Pictorico media setting on canon printers

What is the best media setting to use on canon printers when printing on Pictorico Pro Ultra Premium TPS100? More specifically, I'm printing on a pro-100 and I'm trying to get the best uv blocking it can produce. Can the media setting impact the ink density and therefore the uv blocking effectiveness?

The basic answer is YES, ink density impacts the UV blocking capability of a digital negative. Don't know if it works this way with Canon printers, but with Epson when selecting media type, the further down the list you go the more ink is laid down. You should dive into your driver settings...I'm sure there must be some controls for density. You might want to poke around this site at bit: http://www.easydigitalnegatives.com/. Lots of valuable info there.

Makes sense. I'm actually using EDN and I've been through most of the info there.

The list of media types I see in the driver are these:


I haven't been able to find any info on the ink density laid down for each. I'm guessing glossy uses the least ink, followed by semi-gloss, luster, matte, and fine art paper. Pictorico claims tps100 having a glossy surface, but with 25% more ink absorption than previous versions, so I might try matte, hoping it won't create a mess.

What I do with HP printers is print step tablets (I mostly use the EDN 256 file) on all the media settings, and look at the results. Needless to say, I do this on small sheets. You don't even need to use film; I find cheap photo paper is good enough.

It's usually quite obvious from this which settings are putting down the most ink. This usually narrows the choice down to a few cases that are the most promising, and I run tests on those using film to find the best setting overall.

In Epson drivers you can set a short delay after each pass to allow a brief drying time. Maybe, the Canon will allow same?

Alan9940 said:

Maybe, the Canon will allow same?

Click to expand...

One reason why Epson printers have been very popular with the alt. process crowd is because of the numerous settings in the driver that allow any user to influence printing behavior in a very detailed manner. No other (consumer/prosumer) printer level manufacturer offers this kind of control in their drivers to the best of my knowledge. So what works with Epson, often really only works with Epson and that's it.

revdoc said:

What I do with HP printers is print step tablets (I mostly use the EDN 256 file) on all the media settings

Click to expand...

I think that's a good workaround to figure out how to manipulate ink density on a Canon printer. Personally I would actually use transparency film; probably something cheap like screen printing film, and then evaluate contact prints from a long-range and easy process. I would probably start with something like Van Dyke Brown, because those are quick and easy to make and process and have a reasonably long tonal scale.

zuluz said:

Pictorico claims tps100 having a glossy surface, but with 25% more ink absorption than previous versions

Click to expand...

The limits of ink density that the Pictorico film will take is most likely way beyond what you can dial in on the Canon printers. With Epson printer's you'd have to lay down at least a factor two more ink than on a normal paper. That's well outside the normal range of ink density variation in a printer driver, as far as I know.

Not that there's often a good reason to seek out the limits of Pictorico. Usually you get way sufficient density long before that point is reached!

Good to know that the limits of pictorico is beyond my reach. I will experiment as @revdoc suggested. Thanks!

What kind of alt. printing processes do you envision using your negatives for? I've only found it necessary to go to extremes with processes like salted paper and carbon transfer, the latter under certain process conditions including very low pigment loads and thick tissues. Even with salted paper there's a good chance that one of the default paper settings on your Canon will produce sufficient UV blocking for a decent tonal range. Remember you can always slightly underexpose the print to retain clean whites; you'll sacrifice the blacks a little, but e.g. gold toning can make up for that in part, and it's often not all that problematic if dmax in a print is a little lower than the maximum attainable. The human eye adjusts to it pretty well and 'experiences' it as black nonetheless.

I'm recalibrating everything for the 'simple cyanotype' chemistry (wider tone and lower contrast compared to classic) using the EDN system. My main issue is that I have to reduce exposure by 20% to get proper whites (12m instead of 15m), so I'm probably losing all that extra tonal range that I could have gotten with simple cyanotype.

But beyond all that, I'm just curious to know how to get the max uv blocking capability out of pro-100 when using tps100.

zuluz said:

Makes sense. I'm actually using EDN and I've been through most of the info there.

The list of media types I see in the driver are these:

View attachment 341626
I haven't been able to find any info on the ink density laid down for each. I'm guessing glossy uses the least ink, followed by semi-gloss, luster, matte, and fine art paper. Pictorico claims tps100 having a glossy surface, but with 25% more ink absorption than previous versions, so I might try matte, hoping it won't create a mess.

