Sign in with
Sign up | Sign in
Your question

ICC profiles, "print readiness" & Inkjet vs Lightjet

Last response: in Digital Camera
Share
August 16, 2005 11:42:06 PM

Archived from groups: rec.photo.digital (More info?)

My experience is mainly in designing graphics for the web, but one client
wants some large format print items produced from digital photos. Two areas
I need resources in are ICC color profiles for computers (I'm assuming this
is the area that allows you to "see" your graphics on the computer in a more
"output-accurate" format according to your end product). I would be willing
to invest in software that could help manage this in a "non-complex" way.

Also, information in preparing graphics for printing, (color seperations,
pantone, vector vs bitmap, bit-depth, pixel resolution of image vs output
size, etc).

If anyone knows of internet resources that can help prime a person in these
subjects, it would be much appreciated.

One additional: any comments of output quality comparisons from those
experienced in Inkjet vs "Lightjet" processes?

thanks for any input....

Norman
Anonymous
August 16, 2005 11:42:07 PM

Archived from groups: rec.photo.digital (More info?)

> Norman writes ...
>
>Two areas I need resources in are ICC color profiles for
>computers

You need a color managed program like Photoshop and you need to
calibrate your monitor accurately, ideally with something like the
Gretag Eye-One or the Spyder2 or similar. If your monitor is good (ie
fairly new and not cheap) you can soft proof in Photoshop to printer
profiles and have a fairly good idea of what the print will look like.
It sounds easy enough but there are a lot of problems getting good
profiles ... anyway, these sites will help understand the basics ...

http://www.creativepro.com/story/feature/13605.html (basic idea behind
color management)
http://www.creativepro.com/story/feature/10150.html?ori... (soft
proofing concept)

>I would be willing to invest in software that could help manage
>this in a "non-complex" way.

Once you get the monitor calibrated and characterized and are used to
Photoshop it's fairly transparent, but getting there is not exactly
'non complex' ...

>One additional: any comments of output quality comparisons from
>those experienced in Inkjet vs "Lightjet" processes?

The prevailing wisdom seems to be that highest quality LightJet prints
(from a $130,000 laser printer) are *slightly* superior to Epson
9600/7600/4000 prints when viewed side by side (the Epson printers
range from $5,000 - $1,700 for comparison). The new Epson
9800/7800/4800 printers supposedly reduce the margin even more.

Since beauty is in the eye of the beholder I would urge you to get
sample prints of the same file(s) and examine them closely side-by-side
.... you can get sample prints from an Epson 9600 and a Chromira (poor
man's LightJet, only $40,000) cheaply from this well-known lab ...
http://www.westcoastimaging.com/wci/page/services/more/...
.... there's quite a bit of other interesting info on their site as well
.... you can also send the same image to Calypso for a Lightjet sample
print to compare ... http://www.calypsoinc.com/ ... both of these labs
have printer profiles available so you can soft proof in advance.
There are good discussions comparing the Epson 9600 inkjets to the
laser printers on both of these sites.

If you're doing the WCI sample pack I'd recommend the Fuji Crystal
Archive Matte on the Chromira and Epson Premium Luster Photo Paper with
the 9600 for photo-realistic papers and maybe the Hahnemuhle Photo Rag
308 for a nice fine-art paper.

Bill
Anonymous
August 17, 2005 12:59:46 AM

Archived from groups: rec.photo.digital (More info?)

Bill Hilton wrote:

>>Norman writes ...
>>
>>Two areas I need resources in are ICC color profiles for
>>computers
>
>
> You need a color managed program like Photoshop and you need to
> calibrate your monitor accurately, ideally with something like the
> Gretag Eye-One or the Spyder2 or similar. If your monitor is good (ie
> fairly new and not cheap) you can soft proof in Photoshop to printer
> profiles and have a fairly good idea of what the print will look like.
> It sounds easy enough but there are a lot of problems getting good
> profiles ... anyway, these sites will help understand the basics ...
>
> http://www.creativepro.com/story/feature/13605.html (basic idea behind
> color management)
> http://www.creativepro.com/story/feature/10150.html?ori... (soft
> proofing concept)

Bill,
These are great articles. Can anyone give a link to the
mathematical/table of data for the 1931 CIE Standard Observer
for color vision and making the famous CIE diagram?

