Dan Margulis Applied Color Theory

Adobe RGB for Digital Projectors?

Adobe rgb/digital projection
Posted by: "Ruth Brown"
Thu Mar 12, 2009 6:37 am (PDT)

All, I don't usually post here as I'm usually overwhelmed by the technical expertise and am fully occupied digesting all the knowledge on offer. I'm not a professional but a serious amateur photographer and am involved in writing PS actions for our local members to convert image files to a size/format required for digital projection competitions. The accepted standard here is to use Adobe RGB as most photographers use that space and it is considered less confusing to remain with that. However, I have received a missive from a fellow who is on the 'working committee' to try and standardise things right across the UK and he says: " I am now committed to pushing sRGB as the space to use from camera/scanner to print/projection for all purposes, and for 99% of all authors. Somewhere along the line, people have got the idea that AdobeRGB is better because it is larger, and better for prints. Both perceptually and by photometric measurements, the research I have seen suggests that any extra colours in the AdobeRGB space are rarely present in typical photographs. And, the 'prints' that AdobeRGB was created for are commercial CMYK press output, and not RGB photographic prints.. Using a straight through sRGB colour space is just so much simpler."

I'm not at all sure this is correct and would welcome any comments particularly his assertion about the 'extra' colors and the reference to cmyk press output. Ruth Brown
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: "Jim Bean"
Thu Mar 12, 2009 8:21 am (PDT)

Using a straight through sRGB colour space is
just so much simpler.
hello ruth,
and for many reasons that you did not list... sRGB is the solution to your quest...end of story.

jim bean
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: "Alex Kent"
Thu Mar 12, 2009 8:31 am (PDT)

i'm not going to address the whole AdobeRGB vs. sRGB question, but in the case of your practical application i can answer.

if i understand you correctly: you're preparing a Photoshop Action to automate preparation of images for display via digital projection.

in this specific case, you should convert the images to sRGB in the action.

i suggest this because most software for making slideshows (microsoft Powerpoint for example) doesn't understand colour profiles at all. converting images to sRGB in advance of making the slideshow will reduce the colour change between Photoshop and whatever software is running the slideshow.

someone else can take up aRGB vs. sRGB !

alex kent.
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Michael Jahn
Thu Mar 12, 2009 8:32 am (PDT)

Hi Ruth,

In some ways, you do not have the basic rationale correctly as to why Adobe was asked to develop the "Adobe 1998 RGB" profile in the first place, but regardless of that, you seem to have come to the correct conclusion - sRGB is for doing the best you can, and Adobe RGB is for limiting the color to better simulate what might happen on a printing press - it is not that Adobe RGB offers "more" so much as it is designed to 'change' the colors to look more like it will look when it prints on a high speed printing process where we squish sticky colored oil onto crushed trees.

One can get VERY technical VERY quickly in explaining why one might pick sRGB over Adobe RGB (or, in many cases, why many people insist on Adobe RGB and avoid sRGB at all costs!) - but I like to put it this way...

Why sRGB ? --> If I were taking a Photograph a portrait of you wearing a very saturated Green and DEEP BLUE hat at the kentucky derby, and we were going to go to some RGB friendly ink set commonly found in high end modern inkjet printing systems - I am not in the least bit interested in doing anything EXCEPT to tease the most out of that raster data file and get that on that print.

Why Adobe RGB ? --> If I were going to SELL that hat in a CATALOG, and I am faced with the fact that the catalog will be printed on somewhat 'less than photo-quality' paper, using a high speed heatset offset printing process with SWOP inks and all the images are SCREENED using not so tiny dots that are spaced at 175 dots every inch and each color laid down with the dots at different angles so they do not create a moire patten - and I had MANY OTHER COLOR IMAGES on that same page (some very light pastel hats, some very deep dark toned hats) -- well, I would want to LIMIT the color to FIT into the possible achievable color gamut of this VERY narrow color space, and a great way to START is to immediately assign Adobe 1998 RGB profile so i can begin to see IMMEDIATELY what color and contrast adjustments I might need to do to retain, as much as possible, what colors i can.

Now, i specifically select Green and DEEP BLUE, as there is a tendency of deep refles-ish blues to move toward purple in the CMYK SWOP printing press world, and a saturated green is nearly impossible to reproduce using yellow and cyan. Now, the SOONER you - as a pre-press professional - can see that, well that would better. There are many issues with printing a product catalog that you would never care about - for example, if there model was wearing a pair of dark brown corduroy pants - can I simulate how that will come out on press (I may have to make a mask and increase the contrast to bring out that corduroy pattern) - you see, this is why some of us INSIST on Adobe RGB - it tells us better what will happen using the final print method BETTER.

So, there you have it - 1 image, two different INTENTS (you may have heard out PDF/X-1a and "output intents"- different as night and day as far as I am concerned, and many might suggest that these two scenarios are not easily satisfied simultaneously.

If you would like to better understand some of the history behind why we are here, and what one might to to 'fix' that, I would invite you (or anyone) to read some of Dr. Ed Granger's whitepapers of this subject...

http://www.iqcolour.com/res_whitepapers.html

- most relevant would be the the Slone paper (at the bottom) - This papaer contains the hypothesis and approach behind the IQ Colour system. Dr. Granger discusses how things might have been if the printing industry had never seen a film or color separation. Would it be possible to achieve a color vision that more accurately represented the artist’s (or photographers) intent? related to "ready for magazine and newspaper" imaging standards (international ones anyway) -->
http://www.gwg.org/digitalphotography.phtml

BTW - there is far more to understand than sRGB and Adobe RGB -->
http://www.brucelindbloom.com/WorkingSpaceInfo.html

Hope this helps.

Michael Jahn
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: "Andrew Haley"
Thu Mar 12, 2009 9:44 am (PDT)

Ruth Brown writes:

All, I don't usually post here as I'm usually overwhelmed by the
technical expertise and am fully occupied digesting all the
knowledge on offer. I'm not a professional but a serious amateur
photographer and am involved in writing PS actions for our local
members to convert image files to a size/format required for
digital projection competitions. The accepted standard here is to
use Adobe RGB as most photographers use that space and it is
considered less confusing to remain with that. However, I have
received a missive from a fellow who is on the 'working committee'
to try and standardise things right across the UK and he says:

"I am now committed to pushing sRGB as the space to use from
camera/scanner to print/projection for all purposes, and for 99% of
all authors. Somewhere along the line, people have got the idea
that AdobeRGB is better because it is larger, and better for
prints. Both perceptually and by photometric measurements, the
research I have seen suggests that any extra colours in the
AdobeRGB space are rarely present in typical photographs.

