Canon's take on Neutral Rendering: cobra snakes forever?

[vikcious]

Hi there,

I am closely following on the efforts Doug Gray is putting into taming this PRO series of Canon ImagePrograf. What I am trying to accomplish is to get an as good / smooth as possible Neutral Rendering custom profiles I am building for my Canon PRO-2000 using various (brand and unknowns) types of papers. In assessing the outcome I am using both 'psychophysical' and metrics methods = eyes & tools. For the tools I am very much relying on the ColorThink PRO capabilities, ArgyllCMS, i1Profiler and other tools.

The ColorThink PRO remains the tool of choice and while I am still learning about its features and how it can be used to proof profiles I am still able to draw some not so magical conclusions one of them being...

The typical already snake / cobra shaped curves of all Canon's profiles linearity evaluations and how, no matter how big or small the name (Hahnmuhle, Canson, etc.) being a profile created for the Canon's ImagePrograf series, the result remains ALWAYS the same. See attachements for Canon and EPSON situation.

At the same time, profiles created for the EPSON printers (their professional line of photo printers) is always waaaaay more behaved and better aligned eventually to the neutrals / near neutrals axis, irrespective of the model.

I don't have an EPSON printer at hand to actually visually compare the same picture printed using Canon distorted cobra-shaped curve and one well behaved profile.  Nonetheless, upon visually inspecting my prints (especially the BW ones) I do feel that more could be improved from both the nuances neutrality to the smooth gradation. The darker a gradation becomes, the more prone to visible "steps" it becomes.

So... questions: are these Canon's cobra-shaped curves something that could be straightened up or should I go on and live my life the way it is and abandon this compulsive take on printing perfectionism???

 

[Alan Goldhammer]

Isn't the more proper question, "...is there any visual difference between images printed on the comparable Epson printer?"  I moved from Epson to Canon earlier this years (3880 to Pro-1000) and really don't see much of a difference at all.  I do my own profiling with Argyll and when I compared the Epson profiles with those Canon ones for the same papers, I noticed slightly larger gamut for the Canon but in general no other major difference.   I use various Argyll tools for analysis as I can't rationalize purchasing software that isn't critical to what I do.