A note on ICC profiles. I've had occasional problems with
posterization, particularly in skies when applying ICC
profiles going from 16 bit input to 8-bit lightjet ready
files for Fuji Crystal Archive paper. In those cases, I had to
stop the use of ICC profiles and do it by hand. Has anyone else
seen a problem like this? It has only occurred on
scans of 4x5 transparencies (both drum scans and
Epson 4990; 8 and 16-bit files). I'm using Photoshop CS.

In particular, this image:
http://www.clarkvision.com/galleries/gallery.landscape-...
I've made marvelous lightjet cibachromes from a drum scan. Now that
cibachrome is no longer available on the lightjet (from the
local lab I use), I have made very nice but not what I consider
great prints on Fuji Crystal Archive paper. I just can't seem to get
the green as good as cibachrome (I have gotten what I consider
close enough on other images). The ICC profile posterized
both 8 and 16-bit input. So I've been doing it by hand.
(Most people, having never seen the cibachrome could not tell
any difference or that the greens were off.)

Roger

>
>>I would be willing to invest in software that could help manage
>>this in a "non-complex" way.
>
>
> Once you get the monitor calibrated and characterized and are used to
> Photoshop it's fairly transparent, but getting there is not exactly
> 'non complex' ...
>
>
>>One additional: any comments of output quality comparisons from
>>those experienced in Inkjet vs "Lightjet" processes?
>
>
> The prevailing wisdom seems to be that highest quality LightJet prints
> (from a $130,000 laser printer) are *slightly* superior to Epson
> 9600/7600/4000 prints when viewed side by side (the Epson printers
> range from $5,000 - $1,700 for comparison). The new Epson
> 9800/7800/4800 printers supposedly reduce the margin even more.
>
> Since beauty is in the eye of the beholder I would urge you to get
> sample prints of the same file(s) and examine them closely side-by-side
> ... you can get sample prints from an Epson 9600 and a Chromira (poor
> man's LightJet, only $40,000) cheaply from this well-known lab ...
> http://www.westcoastimaging.com/wci/page/services/more/...
> ... there's quite a bit of other interesting info on their site as well
> ... you can also send the same image to Calypso for a Lightjet sample
> print to compare ... http://www.calypsoinc.com/ ... both of these labs
> have printer profiles available so you can soft proof in advance.
> There are good discussions comparing the Epson 9600 inkjets to the
> laser printers on both of these sites.
>
> If you're doing the WCI sample pack I'd recommend the Fuji Crystal
> Archive Matte on the Chromira and Epson Premium Luster Photo Paper with
> the 9600 for photo-realistic papers and maybe the Hahnemuhle Photo Rag
> 308 for a nice fine-art paper.
>
> Bill
>
Related resources
Anonymous
August 17, 2005 1:24:35 AM

Archived from groups: rec.photo.digital (More info?)

For whatever it's worth, I prefer Lightjet. I like the look of
chemically developed photographic image (vs. inkjet). You also avoid
things like differential gloss and 'bronzing'.
The newer inkjets are apparently better this way, but there is nothing
worse than looking off angle at an inkjet and seeing a wierd color
reflection off the darker areas of the image (due to light reflecting
off the surface of the dense ink....i.e. 'differential gloss'). It
looks really cheap, reminiscent a low quality offset print job at times.
Anonymous
August 17, 2005 11:02:47 AM

Archived from groups: rec.photo.digital (More info?)

>Roger Clark writes ...
>
>A note on ICC profiles. I've had occasional problems with
>posterization, particularly in skies when applying ICC
>profiles going from 16 bit input to 8-bit lightjet ready
>files for Fuji Crystal Archive paper ... Has anyone else
>seen a problem like this?

Hi Roger,

When I got my Epson 4000 I printed some test images on several papers
using various profiles and on one image with smooth gradients of color
I saw weird things like this on the worst profiles. Here's the image
that gives me the most problems ...
http://members.aol.com/hiltonfotography/bill_abstracts/... ...
start at the red flower at the bottom-left and go up to the 3rd and 4th
red blossoms straight above it ... on the really bad profiles these two
blooms actually blend together and there are discrete steps between
shades of red in other blooms instead of smooth gradient (you can't see
the gradations as well in this jpeg which was converted to sRGB but in
the better prints with excellent profiles it's more noticeable).

I talked to the guy who generated the worst profile (3rd party paper,
good quality paper but they are cutting corners with the profile
generation) and could see why what he did might cause this (among other
sins he used 'saturation' rendering instead of 'perceptual' and
'relative colormetric'). I actually get better prints with that paper
when using Epson profiles for a similar Epson paper.

