Epson P900 - Black Enhance Overcoat

[thebestcpu]

If you don’t have a P700/P900 printer, you might want to skip this post.​

I've been testing the print output from the P900 to better understand some of its features beyond the marketing descriptions. This post is the first of several planned posts with this one focusing on the Black Enhance Overcoat (BEO) feature. I welcome any questions or feedback on my approach.

Testing Setup​

To keep things simple, I am conducting grayscale-only testing on Epson Ultra Premium Luster paper.

Here’s a screenshot of the options available in the printer driver:

(Insert image)

The driver provides five preset Print Quality settings, and users can further customize them.

I tested these five preset modes, along with two additional variations, as shown in the following matrix.

Note: The quality mode labels (e.g., “Standard”) do not correspond to fixed output DPI settings. Instead, the DPI values vary depending on the selected paper type. In other words, “Standard” is a relative quality setting, not an absolute resolution setting. (This seems like a marketing decision, but I find it less than ideal as an engineer.)

For this post, I will focus on the results of the options outlined in red within the matrix.

---

Test Details​

I evaluated print output at 1440x1440 DPI and 5760x1440 DPI, both with and without BEO enabled.

Here is the test image I used:

• A 16-bit TIFF containing 100 squares with Lab values ranging from L=0,0,0 to L=99,0,0.
• Paper white was measured for the last point for the L=100 position.

There were four target sheets, one for each of the four modes evaluated in this post.
Using Spotread (from the ArgyllCMS package), measurements were taken with an i1Pro spectrophotometer (default settings).
Printing was done from my MacBook Pro using ColorSync Utility, which allows color-target printing (i.e., no color profile applied).

---

Results​

1440x1440 DPI Output (With & Without BEO)​

The first plot compares print output at 1440x1440 DPI with and without BEO. As expected, prints with BEO enabled appear slightly darker.

To quantify the effect, I plotted the difference between the two curves. The average ΔL (change in luminance) is approximately 0.5 across most of the test squares (up to L = 80), tapering off to zero in the highlights.

This ΔL range aligns with expectations since BEO adds a layer of light gray ink. Interestingly, the effect extends across most of the tonal range, not just the darkest shades. This suggests that BEO slightly reduces black dullness, which aligns with marketing claims that it "lightens the look" by improving sheen and reflectivity.

5760x1440 DPI Output (With & Without BEO)​

The second graph compares print output at 5760x1440 DPI under the same conditions. Here, the ΔL difference is closer to 0.7, following a similar pattern—affecting tones up to L = 80 and tapering off.

Learnings and Possible Further Investigation
- The amount of BEO differs slightly between the two Quality options examined in this post.
- The BEO covers quite a bit of the L range (up to L=80) with the same amount of delta L to that point.
- I did not pursue evaluating sheen differences, which may be complex with the Luster texture.
- I captured Lab's "a" and "b" components, which could be similarly examined/plotted.
- The variations in the delta L are relatively large compared to those seen from multiple readings with Spotread for each target. Multiple measurements per spot could be taken to minimize measurement variability (although relatively small). Other sources of variability could be limits from the printer driver in bit depth, variations in "a" and "b" component vs the calculation from the full spectrum measurements to D50 Lab, or some other systematic difference in printing and the test target. I did try focused external lighting on the i1 Pro and amazingly I did not see variations in the measurement due to that.

You can always give observations/feedback/suggestions.

---

Coming Up Next...​

In my next post, I’ll discuss the enigmatic Carbon Black setting—which, as it turns out, isn’t quite what it sounds like.

John Wheeler

 

[ruuduitdegraaf]

Thanks and interesting. Looking forward for your next episode.
I haven’t experienced that ticking BEO to on makes a considerable difference. I think I will continue to forget the relevance since also your experiment shows the delta is very small..