Building custom profiles using i-1

[Filmnut]

I am trying to build a couple of custom profiles for my printer, a Canon 9000. I have access to an X-Rite I-1 Pro Spectophotometer, and Profilemaker 5 which I decided to use to try to improve on the preset Canon ones. With the heavy photo paper I like to use, the shadows tend to drop off quite quickly. So, I have been making tweeks on the file in Photoshop, or when I print through the printer driver, to get the print I want, but it would be nice to have my first print come out closer to the way I want it.
Now, I am a very experienced printer, both digitally and analogue, and do realize that quality prints don't come easy, which is why I'm trying to do this.
I have printed and analyzed the patches using the I-1, and Profilemaker, and applied them through the Photoshop plug-in, but my results are still not as good as I'd like. I've been fiddling around with this off and on for a few weeks, but I am still not getting what I want.
I do not have a rip station, I rip from my Mac to the printer.
Has anyone out there used these tools to make profiles, and if so, what advice can you give me?
Keith

[Tyler Boley]

Keith, I'm not terribly familiar with the 9000, but it's possible you need to go through thorough testing of the various media settings on the paper you are profiling first. A profile can't pull out values that are already dumped. So look for some separation down in those values when printed with no profile at all from various media settings, find one that balances separation and good density, and profile over that.
Tyler

[Ken Allen]

Hi Keith,

Have an older Gretag EyeOne Pro and was never satisfied with the results with any of my printers including the ipf9000, so I've outsourced that function.

Search the web for Madmanchan color profiles and you'll find the information for Eric Chan. He's an excellent resource.

Best,
Ken

[Filmnut]

Ken;
That's interesting, I'll check it out.
Keith

[Brian Ellis]

If I was creating my own profiles I'd use QTR (www.harrington.com). Not because I have a lot of experience creating profiles and have found QTR to be the best method, only because I'm slightly familiar with using QTR to do it. But if Eric Chan gives contradictory advice by all means do what he says, not what I say.

[pherold]

Building on what Tyler said, try some different media settings. We like looking at profile targets that are printed in a "visual" layout rather than scrambled - as you get a good idea of how the linearization plays out with the different media settings. You can tell at a glance which is printing with more shadow detail, etc.

Also, we have had great luck with our SpectroScan tables. They don't make these measurement instruments anymore, but they come with different filters - and the polarizing filter does a fabulous job pulling detail out of a canvas or matte profiing target. If you can find someone with one of these - or a Barbieri spectrophotometer which also offer polarizing filters - that's what the filter is made for, getting shadow detail into a profile. Here's a little more detail on this in one of our newsletter articles. (http://www.colorwiki.com/wiki/Custom_Profiling_Services#EQUIPMENT)

[Filmnut]

Thanks to all for your tips!
Keith

[neil poulsen]

Another problem can be not using a spectrophotometer with a UV cut filter for papers with optical brighteners, especially for glossy papers.

Pat, where would you draw the line between using the UV cut filter versus the polarizing filter? I imagine it would be for matte papers, as you mention. Which would you use for a matte paper that has optical brighteners?

Also, what target would you recommend using? I've been using the 9.19 target by Bill Atkinson, and have gotten decent results. But, I wonder if I could get better results using a different target.

[pherold]

Yes, you do have to make a choice whether to use the polarizing filter or the UV filter. I supposed I lean toward using the polarizing filter, and then let the software perform its optical brightener correction (if that's available.)

What target you use should depend on what profiling software you're using. ProfileMaker and i1Match seem to need the extra near-neutral patches that are in their targets in order to get you a good neutral gray balance. So, it's best to use a target that was made for them or by them. Monaco seems to do okay without a lot of over-sampling in the neutrals, so their targets work okay for them.

[Filmnut]

I'm using a test chart with 918 colours, but not much of anything for neutrals, it is supported by my system, but there are others, and I am wondering if a different target might give me the results I'm looking for.
Keith