Recently, a color process control manager at a large print production facility wanted to know if there is a more comprehensive chart available for daily digital color evaluations than an 12647-7 proofing wedge. He pointed out the IT8.7-4 has too many patches, and the P2P51 has too many gray finder patches. Reiterating a thought we’ve all had many times, he asked: “Am I overthinking the value of additional patches?”
There is a tradeoff between patch count and how effective a chart is at gathering QC information. There is also something to be said for both extremes; too many patches and too few patches. Too many patches on a noisy (grainy, low screen ruling, etc.) printing device can cause unwanted noise in the measurement data (like using a 1 pixel eyedropper setting in photoshop to determine the dot percentage in a noisy image). Too few patches and you are not sampling enough colors to accurately model how the device is printing.
I just dissected the TC3.5 patch set and found it to be lacking in the 3 color grays. There are not many patches and none are G7 compliant gray patches. In my opinion, this eliminates the TC3.5 for any G7 evaluation. In fact, most of the currently available charts are not very good in the gray areas, especially if you are trying to evaluate G7 compliance. Idealliance built the TC1617 to address this lack of G7 gray patches in the IT8.7-4, but even this chart has too many patches for day-to-day evaluations.
The 3-row 2013 12647-7 chart (the replacement for the 2009 2-row chart) was built as a very good compromise between patch count and patch value. It has a decent number of patches to effectively evaluate print consistency, which includes G7 compliant gray patches, the typical array of CMYKRGB tone ramps, pastel patches, saturated patches, and a good assortment of dirty patches. These dirty patches were purposely built with CMY values and then with 100% GCR values excluding the 3rd color and replacing it with K. This was done because many separations, especially those done with ink reduction products, are made with GCR these days. It’s hard to beat what’s in that 3-row, 84-patch control strip.
While considering charts and patch values, it’s almost more important to note the metrics and tolerances we place on these patches for conformance to specifications. If you look at the metrics we currently use for pass/fail, they are very CMYK printing press centric. Commercial print, specifically offset printing, has been the forefront of most industry standard and best practice development. Therefore much of the data gathering and evaluation is based on printing devices where C, M, Y, and K ink thicknesses are controllable by the operator. This means most metrics are tied to effective control of those ink thicknesses, which is largely irrelevant to the digital world.
We should be asking: “What are we passing and failing?”
For the G7 Colorspace metrics (currently the most stringent) we are evaluating:
- Substrate – Paper color is good to evaluate
- Solid CMYK – Very useful to press operators, but not much of a typical image or job is just solid C, M, Y, or K. This makes these patches poor for evaluating digital print consistency, especially visual consistency.
- Solid RGB overprints – In my opinion, this is more important than Solid CMYK, as overprinted colors are what we see when we look at printed material. Still, these are only the solids, no tints.
- CMY gray balance and tone – This is very important in controlling and evaluating print consistency, although it’s more important in print processes that lay down individual CMYK inks like offset.
- All the other patches (pastels, saturated, dirty colors, skintones, CMYKRGB tints) are all lumped into a single metric called ‘All’ and then given a whopping average ∆E of 1.5 or 2.0 and a worst patch ∆E of 5.0 (95th percentile). That’s huge! A virtual barn door to let almost anything outside of grays and CMYKRGB solids pass.
These are not very visually oriented metrics and tolerances. So the big question to ask is what are you evaluating with your chart, or more importantly, what metrics and tolerances are you using to evaluate your chart? For G7 you could just use a P2P and eliminate the gray finder patches (columns 6-12), because the metrics are really only focused on CMYKRGB solids and the gray patches.
Bottom line, if we are looking for print consistency, we need to look at establishing new metrics that truly help us determine how visually consistent a print is. After a great deal of research, I believe this should be based on a cumulative relative frequency model (CRF) that evaluates all colors in a chart. In a CRF model, each and every one of the patches is relevant to visual consistency and is being counted within the evaluation. I have found the 3-row control strip does an excellent job of evaluating visual print consistency when using CRF. I’ve also performed the experiment in live production many times and have continued to get feedback from users who say using CRF and the 3-row control strip is the best method they’ve found to evaluate visual consistency.
If you would like to see the true power of CRF and real world metrics, try SpotOn! Verify. The trial is free, and our team will help you get started.