share Photo left click on the last one the IMG TAG you will see it go yellow copied, then paste on here so easy
    That way there is no restrictions to size or kbs.
    Thanks. I had been coping the "Direct" link text. From now on, I'll use the IMG link. Thanks again.

  11. #11
    Member
    Threadstarter
    MadMax1412's Avatar
    Join Date
    24 Aug 2010
    Location
    Newcastle
    Posts
    52
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    I think I may have to give up on post-processing photos. It's one thing to know how to see what highlights or shadows are blown out, but it's another thing to know how to interpret what I see plus know what I need to do to correct it. I would need to spend quite a long time doing either a course or on-line tutorial to learn how to use the tools, but the big thing I think I would struggle with is to look at a photo and say "it needs this done (eg vignette adjustment), and that done" etc. I spend a few hours taking a couple of photos but then spend days/weeks trying to make them look better during the small amount of free time I have, so perhaps I should just accept the photos as they are.

    Here's an example. It's the JPG version straight from the Nikon D90. Most of the photo I'm happy with, but it would have been nice to have a bit of a blue sky.

    I opened up the RAW image in the Capture NX-D. I choose Shift-S to show shadows and the whole image goes white. Does this mean it's good as far as shadows?? I undo that and do a Shift-H to show highlights. The shoelaces and a bit of the dress near her waist goes purple, the majority of the photo is black, and the sky is white. What does this bit of purple (and a few spots of red) mean? What do I do if I see this? Is the majority being black a good thing? Based on Arthurking83 wrote above, the sky should be rendered blue, but it's still white. I try reducing exposure a little bit and at -1/3rd, I get a very tiny bit of aqua blue in the corner. Reducing it more gives me a band of black and purple with the majority still being white. This is what I mean by not knowing what I should be doing. Even if I went to a class, they could show me how to do something technically, but how do I learn what I should achieve at each step. I don't think I explained this properly.



    Anyway, thanks for all the replies. I appreciate the advice, but it's just bought home to me how complex post production is.

    Cheers
    Last edited by MadMax1412; 16-05-2016 at 2:33pm.

  12. #12
    Administrator ricktas's Avatar
    Join Date
    24 Jun 2007
    Location
    Hobart
    Posts
    15,129
    Mentioned
    8 Post(s)
    Tagged
    1 Thread(s)
    what you are experiencing is the limits of your camera's sensor. In that most modern (current model) DSLR sensors have a dynamic range of about 14-16 stops of light. The human eye has about 28 stops. ie, from the brightest part of the scene, to the darkest part there is a limit as to how much either way your camera sensor can deal with in one photo.

    You do not have to stop post processing your photos, you need to understand your gear more and its limitations. The issue is not post processing here, it is the original image. Stop using really bright background when you subject are in shade. Ask any wedding photographer what sort of day they would think was perfect for outdoor wedding photos..you will get the answer of overcast..every time.

    Post processing will not fix a badly chosen background for a portrait. You are blaming post processing when you should be blaming yourself for shooting in a situation that was beyond the ability of your sensor.

    Sorry if this is harsh, but you need to be told the truth. Post processing is not the problem.

  13. #13
    Member
    Threadstarter
    MadMax1412's Avatar
    Join Date
    24 Aug 2010
    Location
    Newcastle
    Posts
    52
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    This is a photo taken about 7 minutes after the previous one shown in this thread. It's the JPG as produced by the camera ie unaltered.



    Perhaps it was because I stepped back and there was more dark areas, but for some reason, the sky came out more blue.

    I do know I was changing the settings from portrait to auto etc to try to get a good shot. My main goal was to have the subject sharp and try to blur the leaves etc that were behind her a little bit so she stood out more.

    Looking at the embedded data between a shot like this one where the sky is bluish and others where the sky is white, the bluish ones were at F10 or F11 whereas the white sky ones were at F7 but with both the exposure time was 1/200th and ISO-200. The other difference was the focal length.

    The following 2 photos probably give a good example of just how much the focal length increased/decreased the colour of the sky with everything else being equal (F Stop, ISO, exposure time, etc).


    F11, 1/200 sec, ISO-200, Focal Length=65mm, max aperture 4.8


    F9, 1/200 sec, ISO-200, Focal Length=65mm, max aperture 4.8

    I never took into consideration how the sky would turn out. I was just focused (no pun intended) on trying to get the subject in focus, framed well in the shot (trying to keep in the 1/3rd rule) and trying to blur out the remaining parts of the photo so she "popped".

    So much to learn when it comes to taking photos :-)

    Whilst I'm here, I would just like to share a photo that probably isn't that great technically, but I love and sort of shows what I try to attempt (blurred background, subject off to the side, etc).