I would seriously question such an assertion from someone committed to high-quality work, which presumably includes everyone entering competitions. I can print well outside the sRGB gamut, and it is one of the things that differentiates a gallery-quality print from something you got made at the corner drugstore.

William Eggleston's prints, for example, often used very high colour saturation and he went to a tremendous amount of trouble to get it. Our expert on the working committee seems to be suggesting that such a thing would be inappropriate in a UK competition.

And, the 'prints' that AdobeRGB was created for are commercial CMYK
press output, and not RGB photographic prints.

What is an "RGB photographic print"? I'm pretty sure there is no such thing. Wet colour prints uses CMY dyes, and inkjets CMYK.

Using a straight through sRGB colour space is just so much
simpler."

I'm not at all sure this is correct and would welcome any comments
particularly his assertion about the 'extra' colors and the
reference to cmyk press output.

Adobe RGB is large enough for almost all CMYK press output, but what your man isn't telling you is that high-quality desktop inkjet printers can do better than CMYK press output, and often *much*
better.

The nicest thing about sRGB for projection is that many projectors have an sRGB mode but they don't have an Adobe RGB mode. Also, they probably (almost certainly) aren't capable of displaying Adobe RGB anyway -- they just don't have the gamut. There is no point in using Adobe RGB for anything that you know will be digitally projected.

Finally, ITU-R Recommendation BT.709 is the standard for HDTV, and it requires the sRGB colour space.

Andrew Haley.
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: "Paul D. DeRocco"
Thu Mar 12, 2009 12:21 pm (PDT)

The ideal is for the program that displays the images via the projector to be a color managed application with a proper profile for each projector that's actually used. However, this may not be practical, unless you're talking about prepping pictures for use with a single projector owned by a photo club.

Fortunately, many projectors (and regular displays) have a color space that is pretty close to sRGB, and even if they're not, many have a mode that converts sRGB input to the proper gamut. This means that you're probably best off sending sRGB to the projector.

That said, you can do that in two ways: you can use a non-color-managed application for your slide shows, and ensure that the pictures are in sRGB; or you can use a color-managed application and set the output profile to sRGB, in which case it will properly convert from Adobe RGB or whatever anyone gives you.

Ciao, Paul D. DeRocco
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: J Walton
Thu Mar 12, 2009 1:24 pm (PDT)

On Thu, Mar 12, 2009 at 7:27 AM, Michael Jahn wrote:

sRGB is for doing the best you can, and Adobe RGB is for
limiting the color to better simulate what might happen on a printing
press - it is not that Adobe RGB offers "more" so much as it is
designed to 'change' the colors to look more like it will look when it
prints on a high speed printing process...

I'm a little confused by this. sRGB is the smaller space which more closely matches SWOP in overall size, and Adobe RGB is the larger space. If you really want to limit the range of colors to be mostly what you would find in print you would go with sRGB. If you want to encompass *almost* all of the colors you could reproduce with SWOP and can live with extra gamut that you'll never be able to use, go with Adobe RGB. I don't think Adobe RGB was "designed" to have anything to do with the printing process - it has just become the de facto standard, largely because it starts with the word "Adobe." The fact that it's a decent choice for print has probably kept it in place as the standard.

Why sRGB ? --> If I were taking a Photograph a portrait of you wearing
a very saturated Green and DEEP BLUE hat at the kentucky derby, and we
were going to go to some RGB friendly ink set commonly found in high
end modern inkjet printing systems - I am not in the least bit
interested in doing anything EXCEPT to tease the most out of that
raster data file and get that on that print.

I would think if you were using an RGB-friendly ink set and could go way beyond the gamut of CMYK you'd want to go with Adobe RGB *at the least* and perhaps something like ProPhoto *at the most*. sRGB would limit your possibilities for using that expanded gamut on output.

... a great way to START
is to immediately assign Adobe 1998 RGB profile so i can begin to see
IMMEDIATELY what color and contrast adjustments I might need to do to
retain, as much as possible, what colors i can.

Assigning Adobe RGB to a file that already has sRGB assigned would make things much *more* colorful. Of course, assigning it to a Pro Photo file will have the opposite effect. I would recommend loading your CMYK condition in color settings and hitting Command-Y to preview the separation results.

this is why some of us INSIST on Adobe
RGB - it tells us better what will happen using the final print method
BETTER.

As long as you are previewing the separation using a valid CMYK profile you will get a good preview no matter what your working space. I use AdobeRGB because it was chosen as the working space for our entire WorldGroup. For anything that isn't print-related I use sRGB. But it really doesn't matter that much.

J Walton
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Michael Jahn
Thu Mar 12, 2009 1:24 pm (PDT)

Ruth wrote;

And, the 'prints' that AdobeRGB was created for are commercial CMYK
press output, and not RGB photographic prints.

Andrew wondered;

"What is an "RGB photographic print"? I'm pretty sure there is no such
thing. Wet colour prints uses CMY dyes, and inkjets CMYK."

Jahn comments -- Actually, there are "RGB" inks - some Rotogravure printers very much like OpalTone

http://www.opaltone.com/

http: //pffc-online.com/technical_reports/paper_cmykrgbthe_digital_color/

But obviously this is not the sort of thing that Ruth was refering to - that she would hear people speaking of an "RGB Printer" is simply Semantics - many speak of "I have an RGB Printer" - and mean "I have a printer that you can send an RGB image to, and it prints it, so, it is an RGB printer"

http://photo.net/digital-darkroom-forum/004oSO

Most printing systems do indeed use CMY or CMYK or even CmMmYK (where there are light and darker colorants for cyan and magenta)

There are of course imaging systems that use light to expose to film and photo-sensitive paper - example -- film that is to be developed in the RA-4 process, and expose your data to that film on a Durst Lambda for example.
The film behaves much like photopaper. Agfa makes (made?) this film called Agfatrans and Agfaclear, and there are labs that offer exposing your RGB data on these films. The film could also be exposed in the traditional darkroom way from a color negative. From a quality point of view, it probably doesn't get any better than that, but it is not simple or inexpensive.