At any rate, I would suspect the profile myself, assuming you tried
different rendering intents when you converted. If you can see this
posterization on the screen (ie, before you print but when you soft
proof) then I'd suggest downloading the Calypso LightJet profiles and
see if those show it. Bill Atkinson creates the Calypso profiles and
Bill is pretty much a digital genius at this stuff. If you can't see
the problem on screen but only when you print then I'd pre-flight a
file and have it printed at Calypso and see if their LightJet and their
profiles show the same problem compared to the prints you're getting
now.

Bill
Anonymous
August 17, 2005 6:52:38 PM

Archived from groups: rec.photo.digital (More info?)

> nobody writes ...
>
>Bill, what cannot I understand is how you can survive without a
>Gretag Macbetgh Eye One profiling device.

You mean the $1,000 printer profiler? A friend has one and generated
two profiles for me, one for an Epson paper and one for a 3rd party
paper ... the Epson 4000 profile for the Epson paper was actually
better, I thought, while the Gretag profile for the 3rd party paper was
much better than the profile supplied by that paper's manufacturer. So
basically if you are printing on the Epson papers with the Epson
profiles for the 4000 you don't need this. In the past few months
Epson has brought out 17x22" sheets for three of their better papers so
there's less need for 3rd party papers for me now.

>I couldn't live without one.

Which printer did you buy, the 7600? Which papers are you printing on?

Bill
August 17, 2005 9:08:28 PM

Archived from groups: rec.photo.digital (More info?)

Bill, what cannot I understand is how you can survive without a Gretag
Macbetgh Eye One profiling device. It does cost an arm and a leg, but
without it one cannot walk... :-). I am doing far less photography than you,
at least at the moment, however, I couldn't live without one.
nobody


On 17/8/05 15:02, in article
1124287367.181738.75270@f14g2000cwb.googlegroups.com, "Bill Hilton"
<bhilton665@aol.com> wrote:

>> Roger Clark writes ...
>>
>> A note on ICC profiles. I've had occasional problems with
>> posterization, particularly in skies when applying ICC
>> profiles going from 16 bit input to 8-bit lightjet ready
>> files for Fuji Crystal Archive paper ... Has anyone else
>> seen a problem like this?
>
> Hi Roger,
>
> When I got my Epson 4000 I printed some test images on several papers
> using various profiles and on one image with smooth gradients of color
> I saw weird things like this on the worst profiles. Here's the image
> that gives me the most problems ...
> http://members.aol.com/hiltonfotography/bill_abstracts/... ...
> start at the red flower at the bottom-left and go up to the 3rd and 4th
> red blossoms straight above it ... on the really bad profiles these two
> blooms actually blend together and there are discrete steps between
> shades of red in other blooms instead of smooth gradient (you can't see
> the gradations as well in this jpeg which was converted to sRGB but in
> the better prints with excellent profiles it's more noticeable).
>
> I talked to the guy who generated the worst profile (3rd party paper,
> good quality paper but they are cutting corners with the profile
> generation) and could see why what he did might cause this (among other
> sins he used 'saturation' rendering instead of 'perceptual' and
> 'relative colormetric'). I actually get better prints with that paper
> when using Epson profiles for a similar Epson paper.
>
> At any rate, I would suspect the profile myself, assuming you tried
> different rendering intents when you converted. If you can see this
> posterization on the screen (ie, before you print but when you soft
> proof) then I'd suggest downloading the Calypso LightJet profiles and
> see if those show it. Bill Atkinson creates the Calypso profiles and
> Bill is pretty much a digital genius at this stuff. If you can't see
> the problem on screen but only when you print then I'd pre-flight a
> file and have it printed at Calypso and see if their LightJet and their
> profiles show the same problem compared to the prints you're getting
> now.
>
> Bill
>
Anonymous
August 18, 2005 2:32:16 AM

Archived from groups: rec.photo.digital (More info?)

Bill Hilton wrote:

> At any rate, I would suspect the profile myself, assuming you tried
> different rendering intents when you converted. If you can see this
> posterization on the screen (ie, before you print but when you soft
> proof) then I'd suggest downloading the Calypso LightJet profiles and
> see if those show it.

Hi Bill,
Yes, I can see the problem on screen. I'll try some other profiles.
Thanks
Roger

> Bill Atkinson creates the Calypso profiles and
> Bill is pretty much a digital genius at this stuff. If you can't see
> the problem on screen but only when you print then I'd pre-flight a
> file and have it printed at Calypso and see if their LightJet and their
> profiles show the same problem compared to the prints you're getting
> now.
>
> Bill
>
Anonymous
August 18, 2005 3:56:02 AM

Archived from groups: rec.photo.digital (More info?)