  14. #14
    A royal pain in the bum! arthurking83's Avatar
    Join Date
    04 Jun 2006
    Location
    the worst house, in the best street
    Posts
    7,701
    Mentioned
    2 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by MadMax1412 View Post
    I think I may have to give up on post-processing photos. .....
    I agree with Rick.
    No need to give up on PP, just because you CURRENTLY don't fully understand it.
    It takes some time .. which basically means experience.
    The more you do, the more it starts to make sense .. and hence the more natural it becomes.

    What (I find) difficult in terms of PP is switching between one program and another, and especially when the changes made in one program don't translate, carry over or do the same thing as they do in another app.

    ..... I opened up the RAW image in the Capture NX-D. I choose Shift-S to show shadows and the whole image goes white. Does this mean it's good as far as shadows?? I undo that and do a Shift-H to show highlights. The shoelaces and a bit of the dress near her waist goes purple, the majority of the photo is black, and the sky is white. What does this bit of purple (and a few spots of red) mean? What do I do if I see this? Is the majority being black a good thing?....
    Yes .. kind'a!
    Sometimes it's nice to have some 'blown out' shadows in specific areas(it means contrast!). Just as in real life, having some dark shadows looks real.
    (ps. one of my main pet hates about HDR, and why I prefer not to do them myself)
    eg. the young girl's jacket has what look like black sleeves. In this image they're rendered more a dark grey tone rather than black.
    Of course we don't know, so it's up to you to 'provide those details'.
    That doesn't mean tell us what colour/tone they are, but capture and PP the image to show us those details(ie. "provide those details").

    Now, at the risk of another long arduous reply(that will almost certainly bore you to distraction!)

    When you use the show lost highlights(or shadows too) .. it shows you on a per channel basis .. not just if the total highlights area is blown.
    The blown highlights are obvious, but it's also helpful to know which channel is blown too.
    That is, you have three colour channels(Red, Green, Blue). by default, VNX and CNX-D both show all three colours and the total(luminance) channels in the histogram.

    The reason I mentioned (guessed) blue was that it looked like only the blue channel was blown out, but obviously all three channels were looking at the Shift-H screen shot.
    So with the highlights tool, if you see all white, then this means that all three(RGB) channels have blown out.
    In some images you may see only green, or red or blue(where the white area is shown with Shift-H).
    This obviously means that if you see only one of those colours, then only that channel has blown.

    There's another side of the histogram highlights/shadows tool too tho. Something to be mindful of when assessing images this way.

    If you lose shadow detail in some colour channels, it shows you the opposite colour channel lost!
    So we know that the opposite colour channel of blue is yellow .. it's just something we know(this may be another aspect of PP to read up on, or just experiment with and gather your own experience! )
    So if you expose an image a touch too dark, and an area has lost blue shadow detail, using the Shift-S tool, what you may see, instead of black or white, you may see a yellow area.
    this is common on green areas where they have been exposed dark.
    If this area of dark green is still visible in the image(but is dark), obviously you can see it, so the green channel is OK, but that doesn't mean that the blue channel is as well.

    Remember that all colours in the real world usually are made of all three colours digital primary colours when captured with a digital capture device(which captures in RGB).
    So that blue thing you shot isn't just all blue, it may have a bit of green and a smaller amount of red in it's overall colour blend.

    Another thing to look out for if you use ViewNX2 and CNX-D.
    As you move your cursor across the image(not the screen!) .. only the image .. it displays the RGB value set in real time for that pixel.

    In ViewnX2 you see this information at the top of the image to the left of the [i] icon.
    What you see there is two sets of numbers and as an example, they may be something like 3200,1024(255, 200, 225).
    In CNX-D that similar info is located at the bottom of the page under the image to the RHS. Next to that is also a colour square.
    As you move the mouse cursor it shows you the (R,G,B) value and the actual colour of that pixel site.

    In VNX2, the meaning of those numbers are:
    the first two values are the pixel mapping, that is which pixel the RGB values are for in the actual pixel grid that is the sensor!
    The top most and most extreme left pixel is 0,0 .. so if you move your mouse cursor to the top left, you get close to 0,0. (note its hard to be accurate in pinpointing one specific pixel with accuracy .. unless your mouse is a uber high res type).
    The bottom RH corner is you maximum pixel grid value. So if you had a 6000x4000 pixel sensor(ie., 24Mp) the last value for that set on the RH lower corner will be 6000,4000.

    That info is handy but not as important! The next three values(in brackets separated by a comma) are the important values to take note of.
    Those values in brackets are the individual tone range for each of the three colour channels .. R,G,B remember.
    So in my example of (255,200, 255) simply indicates that Red channel is 255(and hence blown out!) Green channel is 200(a bit high) and Blue channel is 255, and hence also blown out.

    The range of values can only be between 0(blown shadow, and hence dark) and 255(blown highlight, and hence over bright).

    It works in tandem with the histogram and the shadows/highlight tool to show you more info.

