New to digital negatives

Sandy,

The ternary approach does use some black ink, since the the three corners of the triangle are in RGB space, not CMY space. And yes, it can block signficantly more UV light than is possible with the pure binary color-only approach of PDN. I have a UV densitometer that was very useful when I was coming up with this method for the 1280 and the 2200.

I never quite understood why the PDN system goes to such lengths to avoid using black ink entirely. I get very nice smooth and non-grainy results with the 2200 and a green color blend that does have some black ink in the mix. And the adjustment curve is no more 'extreme' than the ones I developed in PDN. Frankly, if you are working with 16 bit files, a curve has to be pretty extreme before there is any noticeable image degradation.

A lot of these systems can all be made to work quite well. The PDN approach is well thought-out and is truly a system. My ternary system is considerably rougher to use at the front end, but is quite useful, especially if you cadge a copy of Chartthrob to analyze your test prints. I developed the ternary system mainly as a freeware approach that I could use in workshop situations where people were understandably reluctant to pay for the PDN system when there could be a distinct possibility that they might never even print palladium again.

sanking said:

Has it really been shown that Clay's ternary-ratio method, and/or your red-green array are capable of producing more UV (or Blue) blocking density than the color combinations of PDN? I don't understand how this is possible, assuming you don't use the blacks, but if it is perhaps you can suggest a specific combination for the Epson 2200 that will produce more UV (or Blue) density than any of the PDN combinations.

Sandy King

Click to expand...

clay said:

Sandy,

The ternary approach does use some black ink, since the the three corners of the triangle are in RGB space, not CMY space. And yes, it can block signficantly more UV light than is possible with the pure binary color-only approach of PDN. I have a UV densitometer that was very useful when I was coming up with this method for the 1280 and the 2200.

I never quite understood why the PDN system goes to such lengths to avoid using black ink entirely. I get very nice smooth and non-grainy results with the 2200 and a green color blend that does have some black ink in the mix. And the adjustment curve is no more 'extreme' than the ones I developed in PDN. Frankly, if you are working with 16 bit files, a curve has to be pretty extreme before there is any noticeable image degradation.

Click to expand...

Clay,

I can see that if you throw black into the mix a lot more transmission density is possible from the 2200 since the black ink by itself is capable of a much higher density than any of the color mixes.

Not sure why black is avoided in PDN. Several years ago I promoted a blend step tablet that had two layers, one consisting of the color that gave the highest denstiy, and another of black. It was possible to adjust density by blending in varying degrees of black. I thought this was a very good solution for the Epson 2200 and discussed it with Mark but I think his preference for getting more density from the 2200 is to increase the amount of ink that the printer lays down.

In my own work with the blended step tablets results appeared to be as smooth and as fine grain as with the green color negatives, so I personally find no reason for not using black inks if they serve to adjust the density to needed levels. Persons working with PDN might find it useful to use a black blend layer over the top of the tonal palette with the new Epson 3800 because from what I have seen the green branch gives a fairly grainy look in pt./pd and the only other possibility in the red branch is much lower maximum density.

Sandy

sanking said:

Has it really been shown that Clay's ternary-ratio method, and/or your red-green array are capable of producing more UV (or Blue) blocking density than the color combinations of PDN? I don't understand how this is possible, assuming you don't use the blacks, but if it is perhaps you can suggest a specific combination for the Epson 2200 that will produce more UV (or Blue) density than any of the PDN combinations.

Sandy King

Click to expand...

As Clay has addressed the issue of black ink I won't go there. But yes, I was also able to produce density greater than PDN's palette and greater than that of a combined CcMmYK Black by using an orange colour (R176, G176, B0) or thereabouts. This is with Epson dye inks however, not dye-pigmented. As you know, this colour is not part the PDN system. My method also excludes blue, the intent was to create as much density as possible that's why it's specifically using just red and green in a matrix. I've made Red-Blue and Green-Blue arrays also so I can run the whole gamut. I took the approach of slicing up the RGB colour cube model on the X and Y axis, basically running all colour possibilities. The interesting thing about [176,176,0] being denser that the CcMmYK black is that I theorise it actually uses less black ink, so it must be getting the density from its spectral properties rather than just opacity. I think I posted the red-green array somewhere a while back, if you want to try it on the 2200 e-mail me off-line and I'll send you a copy.

mkochsch said:

As Clay has addressed the issue of black ink I won't go there. But yes, I was also able to produce density greater than PDN's palette and greater than that of a combined CcMmYK Black by using an orange colour (R176, G176, B0) or thereabouts. This is with Epson dye inks however, not dye-pigmented. As you know, this colour is not part the PDN system. My method also excludes blue, the intent was to create as much density as possible that's why it's specifically using just red and green in a matrix. I've made Red-Blue and Green-Blue arrays also so I can run the whole gamut. I took the approach of slicing up the RGB colour cube model on the X and Y axis, basically running all colour possibilities. The interesting thing about [176,176,0] being denser that the CcMmYK black is that I theorise it actually uses less black ink, so it must be getting the density from its spectral properties rather than just opacity. I think I posted the red-green array somewhere a while back, if you want to try it on the 2200 e-mail me off-line and I'll send you a copy.

Click to expand...

Is it possible that the printer may have been using some black ink at the array you selected? Since we are actually using the subrative colors of cyan, magenta and yellow to produce color density I find it very difficult to understand why one of the primary colors produced by the PDN system, either Green, Red, or Blue would not have produced maximum blocking density. I would like to experiment with your red-green array if you will send me the information. My direct email is sanking@clemson.edu.

Sandy

sanking said:

I would like to experiment with your red-green array if you will send me the information. My direct email is sanking@clemson.edu.

Sandy

Click to expand...

Hey Michael,

Could you send me a copy also? I'm curious to see how your system works.

Thanks,

Don Bryant

mkochsch said:

I think I posted the red-green array somewhere a while back, if you want to try it on the 2200 e-mail me off-line and I'll send you a copy.

Click to expand...

Hi,


Thanks very much for the file you sent via private email. I am looking forwarding to exploring the potential of your method.

Best

Sandy

mkochsch said:

I think I posted the red-green array somewhere a while back, if you want to try it on the 2200 e-mail me off-line and I'll send you a copy.

Click to expand...

I tried the red-green array and it works very well with the Epson 2200. A person with a good knowledge of the exposure scale of their process could probaby quickly figure out the right color from your method by first finding the exposure to print maximum density with the substrate, and then printing the array on our OHP to find the color combination that gives paper white, or just next to it. Or, if you have a transmission densitometer that reads UV you just find the square on the array that produces a DR that matches the ES of your process. The concept is not that different from PDN, but the additional color possibilities do allow a higher DR (at least with the Epson 2200, and the Red-Green array) than one can get with the colors of PDN.

Thanks again for sending me the files.

Sandy King