X-Rite I1 Publisher/I1 Pro OOBE, part 6

I got an e-mail from X-Rite support this afternoon asking for more information. I sent it off.

Meanwhile, encouraged by the fact that I had finally made a profile, albeit a terrible one, I started over. I printed a new 400-sample test target, and was able to measure it with only one bad row. After the measurements Publisher generated a profile with an entirely reasonable looking gamut. I took the next step and asked Publisher to print out the 400 patches that it thought would best enable it to optimize the profile. The test page came out in portrait orientation, and I couldn’t figure out any way to switch it back to landscape. I ignored that detail, printed it, measured it without much drama, and let it generate an optimized profile.

I installed the profile, fired up Photoshop, loaded one of Charlie Cramer’s test images, and printed it. The result was very good, surprisingly good when you consider that the profile was constructed from only 800 color patches (counting both the original profile generation and the optimization step).

A lesson learned: if Publisher doesn’t supply audible prompts during the measurement phase of the profile generating process, stop and figure out how to make it give you those prompts. There is a greater delay between pressing the measurement button on the spectrophotometer and the time that you can actually start to move it than I would’ve thought. That was probably the source of most of my difficulties at the beginning. The weird thing is that I have no idea what I did to get the audible prompts. I hope they’re there for good.

Another lesson:  When you’re done measuring a page, whether I1 Publisher thinks it’s a good page or not, look carefully at the color that Publisher thinks each square should be and compare it to the color that you measured. If they’re wildly off, redo that row. If you see a bad row, you can go back and re-measure it by holding down the right mouse button and dragging the red frame that indicates the row you’re measuring to the row that you want to re-measure.

I may be a convert.