    If we were to use this info on your other images of the young girl:
    eg. the one of her standing on the cairn.
    If you hover your mouse over the sky and set it up high on the darker blue parts, as an example of the RGB values you may see could be something like (50,128,100).
    A set of RGB values like that indicate a mid tone(128 is the mid point between 0 and 255!) There's little to no red, so red is low. Exposure looks good up high so green should be mid range. Blue is mid-dark, so value will be midrange-low.
    As you bring the mouse lower to the horizon, it gets brighter, and less colourful(ie. more white). All three RGB values should increase as you bring the mouse lower to the white horizon area!
    I'm guessing(important to note the guessing part, as I don't have your image on my PC to see it! ... and the jpg you uploaded won't be as accurate) ..
    I'm guessing as you lower the mouse pointer you may find high green(but maybe not blown out to 255), but you may eventually see 255 for the blue channel. Red could be anything, but certainly high!
    That is, as it's white(which means bright), RGB values will all be high(200+ .. most likely 220+)

    The beauty of all this is that it's constant process, there is nothign to guess about it all.
    Exposure sometimes involves a bit of guesswork, but analysing the exposure of the image is not.
    All those numbers and values all remain the same!
    You can obviously alter exposure and change brightness at a pixel level, but all those numbers and values change accordingly too.

    That is, tone values of 0-128 are mid to low brightness levels(0 is a lost brightness value .. ie too dark).
    128-255 are mid to high tone values, where 255 is lost brightness, ie. too bright).

    Note that there is technically nothing wrong with having an image too dark or too bright, and lost highlights and shadows. It's done in photography all the time and to good effect!
    What's important to be mindful of ... is it an appropriate way to show that image.
    That is up to an individual's taste, and mine is as valid as yours which is as valid as the next persons.
    You just have to be confident of two things.
    1. you're 100% certain that you're version of the image is PPed the way you want it to be.
    2. you're screen is showing you the way you think the image is actually displayed.

    1. is easy!
    2. is hard(or can be hard). As a test, use the brightness button on your screen to brighten the screen to maximum and darken it to minimum too)
    Does the image look the same, is it the way you want to show it?
    Those extremes of the screen is a good way to highlight why a calibrated screen is important
    anyhow .. hope some of that makes sense.

  15. #15
    Member
    Threadstarter
    MadMax1412's Avatar
    Join Date
    24 Aug 2010
    Location
    Newcastle
    Posts
    52
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Sorry for the delay in replying. Just a bit time-poor due to work.

    Now, at the risk of another long arduous reply(that will almost certainly bore you to distraction!)
    I appreciate the reply and wasn't bored.

    Thank you for explaining how to interpret the Shadows and Highlights (shift-S and shift-H). As I mentioned earlier, my problem is that it's one thing to interpret the data, but knowing what to do to correct it in PP is what I need to learn. In other words, knowing when to use vignette control, or the sharpening tool or when to use the hundreds of other controls found in PP software.

    Unfortunately with work, my shifts vary day-to-day and week by week so it's difficult to attend a regular class that might teach PP. If I found an online tutorial on how to use a certain package, I assume that the basics would apply to other packages, you'd just need to find the commands that do the same job.

    Appreciate the replies.

  16. #16
    A royal pain in the bum! arthurking83's Avatar
    Join Date
    04 Jun 2006
    Location
    the worst house, in the best street
    Posts
    7,701
    Mentioned
    2 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by MadMax1412 View Post
    ..... As I mentioned earlier, my problem is that it's one thing to interpret the data, but knowing what to do to correct it in PP is what I need to learn. In other words, knowing when to use vignette control, or the sharpening tool or when to use the hundreds of other controls found in PP software.

    Unfortunately with work, my shifts vary day-to-day and week by week so it's difficult to attend a regular class that might teach PP. If I found an online tutorial on how to use a certain package, I assume that the basics would apply to other packages, you'd just need to find the commands that do the same job.

    ......
    Same here with work .. some weeks I have nothing but time to play with, so I get out and get some more stuff .. the next week we have too much work and no time to play with the stuff I just got the previous week .. and I also forgotten that I even got the stuff to play with .. etc, etc.

    Anyhow, re PPing and highlight/shadow details and stuff.
    For me PPing was easy(to learn) .. long and arduous process, as I just clicked buttons and pushed/pulled sliders here and there and kind'a waited to see what would happen to the image by the end of it all.
    It was hard, in the sense that I knew nothing about the processes, so basically just played with as many options till I was satisfied that the image looked the way I wanted it too.

    One last thing: don't get too hung up on having a perfect looking histogram. The idea of it is a 'theoretical ideal', rather than a real world notion of perfection.
    This ideal prefect bell curve histogram can make images look flat and dull in some situations.
    But you're screen needs to be calibrated to have the confidence that WYSIWYG in the image if you do ignore the histogram.

    In terms of using some of the premises from one editing program and translating those editing abilities to other programs .. it's hit and miss. Generally hit! .. but sometimes it can miss.
    The beauty of it all tho is that you'll learn more about the entire process of editing when it misses, as it teaches you what not to do.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •