Most Accurate White Balance Technique Using LR3

Thanks everyone for tipping in and helping! I will give this a good read tonight.

The silly thing about the LR eyedropper (IMO) is, that you cannot specify its sample size.

http://forums.adobe.co​m/thread/854734?tstart​=0
http://feedback.photos​hop.com …ce_eyedropper_t​ool_option

I carry a WhiBal gray card. When in doubt, shoot an image in the same lighting with that. Use the eyedropper, and adjust as needed.

I have often asked someone to stand, holding the whibal in front of them. No one ever balked. They get a kick out of it, in fact.

The problem is greater if you are doing an event with highly varied lighting. then the best is to get target shots with a gray card for all of them. you can set the white balance for each in LR and copy that setting to whichever images are appropriate.

René Damkot wrote in post #13075158
The silly thing about the LR eyedropper (IMO) is, that you cannot specify its sample size.

http://forums.adobe.co​m/thread/854734?tstart​=0
http://feedback.photos​hop.com …ce_eyedropper_t​ool_option

René,
Pick the WB eyedropper tool, place the tool in a section of the photo to be balanced, roll the scroll wheel on the mouse...gives you a range of 5x5 pixels to 16x16 pixels.

That's discussed in this thread: http://forums.adobe.co​m/message/3701576

I've never seen a difference in WB when clicking, using the scroll wheel and clicking again, so I'm pretty certain it doesn't work that way.

Otherwise, these two "clicks" should be drastically different because of the orange being sampled in the first, yet not in the second. They are identical (if I don't move the mouse, which I obviously did in making the screenshots)

René Damkot wrote in post #13076103
That's discussed in this thread: http://forums.adobe.co​m/message/3701576

I've never seen a difference in WB when clicking, using the scroll wheel and clicking again, so I'm pretty certain it doesn't work that way.

Otherwise, these two "clicks" should be drastically different because of the orange being sampled in the first, yet not in the second. They are identical (if I don't move the mouse, which I obviously did in making the screenshots)

Some years ago, TMR Design and myself engaged in studio flash and speedlight output measurements, including the desire to measure White Balance and Tint consistency at various power levels. We gave up trying to characterize WB+Tint back then, due to the issue of pixel sample size and the lack of averaging a zone, in most/all available software which we owned.
FF a few years, and I discovered the scroll recently, and WRONGLY thought this was the averaging feature. Sadly, you pointed out the error in that ASSumpition!

It would be very nice if it did work that way

From what I see scrolling only alters the zoom of the view, so presumably you can be more precise in selecting that single pixel...?

zerovision wrote in post #13066116
xrite colorchecker passport. Here are some reviews and info. If you take the time to learn what it can do, you can use this for very accurate color and warming skin tone as well. Pocket size and easy to carry around with you.

http://www.luminous-landscape.com …lorchecker-psssport.shtml

http://xritephoto.com …uct_overview.as​px?ID=1257

This is what I use when I need accurate white balance/color. From that you can adjust to taste. But shooting artwork doesn't allow for your 'taste'.

The neat thing about the xrite passport is the ability to profile your camera using two light sources and save that as a generic default for processing raw images.

Have another read of this thread again today and just want to say thanks to all the generous POTNers who constantly giving new free lessons.

René Damkot wrote in post #13076103
That's discussed in this thread: http://forums.adobe.co​m/message/3701576

I've never seen a difference in WB when clicking, using the scroll wheel and clicking again, so I'm pretty certain it doesn't work that way.

Otherwise, these two "clicks" should be drastically different because of the orange being sampled in the first, yet not in the second. They are identical (if I don't move the mouse, which I obviously did in making the screenshots)

This was making me nuts last night, until I came to the same conclusion as you - it's doing a balance based on that single pixel, not an average of the "selection" (if you can call it that, since it's not really a selection). It's almost like that WB picker interface is not working as intended - it SHOULD be taking an average, but it's not. Or maybe it's intentionally designed to trick the user, just a big prank played on us by Adobe.

Maybe I'll try the aforementioned advice to crank the color NR way up, and try again.

I normally carry around my PhotoVision calibration target, but my last two shoots were stage performances where it's obviously not possible to use a gray card. I don't want to correct everything - I need to keep SOME stage lighting if only to preserve the ambiance - but at the same time, I'd like skin tones to look pleasing.

I can manage doing it by hand, but it's slow going.