Hope this helps !
<http: //www.agfaphoto.com/en/photo/pdf/products/P-83-e5_en.pdf>
Michael Jahn
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Michael Jahn
Thu Mar 12, 2009 3:23 pm (PDT)

@ J. Walton

I don't think Adobe RGB was "designed" to have anything to
do with the printing process - it has just become the de facto
standard, largely because it starts with the word "Adobe."

jahn comment - I happen to have the Adobe Technical document named "Adobe® RGB (1998) Color Image Encoding"

- a 20 page PDF file available from download for your review;

http://www.adobe.com/digitalimag/pdfs/AdobeRGB1998.pdf

the very first sentance states;

"The Adobe® RGB (1998) color image encoding is defined by Adobe Systems to meet the demands for an RGB working space suited for print production."

sRGB is a RGB color space proposed by HP and Microsoft because it approximates the color gamut of the most common computer display devices with little to no thought about printers or printing. Since sRGB serves as a "best guess" for how another person's monitor produces color, it has become the standard color space for displaying images on the internet. sRGB's color gamut encompasses just 35% of the visible colors specified by CIE. Although sRGB results in one of the narrowest gamuts of any working space, sRGB's gamut is still considered broad enough for most color applications.

Adobe RGB 1998 was designed (by Adobe Systems, Inc.) to encompass most of the colors achievable on CMYK printers, but by using only RGB primary colors on a device such as your computer display. The Adobe RGB 1998 working space encompasses roughly 50% of the visible colors specified by CIE-- improving upon sRGB's gamut primarily in cyan-greens.

As long as you are previewing the separation using a valid CMYK
profile you will get a good preview no matter what your working space.
I use AdobeRGB because it was chosen as the working space for our
entire WorldGroup. For anything that isn't print-related I use sRGB.
But it really doesn't matter that much.

Jahn comments - there is a difference between a discussion of "working space" and "output intent". I think (for the original poster's sake) since she does not seem interested in CMYK, nor interested in previewing CMYK separations, or somehow 'looking at my stuff in a way that simulated how it might look on a printing press" - I think sRGB is her best bet.

--
Michael Jahn
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: J Walton
Thu Mar 12, 2009 5:01 pm (PDT)

On Thu, Mar 12, 2009 at 1:45 PM, Michael Jahn wrote:

I happen to have the Adobe Technical document named
"Adobe® RGB (1998) Color Image Encoding"

the very first sentance states;

"The Adobe® RGB (1998) color image encoding is defined by Adobe
Systems to meet the demands for an RGB working space suited for print
production."

I agree with what Adobe wrote here. They quite clearly define Adobe RGB to be good for print because it is the default working space for North America Prepress II. I also agree that it is good for print. But I would shy away from the word "designed" because from what I understand Adobe RGB was a bit of an accident, especially the green component. If it was truly designed for CMYK output I think it would do a better job with pure cyans and yellows. Even though they don't appear very often in nature (if at all), they DO appear often enough in printed media.

That makes me wonder how "designed" the space really was, which is what I was getting at. I tell you this, I would like to see an RGB space that just barely encompasses GRACOL, but not much more than that. I don't even know if that's possible, but it would be great for edits that will end up in CMYK. I remember some research I did years ago that suggested NTSC might be a better fit than Adobe RGB for CMYK output.

Jahn comments - there is a difference between a discussion of "working
space" and "output intent". I think (for the original poster's sake)
since she does not seem interested in CMYK, nor interested in
previewing CMYK separations, or somehow 'looking at my stuff in a way
that simulated how it might look on a printing press" - I think sRGB
is her best bet.

I agree that the original poster was not interested in CMYK. But that's not what you were talking about, and I was responding to YOUR statement. Actually, it was a single sentence that took the form of a paragraph...

Why Adobe RGB ? --If I were going to SELL that hat in a CATALOG, and
I am faced with the fact that the catalog will be printed on somewhat
'less than photo-quality' paper, using a high speed heatset offset
printing process with SWOP inks and all the images are SCREENED using
not so tiny dots that are spaced at 175 dots every inch and each color
laid down with the dots at different angles so they do not create a
moire patten - and I had MANY OTHER COLOR IMAGES on that same page
(some very light pastel hats, some very deep dark toned hats) -- well,
I would want to LIMIT the color to FIT into the possible achievable
color gamut of this VERY narrow color space, and a great way to START
is to immediately assign Adobe 1998 RGB profile so i can begin to see
IMMEDIATELY what color and contrast adjustments I might need to do to
retain, as much as possible, what colors i can.

I find this suggestion confusing. If you just want to see how something is going to print in a catalog (CMYK), why would you assign Adobe RGB to an image? It would seem that assigning a wide-gamut profile like Adobe RGB would not really "limit the color to fit into the possible achievable color gamut of this very narrow color space." Why not just preview the image using a CMYK profile suitable for your printing process?

It seems that you skipped over my questions about the relative sizes of sRGB and AdobeRGB, but it would be nice if you could flesh that out a bit. I do not say this to be argumentative, merely for clarification. There are plenty of people who follow this list who don't know as much as we do about this subject (perhaps like the original poster), and it is important from an educational perspective to be clear about our advice.

J Walton
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Michael Jahn
Thu Mar 12, 2009 7:34 pm (PDT)

On Thu, Mar 12, 2009 at 3:21 PM, J Walton wrote:

That makes me wonder how "designed" the space really was, which is
what I was getting at. I tell you this, I would like to see an RGB
space that just barely encompasses GRACOL, but not much more than
that. I don't even know if that's possible, but it would be great for
edits that will end up in CMYK. I remember some research I did years
ago that suggested NTSC might be a better fit than Adobe RGB for CMYK
output.

H J. Walton,

Well gee wiz wally ! - okay, you apparently do not like the word "designed" or atleast how i am using it.

( smile )

"With the Adobe RGB (1998) color image encoding, users can represent digital images in a color space larger than typical CRT monitors, without using overranged RGB values. This color space is well suited for professional and consumer digital photography applications."

That is also from the Adobe technical document - I read here that "this is what it is, and this is what is is supposed to be used for."

I would suggest that this is saying that "we designed this to do this" - now, one might argue if they (Adobe) were successful, but that is why they did it.

sRGB was "designed" for displays, and when that did not help us much related to print, Adobe RGB was "designed" for print - of course, they (Adobe and many others) made a few mistakes, but many like the mistakes (since this worked "better" than what they had before) so they kept them in there. Maybe you are not an engineer, and maybe you have never say on a technical committee like CGATS, and maybe you never worked for a vendor who 'develops' stuff, but then again - maybe I use the word "design" wrong.

- maybe I should use the word "engineered' or "developed" - in specifications, this means "we wrote the specification for this purpose, and here is the technical details so you could understand certain terms so you might use it."

As we know, GRACoL did not exist in 1998 when all this was developed by Adobe. Why ? well - this has to do with math. I can say that most of the math is wrong, and many color science people have been complaining.

We really can't have much of a educational 'ah-HA! moment without really understanding the math and how these colorspaces "fit" into a chromaticity diagram, that funny odd shaped leaning horseshoe looking thing that is white near the center (AKA locus) and gets more satuated toward the perimeter...

-- to me, this is a big fat unsolvable mess that requires color management experts to navigate - and I am not alone. many people have stated that it works, or even that it will work "soon" - but this is a house of cards...

http: //www.seyboldreport.com/color-management-predictions-revisited

I find this suggestion confusing. If you just want to see how
something is going to print in a catalog (CMYK), why would you assign
Adobe RGB to an image?

Why use Adobe RGB ? that is what it was "designed" to do, to enable you to see what colors are retained (or changed) or saved and what is lost (or changed) when you convert an RGB image to a PRINT based RGB image - for example, oranges get dirty.

It would seem that assigning a wide-gamut
profile like Adobe RGB would not really "limit the color to fit into
the possible achievable color gamut of this very narrow color space."
Why not just preview the image using a CMYK profile suitable for your
printing process?

Well, sometimes we do not want to edit AFTER we have CONVERTED to CMYK, and we simply want to edit while the image is in RGB, but VIEWING it in an RGB that "fits better" in the "print-like" color space.

It seems that you skipped over my questions about the relative sizes
of sRGB and AdobeRGB, but it would be nice if you could flesh that out
a bit.

okay - go here, and you can see the sizes.

http: //www.brucelindbloom.com/index.html?WorkingSpaceInfo.html

scroll down to see the "sizes" as they relate to each other. I shudder to think that everyone will suddenly start trying to work with ProPhoto RGB because it looks to be the largest one.

on the top and center is the "size" of LAB - for our discussion, that would be all the colors we can actually "see" (including florescence and stuff like that)

- the Adobe RGB (1998) is in the upper left corner of row 1.

- sRGB is the third one in of row 4

While I am not sure exactly what that tells you, and I am sure Ruth is very very sorry she even asked by now !

( smile )

She should not interested in Adobe RGB as far as I can see, as it does not provide her any solution to any problem she is facing - i only mentioned it in the manner I did so she could grasp a scenario where a photographer may pick sRGB one occasion and use Adobe RGB on another occasion. I can share that JC Penneys digital photo-retouchers use Adobe RGB as its working space and a custom profile for its output intent, and requires that their print supplier simulate that output intent, and that it is not TR001.

What they do on the web side, I have no clue, perhaps they sling it back to sRGB or something else - without Adobe FLASH supporting any idea of color management, I guess it matters little.

This is what I think about all that...

http: //www.iqcolour.com/pdf/whitepapers/IQC_ABCINPUT.pdf

on page 5 --> "In the future, all input devices can deliver either XYZ tristimulusor a colorimetrically defined RGB."

http://www.iqcolour.com/pdf/whitepapers/IQC_sloan.pdf

Page 10 -- (basically, we need to make a real world RGB)

so, i pick NEITHER. But I also would need to change the entire world to actually get my way, and that is very unlikely.

*sigh* - i am sure no one wants to hear from me again on this subject, so, i will stop now.

--
Michael Jahn
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Dan Margulis
Thu Mar 12, 2009 7:34 pm (PDT)

J writes,

I agree with what Adobe wrote here. They quite clearly define Adobe
RGB to be good for print because it is the default working space for
North America Prepress II. I also agree that it is good for print. But
I would shy away from the word "designed" because from what I
understand Adobe RGB was a bit of an accident, especially the green
component.

That is so. The Adobe paper, which was written in 2005, is revisionist. It states that the space was introduced in Photoshop 5.0.2, which is false, and indicates without stating it outright that the motivation was for print purposes, which is also false.

The space was introduced, accidentally, in Photoshop 5.0, not 5.0.2. The Photoshop team *thought* that it was using an RGB space called SMPTE-240M, and that's what it was called in Photoshop 5.0. SMPTE-240M is no more designed to be suitable for print than a Boeing 747 is designed to be an SUV. SMPTE-240M was intended as a specification for HDTV. It is, consequently, too big for print.

Unfortunately, in attempting to make this available in Photoshop, a few numbers got scrambled. So, what was identified in Photoshop 5.0 as SMPTE-240M was something else.

In Photoshop 5.0, also, the Photoshop engineers. ignoring the advice of several experts, decided that it would be a good thing for the program to, without warning, convert incoming RGB files into sRGB without notifying the user.

Shortly after the release, it came to the attention of the Photoshop engineers that 1) the experts that they had been publicly berating for doubting the wisdom of converting things into sRGB without warning had actually been right, and 2) the thing identified as being SMPTE-240M in Photoshop 5.0 was not. Hence, the emergency corrective update, Photoshop 5.0.2, which ended the warningless conversions and renamed the thing previously known incorrectly as SMPTE-240M to, for want of anything better, Adobe RGB.

That makes me wonder how "designed" the space really was, which is
what I was getting at. I tell you this, I would like to see an RGB
space that just barely encompasses GRACOL, but not much more than
that. I don't even know if that's possible, but it would be great for
edits that will end up in CMYK. I remember some research I did years
ago that suggested NTSC might be a better fit than Adobe RGB for CMYK
output.

This argument was made most strongly by the late Bruce Fraser. He thought that Adobe RGB (which he repeatedly referred to as "the former Yugoslav republic of SMPTE-240M") was too wide-gamut for CMYK work, and proposed a new space, BruceRGB, as an alternative.

In doing this, he meant, and I agree, that an RGB designed for HDTV use as modified by typographical errors is not likely to be very good for print. The paper being referred to is like the chef who tries to fry eggs, breaks the yolks, and tells the world he meant to scramble them all along.

Dan Margulis
___________________________________________________________________________

Adobe rgb/digital projection
Posted by: "Ruth Brown"
Fri Mar 13, 2009 9:36 am (PDT)

Wow! I'm peering out through the slits of my concrete bunker! I seem to indadvertently set something off here but all the posts/opinions are very interesting to read so many thanks for taking the time and effort to educate me and anyone else in the same position. No, I'm not interested in the commercial printing side of things, my main interest is in getting some slant of whether the quote I posted was a valid statement and it seems opinions are divided somewhat unless I've got totally confused. Turning things over in my mind and practically speaking, I'm asking myself, 'can I tell the difference between a projected image that is Adobe rgb and the same one that is srgb?'. Our club projectors are calibrated/profiled for the room they are used in but on doing some research, few, if any, are capable of offering a choice between color spaces, so, they're going to be displaying srgb{?}. The software that's been written for our digital competitions has the choice in the options but when I questioned the writer of the software, he admitted there wasn't really a choice! I can't say that I *can* tell the difference with any certainty, so, any action for resizing may as well change any Adobe rgb image to an srgb image as I'm seeing it. A colleague and I have actually done a set of actions to cover all the various sizings used locally, the file formats required and whether the photographer requires a white border or not and both image aspects. It looks like some of them can be re-written some time if I can convince those who must be obeyed that srgb is the way to go. As is obvious, I'm not a color guru but keen to learn nonethe less, so you'll have to forgive my simplistic terminology,
thanks all again, Ruth
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Sabo, Lori
Fri Mar 13, 2009 9:36 am (PDT)

What is an "RGB photographic print"? I'm pretty sure there is no such
thing. Wet colour prints uses CMY dyes, and inkjets CMYK

Yes, but in making prints on real old-fashioned photographic paper, the print is made by applying RGB light source, so the input to the printing device is truly and appropriately RGB.

Lori Sabo
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Michael Jahn
Fri Mar 13, 2009 9:37 am (PDT)

Dan wrote;

SMPTE-240M is no more
designed to be suitable for print than a Boeing 747 is designed to be an SUV.

That really matters little in the context of this discussion - Adobe RGB was designed to be a better space for print. That they made mistakes is really irrelevant at this juncture, as it became popular, widely used - so it was "designed" to slove a problem, and even if it happened by accident, it certainly accomplish "something".

In Photoshop 5.0, also, the Photoshop engineers. ignoring the advice of
several experts, decided that it would be a good thing for the program to,
without warning, convert incoming RGB files into sRGB without notifying the user.

As very few users would have really understood that some 'free range" RGB might be better off being forced into a 'defined" RGB, well, I think the decision was a good one, but I will agree that this would be totally unaceptable today and we are always better off with some 'warning' that something is 'changing' - although I am not really sure what that warning dialog box might have said, or if it really did explain what was happening, if many people would have understood it - and they few that did, would have said "gee, thanks" and clicked over that "OK" button ( NERD ALERT -- except me. Or maybe Terry Wyse. We would have busted out our Wyszecki &. Stiles...)

Shortly after the release, it came to the attention of the Photoshop
engineers that 1) the experts that they had been publicly berating for
doubting the wisdom of converting things into sRGB without warning had
actually been right, and 2) the thing identified as being SMPTE-240M in
Photoshop 5.0 was not. Hence, the emergency corrective update, Photoshop
5.0.2, which ended the warningless conversions and renamed the thing
previously known incorrectly as SMPTE-240M to, for want of anything better,
Adobe RGB.

Yes, and that took like what, 7 years - in which time 'what ever it was called' worked better than what we had before. Me, I was thrilled that I could modify CMYK images in Photoshop 2.5 and could have cared less about any of that back then. We were just thirlled that in 1991 we did not have to buy any more scitex imagers at 500k a pop, and we could just use Adobe Phtoshop on a Macintosh.

This argument was made most strongly by the late Bruce Fraser. He thought
that Adobe RGB (which he repeatedly referred to as "the former Yugoslav
republic of SMPTE-240M") was too wide-gamut for CMYK work, and proposed a
new space, BruceRGB, as an alternative.

With all the respect in the world for Bruce, that never really took off, and most people use Adobe RGB if they are in a print environment and using RGB at all - other just convert to CMYK and are not really all that interested in RGB anything.

The paper being referred to is like the chef who tries to fry eggs, breaks
the yolks, and tells the world he meant to scramble them all along.

Well, it is not really a "paper" as in the idea of a 'whitepaper' where they are trying to convince anyone to do anything, it is simply a technical document that describes what it is. I think we all agree that is was a goofy mistake, based on some flakey asumptions, but I would use it all day long over sRGB in most print environments.

And many do.

--
Michael Jahn
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: "Jim Bean"
Fri Mar 13, 2009 10:13 am (PDT)

hello lori,

the enlarger heads that I used were omega products and each utililzed cyan/magent/yellow gel filters... lamps were quartz of various flavors...

same thing with slide duplication... cmy... maybe some of the led printers may have output via rgb...

regards, jim bean
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: J Walton
Fri Mar 13, 2009 11:51 am (PDT)

On Thu, Mar 12, 2009 at 10:02 PM, Michael Jahn wrote:

As we know, GRACoL did not exist in 1998 when all this was developed
by Adobe. Why ? well - this has to do with math.

I get confused sometimes. What does math have to do with when GRACoL was developed?

I can say that most
of the math is wrong, and many color science people have been
complaining.

Can you be more specific? I would like to know what color scientists have been complaining about, since we sometimes do some complaining about them!

-- to me, this is a big fat unsolvable mess that requires color
management experts to navigate - and I am not alone. many people have
stated that it works, or even that it will work "soon" - but this is a
house of cards...

I don't know that we need color management experts to navigate anything. Color management experts will not agree with this, of course, but lots of people are doing high-quality work with a basic understanding of color management. What is *very* important, is an understanding of images, channels, and what looks good. And color management experts don't usually help us very much with that.

http: //www.seyboldreport.com/color-management-predictions-revisited

That's an interesting paper. Thanks for sharing that!

Why use Adobe RGB ? that is what it was "designed" to do, to enable
you to see what colors are retained (or changed) or saved and what is
lost (or changed) when you convert an RGB image to a PRINT based RGB
image - for example, oranges get dirty.

I guess I would disagree with this. I do not think AdobeRGB was designed for print, and even if it was it does NOTHING to tell you what colors are going to be retained when you print anything. Do you know how much of AdobeRGB is completely unprintable? A LOT.

Certainly *assigning* AdobeRGB is not going to tell you anything - it's just going to change the way those RGB values are interpreted. If you assign AdobeRGB to an sRGB document it will get MORE colorful. Does that tell you that when you convert to CMYK or print that catalog it will be more colorful? Of course not.

Well, sometimes we do not want to edit AFTER we have CONVERTED to
CMYK, and we simply want to edit while the image is in RGB, but
VIEWING it in an RGB that "fits better" in the "print-like" color
space.

I think I've found the misunderstanding. You can preview a CMYK conversion without actually converting it. All you do is load the proper CMYK profile and hit Command-Y (or Control-Y on a PC, I assume). Then you get a real-world preview of how that image is going to print without actually converting it. You can do this with AdobeRGB files, or sRGB files, or ColorMatch RGB files, or ProPhoto RGB files.

It actually works really well, and is a much better idea than trying to use AdobeRGB for this purpose.

scroll down to see the "sizes" as they relate to each other. I shudder
to think that everyone will suddenly start trying to work with
ProPhoto RGB because it looks to be the largest one.

I agree. Larger is not better. If everyone switched to ProPhoto most of them would switch back pretty quick.

The idea in choosing a working space is kind of like choosing running shoes. You want the smallest size you can get that comfortably accommodates your feet. A size 15 shoe will probably fit 99% of people in the world, but it's not a good choice because it's bigger than you need. A bigger house than you need doesn't cause you any real problems --a bigger running shoe does. And that's what the working space is like.

A space like ProPhoto or Wide Gamut RGB is not *that* much smaller than LAB. Using it as a working space when your final output is newspaper is ridiculous. You will *absolutely* do lower-quality work by having such a large space. It is also way too big, in my opinion, for any other CMYK work.

You can see this when you assign ProPhoto to an sRGB document. Subtle tones that looked fine in sRGB will look garish in ProPhoto. Which means that subtle curve or saturation moves that would be fine in sRGB will have a HUGE effect in ProPhoto. Now, you could always do a move and then cut back on the opacity, but it's a clumsy workflow.

I am sure Ruth is very very sorry she even asked by now !

I don't know if she is, but I know our conversation has gone beyond her original question. But that's OK, we're on this list to learn from one another. Sometimes a conversation begins in one place but ends somewhere else. As long as it is constructive then the list is serving it's purpose.

She should not interested in Adobe RGB as far as I can see, as it does
not provide her any solution to any problem she is facing - i only
mentioned it in the manner I did so she could grasp a scenario where a
photographer may pick sRGB one occasion and use Adobe RGB on another
occasion.

That makes perfect sense, and I agree that she should be converting anything that is not sRGB to that space for projection.

I can share that JC Penneys digital photo-retouchers use
Adobe RGB as its working space and a custom profile for its output
intent, and requires that their print supplier simulate that output
intent, and that it is not TR001.

That's interesting, and I think that makes sense for their work. Although they might have a better time with fleshtones in sRGB.

http://www.iqcolour.com/pdf/whitepapers/IQC_sloan.pdf

Page 10 -- (basically, we need to make a real world RGB)

Well, a real-world RGB would make a lousy working space for print. The idea is not to capture everything we can see, but to encompass everything the *final output* could produce and nothing more. A working space modeled after human vision would be clumsy indeed.

Dan wrote;

SMPTE-240M is no more
designed to be suitable for print than a Boeing 747 is designed to be anSUV.

That really matters little in the context of this discussion - Adobe
RGB was designed to be a better space for print.

If by "designed for print" you mean designed for something that is not even close to print, then yes, it was designed for print. :-)

That they made
mistakes is really irrelevant at this juncture, as it became popular,
widely used - so it was "designed" to slove a problem, and even if it
happened by accident, it certainly accomplish "something".

I agree that AdobeRGB accomplishes something, and that it is not a bad choice of working space. If you are doing nothing but landscape photography and never need something brighter than reality, sRGB is a better choice of working space. If you do special-effects retouching you might find sRGB to be a little limiting.

In Photoshop 5.0, also, the Photoshop engineers. ignoring the advice of
several experts, decided that it would be a good thing for the program to,
without warning, convert incoming RGB files into sRGB without notifying
the user.

As very few users would have really understood that some 'free range"
RGB might be better off being forced into a 'defined" RGB, well, I
think the decision was a good one,

???

but I will agree that this would be
totally unaceptable today and we are always better off with some
'warning' that something is 'changing' - although I am not really sure
what that warning dialog box might have said, or if it really did
explain what was happening, if many people would have understood it -
and they few that did, would have said "gee, thanks" and clicked over
that "OK" button ( NERD ALERT -- except me. Or maybe Terry Wyse. We
would have busted out our Wyszecki &. Stiles...)

I'm confused again.

Well, it is not really a "paper" as in the idea of a 'whitepaper'
where they are trying to convince anyone to do anything, it is simply
a technical document that describes what it is.

It seems to describe what they want people to think it is, which makes it kind of "sales-y".

I think we all agree
that is was a goofy mistake, based on some flakey asumptions,

All right! Now we're getting somewhere!

but I would use it all day long over sRGB in most print environments.

And many do.

Like me. :-)

J Walton
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: "Laurentiu Todie"
Fri Mar 13, 2009 11:53 am (PDT)

So, is anyone using BruceRGB?

I'd like to, especially if Adobe's "mistakes" were corrected. (I said this without checking again the visual difference between the two RGB spaces)

I would not use sRGB for print, only Web.

Laurentiu
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: "Andrew Webb"
Fri Mar 13, 2009 1:32 pm (PDT)

Lightjet.

<http: //www.cymbolic.com/products/lightjet5000.html>

On Mar 13, 2009, at 9:59 AM, Sabo, Lori arranged some pixels so they looked like this:

Yes, but in making prints on real old-fashioned photographic paper,
the print is made by applying RGB light source, so the input to the
printing device is truly and appropriately RGB.

Andrew Webb

Creative Director
Serious Retouching & Color
303.819.0480
www.seriousretouching.com
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Terence Wyse
Fri Mar 13, 2009 1:32 pm (PDT)

On Mar 12, 2009, at 1:24 PM, Michael Jahn wrote:

Jahn comments -- Actually, there are "RGB" inks - some Rotogravure
printers very much like OpalTone

I don't know of any process that can use RGB inks exclusively as it would be impossible to create yellow. An RGB additive process could of course produce yellow, but not RGB inks. They must be using subtractive primaries (CMY) and supplementing them with RGB inks as they do in some of the newer 10 and 12 ink inkjet printers.

Sutractively yours,
Terry Wyse
______________________________
Terence Wyse, WyseConsul
Color Management Consulting
G7 Certified Expert
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Terence Wyse
Fri Mar 13, 2009 1:32 pm (PDT)

By saying "CMYK", you appear to be narrowly defining CMYK printing as "CMYK offset press printing"? I can tell you that there is plenty of CMYK printing (inkjet primarily) where AdobeRGB is actually too *restrictive* if you want to take advantage of the entire printable gamut of some of these devices. For some CMYK printing processes, one must look to RGB working/editing spaces that are larger than AdobeRGB.

Just thought I'd point that out that offset printing isn't the ONLY CMYK in the world.

:-)

Regards,
Terry Wyse

On Mar 12, 2009, at 10:23 PM, Dan Margulis wrote:

This argument was made most strongly by the late Bruce Fraser. He
thought that Adobe RGB (which he repeatedly referred to as "the
former Yugoslav republic of SMPTE-240M") was too wide-gamut for CMYK
work, and proposed a new space, BruceRGB, as an alternative.

