Banding in radials

I still can’t seem to produce a linear quad without banding.

First, I printed the supplied T2 Cone Curve and linearized it. This caused banding even with 100%/100% smoothing and 100% curve smoothing. I was seeing graininess and tried limiting the ink. This helped resolve the graininess but I was still seeing banding. My theory is that the linearization process was lightening/stretching the inks too much. What I really needed was more/less ink. I decided a better approach would be to create a more accurate starting curve by using ink limiting and then do the linearization process. I did several 21-patch tests with varying ink limits until I had a quad that was closer to linear. I then did a 129-patch linearization. This has produced better results than starting with the Cone Curve but the linearization is still lightening/stretching the inks toward the highlights and causing banding.
In addition, I’m not getting a dMax other users are reporting. I printed the calibration and the darkest I read on T2 is 16.3L. I have been able to get my limited quad close to this.

Furthermore, you may recall, I replaced this printhead and I’ve triple-checked the voltage.


T2 Cone Quad, no linearization


T2 Cone Quad, 129-step linearization


T2 Cone Quad, 129-step linearization, print


T2, all inks limited 80%, 129-step linearization 70% / 20% smoothing


T2, all inks limited 80%, 129-step linearization 70% / 20% smoothing, print


T2, special ink limiting, 21-step linearization, no smoothing


T2, special ink limiting, 21-step linearization, no smoothing, print


T2, special ink limiting, 129-step linearization, 70% / 20% smoothing


T2, special ink limiting, 129-step linearization, 70% / 20% smoothing, print

you are seeing banding for a few reasons.

  1. Your printer has some ink slosh on the left and right of the printed target.
  2. This is causing a step ladder effect with the measurements (you can probably see this if you turn smoothing off entirely).
  3. Your measure smooth settings are not set properly. Set them to 80 and 80 for this particular printer.
  4. After linearization, re-load the .quad into the starting curve, delete measurements entirely, and set ink channel smoothing to 100%.
  5. Use this final curve to print with.

I suggest the 256 patch target. Or the 21x16.

best,
Walker

For anyone who has a similar problem, Walkers’ suggestion worked perfectly! Not sure how much dialing in the ink limiting helped but I’m getting a perfect linearization without banding or graininess. I am still not getting the dMax others claim to be achieving but it’s good enough; I’ll have to investigate this another time.

Thank you Walker, my hero!

not really sure about dMax. seems to be particular to this printer (maybe the electronics in the board).

most people are getting between L* 12.5 and L* 13.89 on basically all different printers but T2 paper is always a little less dMax (similar to Moab Entrada). I get around 14.4-ish on the P800 with this paper/ink combo.

best,
Walker