Click to expand...

Not the way I understand. The amount of ink laid is more a function of how much the paper can take it without pooling - for that the more expensive papers will likely have amount of ink dialled up to maximize the Dmax so they can charge more money for that. In Epsons, they are called Ultra Premium papers. Don't know what the equivalent for that in Canons is called. Glossy does not necessarily mean least amount of ink compared to semi-gloss etc. Also, if you specify a matte paper or fine art paper, it will print with the MK ink - which has different characteristic than the PK. MK in general has a higher UV opacity so if you print on a Pictorico with any of the matte paper selections, the resulting negative will have higher density. However, the MK ink also does not adhere/adsorb well to the coating on the transparency so there is a greater tendency of it pooling and not drying quickly so they can be still wet when they come out of the printer, with a greater propensity for the dreaded "pizza wheels." Also, in general, the MK ink tends to look much more grainier. All in all not something I wanted to mess with. Again my experience is on an Epson. Canon behavior may be different. So testing on your own is required. You don't even have to do whole step wedges as suggested - just print a bunch of (0.0.0) squares individually, side-by-side with different paper selections on the same transparency and print them all together with the process of your choice and see which one comes out the cleanest.

:Niranjan.

Great info @nmp, thanks. So, all things considered, is there a generally preferred media type that all you Epsonites use to get the best uv performance on tps100?

zuluz said:

Great info @nmp, thanks. So, all things considered, is there a generally preferred media type that all you Epsonites use to get the best uv performance on tps100?

Click to expand...

You will be hard pressed to get a consensus agreement to this question. Everyone probably has their own quirky way of doing things. I use Epson Ultra Premium Clossy settings. I think though, if you are going to use a colorized negative and if the color is significantly away from 100% black ink, it probably does not matter whether you used the the paper selection that gives the highest amount of black ink dumped at (0,0,0) step. Your best blocking color could be (R1, G1, B1) for paper selction 1 and you will get another (R2, G2, B2) for paper selction 2, but they both should give approximately the same UV opacity. So if you are using the color matrix from Peter Mhrar, just do the test with best glossy paper selected with the corresonding icc profile and find the color with the most blocking density and then use the same conditions for making your real negatives.

:Niranjan.

zuluz said:

is there a generally preferred media type that all you Epsonites use to get the best uv performance on tps100?

Click to expand...

I think most Epsonites rely on custom paper profiles where they manually manipulate ink densities, or even QuadTone RIP, which allows the creation of a custom ink curve for each individual ink channel. There simply are not substitutes for this for Canon printers to the best of my knowledge.

The problem with the predefined paper profiles, as you've found out, is that they're a black box and you don't know exactly what you're getting.

I think the test strip approach suggested by @revdoc is really the way to go!

I finally got a chance to print the various media types to see which blocks UV the most on a Canon Pro-100 using Pictorico TPS-100:



Looks like "Matte Photo Paper" is the one to use. The other ones are more or less the same and not as good. I'm also using H=330 which seems to be the optimal color per EDN.

It's a little surprising to me that even though matte blocks UV the most when K=100, it blocks less than others when K<75.

The other notable thing is that the jump from K=100 to K=95 is very abrupt which means the compensation curve needs to stretch that 5% quite a lot.

zuluz said:

It's a little surprising to me that even though matte blocks UV the most when K=100, it blocks less than others when K<75.

The other notable thing is that the jump from K=100 to K=95 is very abrupt which means the compensation curve needs to stretch that 5% quite a lot.

Click to expand...

Yeah, what you're seeing here is that blocking power is not necessarily the only criterion to take into account. I also notice the rapid drop-off when using the Matte setting; note how it has one less step to its tonal range on the right. This means it is just a little more contrasty than the other channels, and indeed, it seems it's also very non-linear.

Keep in mind that curves for optical density (regular prints) are generally quite different from negative curves to begin with. What may be going on here is that pure black is given an additional boost (strongly upswept curve at one end) to make solid blacks as convincing as possible even on matte papers with their lower reflective density. At the same time, the rest of the curve may be shaped quite differently. Since these curves are embedded into the printing system itself, you never really get to see them. Using QTR with an Epson printer does give some glimpses of what really happens. It's interesting for sure.

One thing that concerns me a little is that for the printing process you've chosen here (New Cyanotype or Simple Cyanotype?) you don't have any ink that fully blocks. Hence, you're stuck with either some kind of double-patterning where you overlay two negatives on top of each other (but this will give problems with sharpness), or with a low dmax in your prints. On the other hand, you may get different results with a different paper. Maybe you can find a paper that happens to give a nice tonal scale with your particular combination of digital negatives, exposure etc.

What kind of printing processes do you intend to use these negatives with? For e.g. classic cyanotype and Van Dyke Brown, it looks to me you should be just fine. It'll be the long-scale processes like New Cyanotype and salted paper where you appear to be running into limitations.

zuluz said:

I'm also using H=330 which seems to be the optimal color per EDN.

Click to expand...

What do mean by this: H = 330, what about S & B. Did you actually print a color matrix?

:Niranjan.

nmp said:

What do mean by this: H = 330, what about S & B. Did you actually print a color matrix?

:Niranjan.

Click to expand...

I used EDN (Easy Digital Negative) ColorBlocker to identify the optimal color and used the generated LUT file.

OK. It's a little confusing to me. Ideally you would want to do the color blocker after you have done the paper settings test. Because the best color blocking color will vary somewhat with different paper settings. If you decide to use matte paper settings, and if you are not satisfied with the K=100% opacity, then use the color blocker test to get the optimum color with greater UV opacity using the same settings.

Also, the color at K=100 in all cases above seems to be RGB of 0,0,0. May be I am wrong, it's hard to tell and correct me if so. But if you are using an optimum color, then the K=100% should be that color.

:Niranjan.

koraks said:

What may be going on here is that pure black is given an additional boost (strongly upswept curve at one end) to make solid blacks as convincing as possible even on matte papers with their lower reflective density. At the same time, the rest of the curve may be shaped quite differently.

Click to expand...

Yes, that seems to be what's going on.

koraks said:

One thing that concerns me a little is that for the printing process you've chosen here (New Cyanotype or Simple Cyanotype?) you don't have any ink that fully blocks. Hence, you're stuck with either some kind of double-patterning where you overlay two negatives on top of each other (but this will give problems with sharpness), or with a low dmax in your prints. On the other hand, you may get different results with a different paper. Maybe you can find a paper that happens to give a nice tonal scale with your particular combination of digital negatives, exposure etc.

Click to expand...

You're right, I'm pretty much resigned to a lower dmax with this printer (about 15-20% below dmax by my estimates). Right now I'm just trying to get a half-decent linearization. I have no idea if there are other ink sets that work with PRO-100 and offer better UV blocking.

koraks said:

What kind of printing processes do you intend to use these negatives with? For e.g. classic cyanotype and Van Dyke Brown, it looks to me you should be just fine. It'll be the long-scale processes like New Cyanotype and salted paper where you appear to be running into limitations.

Click to expand...

I'm using the Simple Cyanotype on COT320.

nmp said:

OK. It's a little confusing to me. Ideally you would want to do the color blocker after you have done the paper settings test. Because the best color blocking color will vary somewhat with different paper settings.

Click to expand...

You're probably right and I'm definitely lazy! I may go back and redo the color blocker test.

nmp said:

Also, the color at K=100 in all cases above seems to be RGB of 0,0,0. May be I am wrong, it's hard to tell and correct me if so. But if you are using an optimum color, then the K=100% should be that color.

Click to expand...

This is where I am confused. I'm letting the generated LUT file do the coloring, and at K=100 it is definitely RGB of 0,0,0 (not some color). That and K=0 seem to be the exceptions, while all other Ks are a shade of the color (H=330).

Here is the LUT file generated by EDN:

#Created by: Easy Digital Negatives
TITLE "Easy Digital Negatives"
LUT_1D_SIZE 3
DOMAIN_MIN 0.0 0.0 0.0
DOMAIN_MAX 1.0 1.0 1.0
#LUT data points
0.000000 0.000000 0.000000
0.800000 0.000000 0.400000
1.000000 1.000000 1.000000
#END data

zuluz said:

You're probably right and I'm definitely lazy! I may go back and redo the color blocker test.


