ColorMunki users, paper profiles?

[buggz]

Hello,
Any other ColorMunki users here generating paper profiles?
It seems to be working really good for me.
Though, I have a question about tuning the profiles.

I know when I optimize the profiles using a source image, the gamut warning decreases a LOT when soft proofing in CS5.
I am wondering how many times can this be performed to improve the profile?
Of course, using different source files with different color pallets.
I am wondering if the improvements start to decrease with continued optimization?
And at some point is not really noticeable?
Would it be better to use test chart images for this tuning, or not?
I originally thought the optimization via an image source was just for that image.
Though, the software doesn't complain when I continue to optimize a profile with different images.
Am I wrong here?
I will have to ask Xrite support these questions and post back.
Though, in the mean time, if anyone knows something, please post.
Thanks!

[Alan Goldhammer]

I have profiled all the papers that I use with ColorMunki and find that it does a wonderful job.  I have compared every profile that I have made with the paper manufacturer's available profile using Jack Flesher's standard print (which is a derivative of Bill Atkinson's and is at:  http://www.outbackprint.com/printinginsights/pi049/essay.html ).  I found the ColorMunki profiles equivalent or in some cases better than the manufacturer's (especially with the tear-drop arch shot with the sky).  In almost every case I did not need to do any fine tuning following the standard method of profile generation.  I did have one set of landscape photos that benefited from slight tweaking and did so.

[PhilipCummins]

EdJune: Apparently (and it's been a while) the earlier versions of the ColorMunki software had issues with the optimisation so you could only do a few additional passes, this should be fixed in 1.1.1. In theory you should be able to pass multiple images of different colours (ideally, ones that aren't printing well) to attempt to get the ColorMunki software to fine tune the profile - it's been a while since I've tried this myself however. In theory successive optimisation passes with different images should improve the quality & accuracy of your printer's profile over time until there are no real improvements to be made. I'd recommend keeping a copy of pre/post optimised profiles so you can compare with soft proofing to find which one you prefer and also to see if the quality is decreasing from the increased number of optimisations.

[Scott Martin]

The Color Munki output profiles are good, and the RelCol and AbsolCol intents perform extremely well, even in demanding press proofing workflows. The Perceptual rendering, however, is a bit of a disappointment and that effects those in photography and fine art markets, especially where saturation is important. Other profiling packages, especially Monaco Profiler and it's upcoming replacement i1Profiler perform far better in this respect (and they don't support the Munki hardware). Excellent perceptual rending will be one of those lines that divide consumer and pro level solutions.

[digitaldog]

Considering the price point, the unbelievably few patches you have to measure, the product makes shockingly good profiles.

In the early 90’s, measuring one patch at a time (hundreds of patches) using a ColorTron, I couldn’t even imagine such a product <g>

[Czornyj]

Monaco Profiler and it's upcoming replacement i1Profiler perform far better in this respect (and they don't support the Munki hardware). Excellent perceptual rending will be one of those lines that divide consumer and pro level solutions.

Speaking of i1profiler - wasn't it supposed to be available a while ago? Anyone knows when at last will it be available?

[langier]

We've come a long way...1970s and 1980s with a densitometer and hours of reading step wedges, to hundreds of patches with the spectrometer to two sheets with the Monki!

It works for me, generally the first time.

[David Good]

My understanding of the Optimization process is that it does not increase the gamut volume but instead just shifts it towards the colors of the reference file. I am not sure, however, what happens when it's presented with complimentary colors.  

[gdi]

I have an Eye One Photo set that I bought a number of years ago, can it compare with the ColorMunki for paper profiles?  Has time and technologies made the i1 obsolete?  I have been using vendor and Atkinson profiles for my old 7600,  but I will be getting a Canon ipf and foresee the need to create my own profiles. 

Thanks

[Rhossydd]

Speaking of i1profiler - wasn't it supposed to be available a while ago? Anyone knows when at last will it be available?

"X-Rite previously announced a ship date of mid-November for the new i1 Professional Solutions. However, based on feedback from independent industry experts who serve as beta testers and X-Rite’s own testing, several useful enhancements are currently being incorporated into the product. The new i1 Professional Color Management Solutions are now expected to ship in February 2011."
http://www.xrite.com/custom_page.aspx?PageID=200

[digitaldog]

Speaking of i1profiler - wasn't it supposed to be available a while ago? Anyone knows when at last will it be available?

It will be available when we are finished with it <g>

Seriously, you don’t want to rush this process, you want it to be done right.

[digitaldog]

I have an Eye One Photo set that I bought a number of years ago, can it compare with the ColorMunki for paper profiles? 

The good stuff about Munki is more about the profiling technology than the instrument. IOW, if you could use the EyeOne Pro to read the 100 patches and use the new color engine, it would probably be very similar. There are some newer instrument technology inside the Munki but the EyeOne is still a very useful product I think you’ll be using for awhile.

[gdi]

Thanks Andrew - finally something that paid a small fortune for still has utility!

[buggz]

Hello,
Thanks for the replies and discussion.
I have the following information from another group:
>>
From my field experience 3 to 5 times refinement target is ok. After that number the profile starts to degenerate and create bumps and holes (when viewed with Gamutvision for example) and noticeable banding.
I always use portraite pictures to optimize profile for portrait and B&W+Landscapes to optimize for other pictures

And a site w/ really good review/information, and some verbiage from Xrite:
http://www.northlight-images.co.uk/reviews/profiling/colormunki_printing.html

And, then the reply I received from Xrite:
Ed,
Here is a help desk article written to address your question:
X-Rite Document: http://www.xrite.com//product_overview.aspx?ID=1115&Action=support&SupportID=4430