In doing this, he meant, and I agree, that an RGB designed for HDTV
use as modified by typographical errors is not likely to be very
good for print. The paper being referred to is like the chef who
tries to fry eggs, breaks the yolks, and tells the worldt he meant
to scramble them all along.
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Terence Wyse
Fri Mar 13, 2009 1:32 pm (PDT)

Working Space Wars! :-)

BruceRGB had it's day and is still viable but many folks consider the "eciRGBv2" RGB working space to be ideally suited for print work (the offset/flexo kind, not necessarily inkjet or photo printing). Personally, I think it's superior to AdobeRGB and is more "right-sized" for RGB images intended for offset printing. If you want me to go all Techno-Color-Geek-o on why that is the case, I can but I'm a color-lover, not a color-fighter.

:-)

Regards,
Terry Wyse

______________________________
Terence Wyse, WyseConsul
Color Management Consulting
G7 Certified Expert
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: J Walton
Fri Mar 13, 2009 3:53 pm (PDT)

On Fri, Mar 13, 2009 at 12:46 PM, Terence Wyse wrote:

BruceRGB had it's day and is still viable but many folks consider the
"eciRGBv2" RGB working space to be ideally suited for print work (the
offset/flexo kind, not necessarily inkjet or photo printing).
Personally, I think it's superior to AdobeRGB and is more "right-
sized" for RGB images intended for offset printing. If you want me to
go all Techno-Color-Geek-o on why that is the case, I can but I'm a
color-lover, not a color-fighter.

I suppose that makes me both a color-lover and a color-fighter! I fight *because* I love... ;-)

I for one would be grateful for you to lend your expertise to this discussion. If you can go Techno-Color-Geek-o without leaving us all behind or getting us confused, we may just learn something after all!

So, here's a question... "How is eciRGBv2 better than AdobeRGB for offset printing purposes? Why might some prefer it over BruceRGB?" Obviously anything else you want to comment on is open but I'd love to hear those things answered.

I seem to remember seeing something on Lindbloom's site that compared RGB working spaces with some sort of traditional CMYK gamut, but I went back today and didn't see it there. I wonder if I am just "mis-remembering".

J Walton
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Michael Jahn
Fri Mar 13, 2009 3:54 pm (PDT)

Hi TW,

OpalTone's RGB inks are meant to "add" to CMYK, not replace - sorry, should hve been more clear on that. The basic idea AFAIK is to extend the gamut of CMYK printing (kinda sorta like Hexachrome) so gravure packaging printers can avoid down presstime that is required during wash out (between spot color jobs) - the idea being that you take spot color jobs in and 'separate" them into CMYKRGB 7 color jobs and run that (instead of stopping the press and switching spot color inks for job to job to job)

I only mentioned that because Andrew Haley had never thought that RGB inks actually existed or were used - but i think we all agree that they would not be used in projectors. Actually, using inks in projectors would require some sort of rain suit be worn by the audience.

Michael "The only Lab I like is my Labrador" Jahn
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Michael Jahn
Fri Mar 13, 2009 4:36 pm (PDT)

On Fri, Mar 13, 2009 at 12:45 PM, J Walton wrote:

"How is eciRGBv2 better than AdobeRGB for
offset printing purposes? Why might some prefer it over BruceRGB?"

from;

http://www.eci.org/doku.php?id=en:colourstandards: workingcolorspaces

The most relevant improvement is that luminance is now encoded in an
equidistant way – ‘conversion losses’ between data and the human eye
are thus a matter of the past: The gamma of 1.8 has been replaced by
an L* characterization as used in the theoretically ideal CIELAB color
space. This improved encoding efficiency brings with it substantial
advantages in the shadows, as the risk of posterization effects –
especially while retouching – is significantly reduced. Errors caused
by colour space conversions – e.g. banding or reversal – are minimized
as much as currently technically feasible.

<!--snip--!>

sRGB as a color space has serious weaknesses - there are a lot of
colors today’s printing presses as well as other output devices like
photo printers, large format printers and many inkjets (not to speak
of up to date monitors or digital cameras) can produce that cannot be
stored in sRGB. If quality is important, sRGB is not an option.

hope that helps.

Michael Jahn
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: J Walton
Fri Mar 13, 2009 4:36 pm (PDT)

On Fri, Mar 13, 2009 at 1:49 PM, Michael Jahn wrote:

Michael "The only Lab I like is my Labrador" Jahn

Ha! I got a kick out of that one.

While I'm a big fan of LAB as an editing space, since there are things you can do there that cannot be done anywhere else, I gotta respect a dog lover who works his convictions into his middle name.

Sincerely,

J "I love LAB almost as much as my dog Buster" Walton
http://picasaweb.google.com/solomentej/TheBusterFiles#
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Dan Margulis
Fri Mar 13, 2009 4:37 pm (PDT)

Andrew Haley writes (about the use of sRGB as a customary working space)

I would seriously question such an assertion from someone committed to
high-quality work, which presumably includes everyone entering
competitions. I can print well outside the sRGB gamut, and it is one
of the things that differentiates a gallery-quality print from
something you got made at the corner drugstore.

Since I track people's practices by polling each ACT class and most of my Photoshop World sessions, I can comment fairly definitively on this. Adobe RGB is a perfectly reasonable choice, is used by many at the highest levels of sophistication, and I have no quarrel with those who choose it. It is, however, a fact that the more sophisticated the sample in terms of image manipulation ability, the more likely one is to encounter sRGB as the primary workspace.

Over the last two years, in both my ACT classes and at Photoshop World, Adobe RGB users outnumber sRGB users by around 3 to 1. Use of anything else is rare.

The super-advanced courses that I am currently teaching do not consist exclusively of color gods, but there is a high proportion of top predators--people from a variety of disciplines, from many different countries, who have risen to the top of their field by demonstated superiority in technique. I have been in the imaging industry for a very long time and have not yet encountered so much talent in the same room. Getting to share technique with them is an experience both humbling and exhilarating.

This group has in principle less reason to use sRGB, because it prepares images for every conceivable output condition. Well over half has critical output requirements in both CMYK and RGB, and only 20 percent are essentially CMYK-only (compared to nearly 40 percent in standard ACT classes).

The breakdown of the 15 people here with respect to customary RGB workspace is 10 sRGB, five Adobe RGB, others none.

Dan Margulis
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Dan Margulis
Sat Mar 14, 2009 5:35 am (PDT)

Terry writes,

By saying "CMYK", you appear to be narrowly defining CMYK printing as
"CMYK offset press printing"?

The allegation was that Adobe RGB was designed for "print". As it (under an assumed name) was placed in beta versions of Photoshop in 1997, "print" must be interpreted in terms of what it might have meant at that time. "CMYK offset press printing" is IMHO a a reasonable understanding of the meaning of "CMYK", granted a 1997 time frame.

I can tell you that there is plenty of
CMYK printing (inkjet primarily) where AdobeRGB is actually too
*restrictive*

Of course you can, just as you can tell me that the current president of the United States is William Tecumseh Sherman, or that color management was invented by Ottmar Mergenthaler.

Dan Margulis
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Stephen Marsh
Sat Mar 14, 2009 5:35 am (PDT)

Ruth Brown wrote:

...to convert image files to a size/format required for digital
projection competitions. The accepted standard here is to use Adobe
RGB as most photographers use that space and it is considered less
confusing to remain with that.

Ruth, I take it that you are talking of Adobe RGB as being the working/editing space used for processing images. That is of course only one part of the production chain. The next stage would be converting the Adobe RGB files into a space suitable for projection. I believe that this is where a lowest common denominator such as sRGB is being touted, as the projector likely has behaviour that is designed to work well with sRGB or a similar monitor-like space. Is it possible to profile the projector (are the output and viewing conditions repeatable?) and to convert Adobe RGB working files via a *perceptual rendering intent* into this projector output space? sRGB uses a relative colorimetric intent, clipping out of gamut colours and possibly reducing detail in saturated areas.

Or does the projector have an Adobe RGB setting?

What would be nice is to have the ability to use perceptual gamut mapping performed independently by the CMM and not to be limited by relcol transforms in simple matrix profiles. That way going from Adobe RGB or larger to sRGB could retain detail.

As an RGB WS example, PhotoGamut RGB has a perceptual transform, however I would not recommend this space for most work.

Of course one can manaully work around such limitations with channel blends and other advanced moves.

I am not going to wade in on the other points on the pros/cons of Adobe RGB vs. sRGB, or the history/intent of Adobe RGB in this post as I would prefer to explore the practical aspects of your situation first.

www.ledet.com/margulis/ACT_postings/ProfilingandProofing/ACT-sRGBvAdobe_2004.htm

www.ledet.com/margulis/ACT_postings/ProfilingandProofing/ACT-I-Switched.htm

www.ledet.com/margulis/ACT_postings/ProfilingandProofing/ACT-sRGB-vs-Adobe.htm

www.ledet.com/margulis/ACT_postings/ProfilingandProofing/ACT-RGB-Numbers.htm

Sincerely,

Stephen Marsh
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Howard Smith
Sat Mar 14, 2009 10:14 am (PDT)

On 3/13/09 Michael Jahn included the following link in his 6:10 PM post:

http://www.eci.org/doku.php?id=en:colourstandards: workingcolorspaces

I downloaded the ISO Coated V2 (ECI) profile and tested it with a limited number of images, using Convert to Profile to convert from RGB to CMYK.

While I am certainly no expert on CMYK or on offset printing, my impression is that overall color is definitely improved. Even blues are a little brighter and more true than with Sheetfed Coated CMYK. Reds, greens, and yellows are not only brighter but a closer match to the original RGB. Of course this is referring to appearance on my monitor, not to any printed output.

Now I'm curious to hear from those who are experts in the field of CMYK printing. Most of all it would be nice to get opinions concerning the feasibility of using this profile for CMYK printing on coated stock.

Howard Smith
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Howard Smith
Sat Mar 14, 2009 10:14 am (PDT)

How interesting! If I convert an RGB image to eciRGB v2, then to ISO Coated
V2 (ESI), the blues are significantly less good than if I convert to Adobe RGB 1998, then to ISO Coated V2 (ESI). Both give better results than converting to Adobe RGB and then to Sheetfed Coated V2.

Original link in Michael Jahn's post:
http://www.eci.org/doku.php?id=en:colourstandards: workingcolorspaces

Howard Smith
___________________________________________________________________________

Re: Adobe rgb/digital projection
Posted by: Stephen Marsh
Wed Mar 18, 2009 7:31 am (PDT)

Howard Smith wrote:

I downloaded the ISO Coated V2 (ECI) profile and tested it with a
limited number of images, using Convert to Profile to convert from RGB
to CMYK. [snip] Now I'm curious to hear from those who are experts in
the field of CMYK printing. Most of all it would be nice to get
opinions concerning the feasibility of using this profile for CMYK
printing on coated stock.

Howard, do you prefer the preview or the numbers, or both? How do you feel if you assign your regular profile to the image after conversion to ISO (if your regular profile provides an accurate preveiw of expected printing conditions)?

I have more than a few "ISO" profiles loaded (ECI or from other sources). There is quite a lot of variation and some perform better than others, depending on the rendering intent and the original (there are ISO conditions for flatsheet with higher total ink limits than for web offset, numbers are used in the specifications to differentiate the various "ISO" conditions).

Howard, as with any press profile, if the profile reflects the actual print conditions - results will be "accurate". As with any profile, one can still use it as a "generic" conversion if one prefers the final results, even if the final conditions are not as per the profile (such as with the case of Custom CMYK which does not reflect true printing conditions, however it may provide an acceptable final result).

How interesting! If I convert an RGB image to eciRGB v2, then to ISO
Coated V2 (ESI), the blues are significantly less good than if I convert
to Adobe RGB 1998, then to ISO Coated V2 (ESI). Both give better results
than converting to Adobe RGB and then to Sheetfed Coated V2.

What is the RGB before going to ECI RGB? The v2 ECI RGB does not use gamma, it uses L* (lightness as in Lab mode). Perhaps try the same test using ECI RGB v1, which used 1.8 gamma and not L*.

Of course, due to the nature of working space profiles and their gamut and other variables, it may not be best to convert from the original RGB to another RGB before going to CMYK. In theory an "unnecessary" intermediate profile conversion does not help, although there are exceptions which I have demonstrated on this list in the past. Due to the design decision by Adobe to limit ACR to hardwired RGB working spaces - one can't select ECI RGB and if one would prefer to edit in this space, one has to render the raw file to ProPhoto or Adobe RGB, before going to ECI RGB. Other developers of raw converters may allow the rendering of raw data to any installed ICC profile and not limit the choice to only the profiles that the developer considers that users should have access to (some allow direct output to CMYK from the converter, over and above the choice of any RGB).

Regards,

Stephen Marsh