This is where I am confused. I'm letting the generated LUT file do the coloring, and at K=100 it is definitely RGB of 0,0,0 (not some color). That and K=0 seem to be the exceptions, while all other Ks are a shade of the color (H=330).

Click to expand...

Every new version of EDN gets a little more complicated, I am not that familiar with the latest one so I can't say what you did there to get a black block at k=100%. Simple still works. Using the paper setting you think you are going to use (either PK or MK) print the color matrix on the transparency and print the cyanotype. Then let EDN give you the optimum color (you can also eye-ball it) - say R,G,B = A,B,C. Then print your inverted step wedge with the colorized layer on top - it is a fill layer with fill color of ABC. Change the blending mode to Screen. You will see that the darkest step will now be not 0,0,0 but A,B,C. Lightest step should still be 255,255,255.

:Niranjan.

zuluz said:

You're probably right and I'm definitely lazy! I may go back and redo the color blocker test.


This is where I am confused. I'm letting the generated LUT file do the coloring, and at K=100 it is definitely RGB of 0,0,0 (not some color). That and K=0 seem to be the exceptions, while all other Ks are a shade of the color (H=330).

Here is the LUT file generated by EDN:

#Created by: Easy Digital Negatives
TITLE "Easy Digital Negatives"
LUT_1D_SIZE 3
DOMAIN_MIN 0.0 0.0 0.0
DOMAIN_MAX 1.0 1.0 1.0
#LUT data points
0.000000 0.000000 0.000000
0.800000 0.000000 0.400000
1.000000 1.000000 1.000000
#END data

Click to expand...

I didn't see this earlier, but I tried out the above LUT file and sure enough it gives 0,0,0 for the 100%K block. Either there is a problem with the program (I am not able to generate a lut file myself for some reason using a sample scan - may be related to some settings in my browser) or the best blocking color is indeed black. Can you share the scan of the HSB print that you used to calculate the LUT. We can tell from that if it is the latter.

:Niranjan.

Dump the whole color blocking thing. Unless you have the equipment to measure your neg, color blocking is a waste of time. You won't get a good result since somewhere there will be a transition from the color to black and that won't go well. It is obvious from your example. Color blocking is a nice "theory" but when the rubber hits the road it is pretty useless I think. It creates more problems than it solves.

You've established that with your Canon dye printer the matte setting blocks the most UV, which appears to be significantly more than the regular black. You are almost there. The only problem I forsee is if Canon uses a pigment black for matte. You might have the same transition problem as color blocking when the dye black takes over. you might want to look up if that is the case (if matte is pigment).

Print out a target that is 0-255. There are some on the internet. Then use it to establish a few parameters. Find your exposure (if you want max blue) by printing for the time that there is no difference between the edge of the Pictorico and the paper blue. Once you have established that printing time, you can look at all of the squares and figure out your printing range. Find where the shadow areas start to separate, and where the highlights do as well, and those are your end points. Hopefully you will still have some white otherwise your ink isn't dense enough. Once you have established your end points, put them in your curve or levels in Photoshop. You'll want to use output levels. Easier with a levels layer. Once you have your parameters set print a new target with the new parameters to make a curve. I've found the simplest way is to just use gamma, 1.8 to 2.2 works well usually for my Canon pigment printer. The higher the setting the brighter the print. I stopped using curves a while back. There is very little difference in the end print.

That is all you have to do. Easy peasy. No complicated math or LUTs or anything else. Maybe I should start a website.....

I need to redo my own parameters because since I've made them for myself a few years ago I've changed a lot of things. Maybe I'll do a tutorial but I use LAB these days so it would probably be too confusing.

I think I can agree with you in general, that the transition from color to black can be problematic. But that's the same for grayscale as well, at least for my printer. So the theory goes that you might as well pick a color that gives the longest range of shades, that way you get better linearity. I think I'm going to go through the rest of my calibration and see what's the best this can do.

Would love to see a tutorial on your method with examples.

nmp said:

Can you share the scan of the HSB print that you used to calculate the LUT.

:Niranjan.

Click to expand...

I decided to redo the HSB print with the "matte" setting. EDN is now giving me H=40 as optimum, so it's all over the place. I've seen H=330, H=270, H=220, etc. I'll have to do more samples to make sure this is right, but here's the scan: