Results 1 to 8 of 8

Thread: ETTR + uniWB + DPP = ??? (Newbie guide?)

  1. #1
    Junior Member
    Join Date
    Dec 2008
    Posts
    14

    ETTR + uniWB + DPP = ??? (Newbie guide?)



    OK... long time lurker, first time poster (be gentle)...


    I've been following the posts about exposure, ETTR, uniWB (never heard of that till 2 weeks ago!) with some interest and thought I'd see if I could give this a go... but I can't find a guide for numpties/newbies that bootstraps you...


    What I have to work with : 400D, 18-55IS (kit lens from a 450D), DPP and Photoshop Elements 2! So my computer equipment is limited...


    This is what I have done so far, what have I missed, what should be I be looking for...


    1. Took a shot of my lens cap, set that as the WB.
    2. Took lens cap off(!). Took a picture. It was all green in on the preview.
    3. Reviewed the picture in camera with the RGB histogram.
    4. Checked that none of the channels were clipping to the right, increased exposure, took picture
    5. Iterated till clipping, took one step back.
    6. Clicked away like a mad loon (sorry, carefully considered my composition and took a perfect shot)
    7. In DPP opened the picture(s) and set the WB using temperature (5000k) ... should I have taken a picture of something white before step 6 to help set the WB?
    8. Adjusted the brightness down, pulling the histogram away from the right.
    9. Saved out as jpeg 10 quality
    10. What now...?

    <div>I took a couple of autowb shots of the same scene and the one produced above seem better... what should I expect to see? The shots should have less noise? The shot should be sharper? In PSE(2) how far should I zoom in check the shots and what should I be doing (levels, pulling the mid point back?)</div>

  2. #2
    Senior Member
    Join Date
    Dec 2008
    Location
    Vancouver, Washington, USA
    Posts
    1,956

    Re: ETTR + uniWB + DPP = ??? (Newbie guide?)



    Quote Originally Posted by uberbloke
    should I have taken a picture of something white before step 6 to help set the WB?

    Yes.


    Quote Originally Posted by uberbloke
    what should I expect to see? The shots should have less noise?

    Yes. Just less noise. Less noise means more dynamic range and higher color depth.


    Quote Originally Posted by uberbloke
    The shot should be sharper?

    No.


    Quote Originally Posted by uberbloke
    n PSE(2) how far should I zoom in check the shots and what should I be doing (levels, pulling the mid point back?)

    I do it at "fit on screen" zoom.

  3. #3

    Re: ETTR + uniWB + DPP = ??? (Newbie guide?)



    Daniel,


    I recently read the thread on RAW histograms, and now this thread gives me a new thought. I'm typically out shooting my daughter on the soccer field. I'm now thinking that my set-up should include two additional shots, first with the cap on to set WB so the histograms are more accurate. The second should be on my handy-dandy grey card so I can "set" the WB in post. Given that I have a 30D, what other settings in Picture Styles should I set to get the most accurate RAW histogram?


    I generally go manual on sunny days for the consistency. I go Av when the light is changing, but I don't like the wildly (to me at least) varying exposures I get. I'm also going to try spot metering the next time I'm in Av to see if the exposures are less affected by how much sky I have in the back ground. Any thoughts or suggestions?


    Thanks


    David

  4. #4
    Senior Member
    Join Date
    Dec 2008
    Location
    Vancouver, Washington, USA
    Posts
    1,956

    Re: ETTR + uniWB + DPP = ??? (Newbie guide?)



    Quote Originally Posted by dmckinny
    Given that I have a 30D, what other settings in Picture Styles should I set to get the most accurate RAW histogram?

    I'm not certain, but I would suggest the "neutral" profile, with 0 saturation and 0 contrast. I don't know if AdobeRGB is better than sRGB on that camera or not.


    Quote Originally Posted by dmckinny
    I generally go manual on sunny days for the
    consistency. I go Av when the light is changing, but I don't like the
    wildly (to me at least) varying exposures I get. I'm also going to try
    spot metering the next time I'm in Av to see if the exposures are less
    affected by how much sky I have in the back ground. Any thoughts or
    suggestions?

    I like to use Av in rapidly changing light as well. Use exposure compensation to do ETTR in Av mode.

  5. #5
    Senior Member
    Join Date
    Dec 2008
    Posts
    505

    Re: ETTR + uniWB + DPP = ??? (Newbie guide?)



    Quote Originally Posted by Daniel Browning
    I'm not certain, but I would suggest the "neutral" profile, with 0 saturation and 0 contrast.

    Daniel, I'm having a real hard time with the "I'm not certain" part. LOL []


    I went back to the Luminous Landscape and read up on the article "Settings for a AccurateHistogram". I realize now what the reduced contrast setting actually does. It "compresses" the historgram. Because the contrast is set to minimum...( that's -4 on my 5D) you see more of the histogramon the LCDthen you would with the contrast turned up. Just try it...it's really cool. It has no effect on the raw file at all. The LCD image looks washed out but that's why it works so well...I think. Now , just to put it to use and see if it practically makes a difference.



  6. #6
    Senior Member
    Join Date
    Dec 2008
    Location
    Vancouver, Washington, USA
    Posts
    1,956

    Re: ETTR + uniWB + DPP = ??? (Newbie guide?)



    The thing is that we don't want contrast reduced too much. The ideal raw histogram has only the gamma curve applied (what Photoshop incorrect calls "linear"), no extra tone curve. If you reduce contrast (inverse "S" tone curve), you make it seem like values are lower than they actually are, and it's even possible to have clipping hidden by it. The reverse is obviously bad (normal contrast-enhancing "S" curve), because it shows clipped highlights where there are none, and makes values higher than they actually are.


    On Nikon, -2 or -3 contrast is the point where contrast is actually neutralized, and you're getting just (or mostly) the gamma curve. But on Canon a setting of 0 is neutral contrast. I could be wrong, though.

  7. #7
    Senior Member
    Join Date
    Dec 2008
    Posts
    505

    Re: ETTR + uniWB + DPP = ??? (Newbie guide?)



    Quote Originally Posted by Daniel Browning
    If you reduce contrast (inverse "S" tone curve), you make it seem like values are lower than they actually are

    That is what you "are" doing. The question is how accurate is it. As I stated in another post, it works good in some situations and not so good in others. The "good" was indoors underlimited dynamic range lighting. The "not so good" was in outdoor under wider dynamic range lighting. I imagine that somewhere there's a middle of the road depending on the situation and that would take some experimented under various conditions to come up with. If I have any exrta time I'll do some experimenting. I am however already quite crippled with a under performing archaic LCD. [] whaaa!

  8. #8
    Junior Member
    Join Date
    Dec 2008
    Posts
    14

    Re: ETTR + uniWB + DPP = ??? (Newbie guide?)



    Thanks for all the responses... I'll keep trying...

Posting Permissions

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