"Roger N. Clark (change username to rnclark)" <username@qwest.net>
wrote in message news:4302A822.6010004@qwest.net...
SNIP
> Can anyone give a link to the mathematical/table of data for the
> 1931 CIE Standard Observer for color vision and making the famous
> CIE diagram?

This should help:
http://www.fho-emden.de/~hoffmann/ciegraph17052004.pdf
and from that document you can use:
http://www.fho-emden.de/~hoffmann/ciesuper.txt
to produce a Postscript version in Photoshop.
And the following may be of background interest as well:
http://www.fho-emden.de/~hoffmann/ciexyz29082000.pdf
http://www.fho-emden.de/~hoffmann/cielab03022003.pdf

Other good sources for colormanagement calculations and formulae:
http://www.brucelindbloom.com/index.html?ColorCalculato...
http://www.efg2.com/Lab/Library/Color/Science.htm

A good source for Color matching function data and chromaticity
coordinates:
http://www.cvrl.org/

> A note on ICC profiles. I've had occasional problems with
> posterization, particularly in skies when applying ICC
> profiles going from 16 bit input to 8-bit lightjet ready
> files for Fuji Crystal Archive paper. In those cases, I had to
> stop the use of ICC profiles and do it by hand. Has anyone else
> seen a problem like this?

Printer profiles need to be made with a large number of control
patches. The better printer profiles I can make (EyeOne Photo
spectrophotometer) are based on at least 918 patches.

Bart
Anonymous
August 18, 2005 3:56:03 AM

Archived from groups: rec.photo.digital (More info?)

Bart van der Wolf wrote:
>
> "Roger N. Clark (change username to rnclark)" <username@qwest.net> wrote
> in message news:4302A822.6010004@qwest.net...
> SNIP
>
>> Can anyone give a link to the mathematical/table of data for the 1931
>> CIE Standard Observer for color vision and making the famous CIE diagram?
>
>
> This should help:

Thanks Bart.
This is great stuff! I found this link from one of yours:
http://members.cox.net/astro7/color.html
It includes programs.

Roger

> http://www.fho-emden.de/~hoffmann/ciegraph17052004.pdf
> and from that document you can use:
> http://www.fho-emden.de/~hoffmann/ciesuper.txt
> to produce a Postscript version in Photoshop.
> And the following may be of background interest as well:
> http://www.fho-emden.de/~hoffmann/ciexyz29082000.pdf
> http://www.fho-emden.de/~hoffmann/cielab03022003.pdf
>
> Other good sources for colormanagement calculations and formulae:
> http://www.brucelindbloom.com/index.html?ColorCalculato...
> http://www.efg2.com/Lab/Library/Color/Science.htm
>
> A good source for Color matching function data and chromaticity
> coordinates:
> http://www.cvrl.org/
>
>> A note on ICC profiles. I've had occasional problems with
>> posterization, particularly in skies when applying ICC
>> profiles going from 16 bit input to 8-bit lightjet ready
>> files for Fuji Crystal Archive paper. In those cases, I had to
>> stop the use of ICC profiles and do it by hand. Has anyone else
>> seen a problem like this?
>
>
> Printer profiles need to be made with a large number of control patches.
> The better printer profiles I can make (EyeOne Photo spectrophotometer)
> are based on at least 918 patches.
>
> Bart
August 18, 2005 3:25:54 PM

Archived from groups: rec.photo.digital (More info?)

winhag@yahoo.com wrote:
> For whatever it's worth, I prefer Lightjet. I like the look of
> chemically developed photographic image (vs. inkjet). You also avoid
> things like differential gloss and 'bronzing'.
> The newer inkjets are apparently better this way, but there is nothing
> worse than looking off angle at an inkjet and seeing a wierd color
> reflection off the darker areas of the image (due to light reflecting
> off the surface of the dense ink....i.e. 'differential gloss'). It
> looks really cheap, reminiscent a low quality offset print job at times.
>
Really, this issue is solved. You don't see it with Canon / HP dye
photo printers on gloss/semi gloss media or with the R800/1800 with GOP,
and it is apparently a non -issue with the R2400 /4800 etc.

For a comparison of the subjective qualities you may find the following
link interesting:
http://www.luminous-landscape.com/reviews/printers/ligh...

Now, to put that in perspective, it was written over 5 years ago. Huge
advances have been made in quality of output from photo quality inkjet
printers by all manufacturers since then. There is no contest.
!