[argyllcms] Re: Number of patches - well behaved printer?

  • From: Terry Ritz <t.ritz@xxxxxxxxxxxxx>
  • To: <argyllcms@xxxxxxxxxxxxx>
  • Date: Wed, 23 Feb 2011 01:08:53 -0600

> edmund ronald wrote:
>>   I wasn't asking you, who are a specialist doing proofing; I was
>> asking the OP who appears to be a photographer doing fine art work.
> Edmund, nowhere did the OP (Terry Ritz) specify what the end use was,
> nor did the post from Don Craig that you responded to. So you
> can't assume that it is or is not photography, or even
> one and not the other. Now proofing tends to be the most critical
> use because there is a reference print to objectively and visually
> match to, but I hardly think that it's unjustifiable that some
> Photographers aspire to produce the best possible prints as well.
> Graeme Gill.

To answer the question above, I am indeed a photographer doing fine art
work. Having said that I'm very fussy, and I'd like to get the most out of
the printer that I can. I've never made a profile from patches that didn't
sit for a day. It's just not something I would even think of doing. I also
print black and white work on a separate printer using dedicated monochrome
inks and QTR.

I print my own work, but I also produce prints for other photographers. This
is a part-time endeavour, which is why I opted for a ColorMunki rather than
an EyeOne. My plan was to start with the ColorMunki software and then move
up to an Argyll solution. I felt that this combination would get me
reasonably close to the higher-end product. So far it seems to be doing
that. I particularly like that I can target the printer profile for a
specific file working space. I use Joseph Holmes Dcam, which is a larger
working space that is a bit outside of the mainstream.

So, I am new to Argyll, and also new to the Canon 8300. A recent upgrade
from an Epson 3800 was to facilitate making canvas prints, and also to give
me a better platform for my b/w work (the 3800 will be redeployed).

I initially made CM profiles for my papers on the 8300. So far I have also
made Argyll profiles for 3 of 5 papers. With two papers (Epson Hot Press
Natural and Breathing Color Lyve canvas) my Argyll profiles are in and
around 5% larger in gamut, with smoother gamut shapes, than the CM profiles.
However, I am having problems with a fiber based gloss media, Cone Type 5.

Type 5 is a Jon Cone product. It's very similar to Silver Rag (no OBA's /
warm base colour) but with a baryta coating. The paper is very sharp and
warm, with a beautiful artistic texture.

On my first attempt at profiling this paper using the ArgyllCMS I had some
very high read errors. I have been using a guide consisting of a mylar base
with a slit for the CM aperture, framed on either side by rulers to guide
the instrument. The CM was binding and my read was not smooth, so of course
I had errors. I addressed this and on the second attempt the error rate was
much lower, but the profile had a blue cast. I recently finished reading the
same patches a third time and things are looking better now, but the gamut
is smaller than the CM profile's gamut.

This is part of why I asked "how many patches". The Type 5 profile has been
created using a target with 1053 patches. The gamut is smaller than the CM's
software profile, and I had the following errors reading it:  peak err =
4.117913, avg err = 0.643046, RMS = 0.791215.

With Epson HPN and Lyve canvas I used 1176 patches, and had the following
read errors:

(HPN) peak err = 2.039644, avg err = 0.383221, RMS = 0.451281
(Lyve) peak err = 2.521818, avg err = 0.502227, RMS = 0.609855

I realize that the error rate may not directly relate to the number of
patches. More importantly, perhaps the number of patches chosen was too few
to properly determine the gamut of this paper. Then again, there may
considerations I'm unaware of when profiling fiber gloss media.

I have appreciated everyone's thoughts and suggestions. Any other insights
on profiling Type 5 would be appreciated.


Other related posts: