[argyllcms] Re: A better way to measure paper white?

On 2009 Oct 31, at 12:35 PM, Gerhard Fuernkranz wrote:

Well, in order to achieve a good illuminance uniformity, I would also
recommend a symmetric setup with (at least) two equal light sources.

Thanks for the charts, and for the confirmation that I'm not crazy.

My setup might be overkill, but it's not much more hassle to set up and tear down than a one- or two-light setup, and it's mentally easier. I visualize (and measure) a 7' cube. Strobes with reflectors only (no umbrellas, etc.) go at the corners of the cube, each pointing across at the opposite corner and however far down the modeling light shows the most even illumination of the art (placed on the ground in the center) if that were the only light. The center pole of the tripod gets mounted crosswise and fully extended; the tripod gets weighted down to prevent tipping, and the camera goes on the end facing down. Then there's a bunch of fiddling to adjust height, get everything parallel (a spirit bubble level is invaluable), and somewhat centered. Shutter to 1/100 or thereabouts, aperture to whatever that lens's optimum is (usually f/8), and adjust the lights for proper exposure (with a gray card, checking that ProPhoto R=G=B=100, preferably across the whole field of view). Shoot a color target, shoot the art, swap out the art, repeat until done, do a sanity check in Photoshop, tear down the set. Process the RAW files, including the target, use the target to create a profile, apply said profile to the pictures and convert to whatever color space best matches the art (usually either ProPhoto or sRGB, sometimes AdobeRGB), and finally crop / sharpen / straighten / etc. in Photoshop. (And then, often, profile a piece of the same paper to print a reproduction on.)

Here's an example from my latest session, a semi-abstract watercolor my Mom did of some wheat. You can tell from the shadows cast by the rulers (used to hold down the paper which really, really wanted to curl) that the light is nearly uniform, but the specular reflections on the metal also indicate I could have done better. But for the intended purpose, this was plenty ``good enough.''

Cheers,

b&

image/jpg

Other related posts: