Profile Problem

[bob carnie]

Hi Folks

here is one for the PS Guru's

When working on a file I first assign adobe 1998 , then I convert to profile ie High Cotton White. At the start I am in RGB but for various reasons**please I have my reasons** I switch mode to LAB. In LAB I do my work and then go back to RGB for final adjustments and printing on my Lambda.

I have noticed that I lose my settings with this mode change and basically I go back and assign adobe 1998 and convert to my high cotton white.
what perplexes me is I always get a slight contrast/colour shift when I do this.

Am I missing something very basic here?
Is there a way for the profiles to stay intact with mode changes?

your help is appreciated

[Kirk Gittings]

I see something somewhat similar, losing my softproof settings, through CMYK conversions. I wish I knew the answer.

[Tyler Boley]

Bob, I'm not sure I follow what you would have preferred to occur..

are you saying that when you convert back to RGB (which will always be to the RGB working space in the color settings when using the image> mode method) from LAB you want that conversion to be to the High Cotton White you converted "from"?
If so, rather than doing a mode change to RGB, you need to edit> convert to profile selecting High Cotton White from the list.
Or am I missing the point?
Tyler

[bob carnie]

Tyler
when I apply adobe 1998 then convert to lets say high cotton white. I see an asterk like this rgb16* when I change modes to either cmyk or lab do any # of adjustments then come back to rgb mode for printing the asterk is gone and I believe the profile I originally converted to as well .. which looks like this rgb16

To me the asterik implys that there is an assigned specific profile and is important.

The loss of the asterik with a mode change implys to me that the profile I have selected in my beginning steps seems to get lost with the mode change.

currently I just re edit >assign profile edit convert to profile
this is where I see a slight change which I then make a adjustment to tweak back contrast specifically.

I am wondering how do you keep a profile on the steps of image adjustment when I change modes.
hope this clarify's a bit


bob





Bob, I'm not sure I follow what you would have preferred to occur..

are you saying that when you convert back to RGB (which will always be to the RGB working space in the color settings when using the image> mode method) from LAB you want that conversion to be to the High Cotton White you converted "from"?
If so, rather than doing a mode change to RGB, you need to edit> convert to profile selecting High Cotton White from the list.
Or am I missing the point?
Tyler

[bob carnie]

Kirk
I believe this is a mode change specific issue, which is beyond my current skill sets.
It really is a pain in the ass as I have a order of moves that I believe are good for my workflow, but this issues bugs me.

Bob

I see something somewhat similar, losing my softproof settings, through CMYK conversions. I wish I knew the answer.

[Kirk Gittings]

Your are right...sort of as a profile conversion, but changing from RGB color to CMYK as a Mode change does the same thing. Right? It seems to me a related issue.

[Francisco Disilvestro]

Hello,

I think Tyler had it right, instead of a mode change from LAB to RBG, when you are in LAB mode Edit>Convert Profile to High cotton white.

If you just change mode from LAB to RGB the profile will be what you have in Edit> Color settings as RGB working profile.

Another way (also suggested by Tyler) will be to assign High Cotton White as your working RGB profile in color settings

The asterisk in the color profile is just a warning that you are working whit a different profile than your default working profile.

The same will happen with a mode coversion to CMYK

[Stephen Best]

I am wondering how do you keep a profile on the steps of image adjustment when I change modes.


I'm not sure I follow what you're trying to do, but Photoshop keeps no record of what space you were in before the change to Lab. You'll have to specify the space you want in Convert to Profile. If you instead do a Mode change, it will pick up the default RGB space from your Color Settings. Note that the colours you get won't necessarily be those in the Lab space, nor those displayed on your monitor before the switch. Photoshop does a colorimetric conversion from Lab to RGB which will change the L* for those colours out of gamut in the destination space ... this is maybe what you're seeing.

[bob carnie]

Hi Stephen

Is there a way to make sure my settings follow through different mode changes.???

I work with *as many here* on a daily basis well over 10 different materials which we have made profiles for.
My working method to date is to open an file almost always in RGB, first in edit drop down I assign profile -Adobe 1998 . I then in edit drop down, convert to profile , at this stage I enter the type of paper I will be using,it can be a number of different papers but lets say High Cotton White, which I have profiled.
From here I nuetralize the image and make minor adjustments to the file.

When I am happy with my work I then change modes to LAB to make contrast and sharpening adjustments, I may make some basic colour adjustments , but in LAB I am particularly building my image contrast, density and sharpening.** I have never checked if the High Cotton White profile is still on, which I will play with tommorow.*

After I am happy here I change modes back to RGB and do final adjustments to the file.
When I am back in RGB this is where I notice that I have seemed to lost my profile and basically to date have been going back into the Edit drop down and redoing what I initially did by assigning and then converting back to Adobe 1998 and the paper profile I wanted to be in all along.

When I re convert to the paper specs , this is when I notice the slight change , which is usually a mild contrast change , and this is easily corrected but has always perplexed me.

Fransisco and Tyler.. are you suggesting that when I change modes I should double check my settings in Edit drop down when I get in LAB??
I never thought to do this as I always assumed that in PS once you assigned a profile it will carry through , mode to mode to mode.
When I go from RGB to Lab , I am now guessing that I lose the High Cotton White profile and any adjustments I make in LAB are basically blind adjustments?

I hope I am not confusing everyone here and I am not a total knob here an missing something really simple.


thanks
Bob



I'm not sure I follow what you're trying to do, but Photoshop keeps no record of what space you were in before the change to Lab. You'll have to specify the space you want in Convert to Profile. If you instead do a Mode change, it will pick up the default RGB space from your Color Settings. Note that the colours you get won't necessarily be those in the Lab space, nor those displayed on your monitor before the switch. Photoshop does a colorimetric conversion from Lab to RGB which will change the L* for those colours out of gamut in the destination space ... this is maybe what you're seeing.

[Francisco Disilvestro]

When you are in LAB mode there won´t be a profile assigned, so what I suggest is to directly convert to your desired profile (High Cotton White) instead of change mode from LAB to RGB,

So in your workflow, from Adobe RGB > convert to High Paper White > adjust > change mode to LAB > adjust > convert to High Paper White (instead of change mode to RGB)

This way you don´t have to be worried what your default settings are.

[Stephen Best]

My working method to date is to open an file almost always in RGB, first in edit drop down I assign profile -Adobe 1998 . I then in edit drop down, convert to profile , at this stage I enter the type of paper I will be using,it can be a number of different papers but lets say High Cotton White, which I have profiled.

I'm not sure why you're assigning a space. This should be set by the scanner software or raw conversion. Also, be aware that printer spaces aren't gray-balanced nor linear and while you can edit in them you have to be very careful. It's something I did in the past to nail the black point and counter saturation losses from perceptual transforms but now have a better printer and colour management tools and no longer feel the need. Regardless, the conversion to a printer space is something you want to do last, then print with No Color Management (but not on a Mac with CS4 but this is another can of worms).

What you appear to want though is for Photoshop to retain some notion of the destination RGB space while you're in Lab which it won't do. You can create actions to convert to the space you want (or just set them as the default in Color Settings). Using Lab mode outside of the small gamut world that Margulis inhabits (where the monitor, working and printer spaces are all of comparable size/shape) isn't that practical ... as you're beginning to appreciate.

[Tyler Boley]

Hi Stephen

....
When I am happy with my work I then change modes to LAB to make contrast and sharpening adjustments, I may make some basic colour adjustments , but in LAB I am particularly building my image contrast, density and sharpening.** I have never checked if the High Cotton White profile is still on, which I will play with tommorow.*
...

at this point it's a total non issue, PS has no way of knowing or really caring what space or profile you previously came from, to your LAB part of the workflow. Whatever editing you are doing in LAB at this point is doing what it is doing with no regard to what space the data was previously tagged.


...
After I am happy here I change modes back to RGB and do final adjustments to the file.
When I am back in RGB this is where I notice that I have seemed to lost my profile..

yes, this is because of doing a mode change, which has nothing to do with your profile, only the working RGB space. So to change to your RGB Paper profile, from hence you cameth fortnights past... you need to do edit> convert instead of image> mode> RGB, as others here have mentioned as well.
You'll still be going to RGB, since that is the mode your printer profile is, but the RGB you wanted.
The steps you are taking result in the slight visual shift, because they involve another "assignment", which tells Photoshop to display the data differently.

The only way to have PS "track" (or seam to) these mode changes is to have your paper profile selected as your working space so any mode change goes to it, instead of a working space, as Francisco mentioned, but this would not be recommended for a variety of reasons, one of which is that you have several RGB paper profiles and would constantly have to change the setting to accomidate.
If I do have my head around your dilemma, I really think the solution is to accept the fact that PS, after any profile or mode conversion, retains no memory of where it came from to utilize in future conversion "back" or whatever, and make that one change, at the LAB to RGB change, using edit> convert to profile> your paper profile, rather than a simple mode conversion, and all should be well.
Maybe?
Tyler

[Joanna Carter]

Can I start by asking what is the source of the file?


When working on a file I first assign adobe 1998 , then I convert to profile ie High Cotton White.
Why do you bother to assign a profile only to convert to another? You should only assign a profile when doing something like correcting the colours to match a source for which you have a profile like a scanner.

Also, I would ask why you are using, what looks like, the profile for a paper as your working profile? Yoiu should only be using the paper profile for proofing, by adjusting the Proof Setup on the View menu.


At the start I am in RGB but for various reasons**please I have my reasons** I switch mode to LAB. In LAB I do my work and then go back to RGB for final adjustments and printing on my Lambda.
My working space is ProPhotoRGB; I have just switched to LAB mode and back again. The result is that the profile of the file changed to the LAB profile, then back again to ProPhotoRGB.


I have noticed that I lose my settings with this mode change and basically I go back and assign adobe 1998 and convert to my high cotton white. what perplexes me is I always get a slight contrast/colour shift when I do this.
What is causing the colour shift is more than likely the business of both assigning and then converting, again, to another profile. Basically, you started out by converting to your "paper" profile, which you shouldn't be doing; then you assign back to the AdobeRGB before converting again!!!


Am I missing something very basic here?
Is there a way for the profiles to stay intact with mode changes?
Yes, stop trying to use a "paper" profile whilst working on the file, only use it for proofing via the View|Proof Setup menu.

Assigning a profile does not alter the file in any way, it simply affects the way you see the image on screen. Re-assigning a profile will always cause a visual change, whereas converting will not affect what you se but does change the file.

[bob carnie]

This makes sense, now what happens when I then change mode to RGB?
I am going to work on this today but the last step is where I seem to be losing the profile. Is it a simple fact that I have to live with the convert to High Paper White each time I change modes?

When you are in LAB mode there won´t be a profile assigned, so what I suggest is to directly convert to your desired profile (High Cotton White) instead of change mode from LAB to RGB,

So in your workflow, from Adobe RGB > convert to High Paper White > adjust > change mode to LAB > adjust > convert to High Paper White (instead of change mode to RGB)

This way you don´t have to be worried what your default settings are.

[bob carnie]

I convert to profile for the very simple reason that as I am making adjustments to any file , I want to see the actual final result while doing so. If I apply the profile at the end stage basically all my edits are blind** unless I am missing something here**, the type of original does not matter, phase capture, cannon/nikon capture, scan from various scanners. They all drop the profile during mode change.
I am now thinking that I will have to convert to profile in each mode as some suggest.

I should say that this is not causing me problems with my printing, as I am getting first run dead nut monitor to final print results on all my papers. * currently we have around 15 different stocks we work with on any given day*
If I do not apply the High cotton white at the beginning stage and end stage my results fluctuate and I do not get great results.
I do understand the colour gamut issues between the different modes , but my changes are generally from rgb - lab*much larger space* and back therefore I am not afraid of loss due to gamut restrictions.
If I was going from rgb - cmyk then it would be another issue , as I may be affecting some colours by going into a smaller space.
Creating an action to convert to profile should not be an issue each time I change modes, I was just perplexed why this happened and wondered if anyone else here saw this as a problem and knew a quick answer.

for me it has been convert back to profile at the last stage, see the change and adjust. This usually is a contrast change which Stephen explained.
Most of my Lab work is contrast, density and sharpening work so I have not seen a critical failing or change in colour.
I assume I am seeing a bit of the reverse, contrast density change when this happens.

To me a profile is:

A guide to what the colour gamut and contrast of any paticular paper will exhibit.
It shows me the what will happen to different colours when paper stocks are used.
As well it allows me to work on a file at the beginning stage ** not the end stage, as suggested above** with the confidence that what I am seeing on screen will be as close to accurate as possible.

By applying a profile at the end stage I believe one is working blind to the paticular nuaunces of paper being used while editing. Without seeing actual predictable colour and contrast how in the world does one do any degree of dead nuts colour match.
If someone here can explain how applying a profile at the very last stage and then not re edit and still make a great first test , I am all ears .


Now I may be wrong with my assumptions but my daily workflow tells me that having a profile attached while editing is the way to go. Without it I think I am working blind.

I probably am old school that I want to see a test print before any final prints are made , but to me this is a step that we do on all work and over the last couple of years I have seen enough file - print to appreciate a test print.

Years ago , I worked at a few very large wedding lab. The single most defining area was the VCNA workstation which was used to edit each negative to a good starting balance. Each printer needed to run a calibration and then when balanced one was able to edit and print out a paticular set of numbers that showed the current negative 's devience from normal. By applying this number to a translator and then nulling the enlarger to nuetral for colour and density we were able to get very good first tests to work from.

Thirty years latter I am seeing basically the exact same thing except in my case the profile I have made for a paticular paper and chem's/inks sets now replaces the VCNA machine. I still have to apply this profile while editing the image to expect good reproductions when using various material.

I hope I am not boring the crap out of you all, but all of these questions are related to those who are printing onto various papers and need to have very good first test prints that mimic what is seen on the monitor.

-am I to assume that Yes PS cannot hold onto a profile with Mode changes?
-if it can is it by creating actions to do so when changing modes?
-should I get over it and just keep on doing what I do ?
-will adding a profile as a very last step give me the results I desire?*monitor-print accuracy*
- if so would someone please explain in detail how this would work.

[Francisco Disilvestro]

Profiles are related to a color space, so when you convert from LAB to High Paper White the result will be in RGB. No need to change mode to RGB, it already is in RGB.

Change mode to RBG is the same as Convert to defalut RGB working profile.

[Joanna Carter]

I convert to profile for the very simple reason that as I am making adjustments to any file , I want to see the actual final result while doing so. If I apply the profile at the end stage basically all my edits are blind** unless I am missing something here**
Yes, you are missing something here. You are not meant to assign or convert your image to the print profile. You should be working in a wide gamut profile like ProPhotoRGB or, at least, AdobeRGB. Use the Proofing Setup to allow you to see how the final print profile will affect the image.


the type of original does not matter, phase capture, cannon/nikon capture, scan from various scanners. They all drop the profile during mode change.
I am now thinking that I will have to convert to profile in each mode as some suggest.
Your workflow for a camera should be to convert from the camera profile to something like ProPhotoRGB and work on the image, converting to and from LAB mode as necessary, but using the Proofing Setup to apply the printing profile.

Scanning is similar but, usually, you assign the scanner profile after opening the image in Photoshop but prior to converting to ProPhotoRGB. Then you must not convert to your printer profile, simply set the Proofing View.


If I do not apply the High cotton white at the beginning stage and end stage my results fluctuate and I do not get great results.
You are getting fluctuating results on printing because you need to proof and print using a separate profile for each paper.


I do understand the colour gamut issues between the different modes , but my changes are generally from rgb - lab*much larger space* and back therefore I am not afraid of loss due to gamut restrictions.
Which is why I suggest using PRoPhotoRGB as your default working profile, it is also a much wider gamut.


for me it has been convert back to profile at the last stage, see the change and adjust. This usually is a contrast change which Stephen explained.
It is not converting to a profile that causes the contrast change, it is assigning a profile that causes the visual change but does not change the file.


To me a profile is:

A guide to what the colour gamut and contrast of any paticular paper will exhibit.
It shows me the what will happen to different colours when paper stocks are used.
As well it allows me to work on a file at the beginning stage ** not the end stage, as suggested above** with the confidence that what I am seeing on screen will be as close to accurate as possible.

By applying a profile at the end stage I believe one is working blind to the paticular nuaunces of paper being used while editing. Without seeing actual predictable colour and contrast how in the world does one do any degree of dead nuts colour match.
If someone here can explain how applying a profile at the very last stage and then not re edit and still make a great first test , I am all ears .
Then you need to listen to what is being said ;)

Definitive workflow; try this at least once before you do anything else and let us know what happens:

1. Set your default working colour space to ProPhotoRGB

2. Use Image RAW to open an image from your camera.

3. If the import didn't convert to ProPhotoRGB, Convert to ProPhotoRGB.

4. On the View menu, choose Proof Setup|Custom… and choose your paper profile.

5. On the View menu select the Proof Colors menu item ticked

6. Make any edits that you don't want to do in LAB mode.

7. Convert to LAB mode.

8. Make any edits in LAB mode.

9. Convert back to RGB mode (the profile should, once more, be ProPhotoRGB.

10. When you are satisfied with your image, Print it ensuring that the Print dialog shows the Color Handling as "Photoshop Manages Colors" and select your printer profile.

[bob carnie]

I just tried that and you are right, went right back into rgb


Profiles are related to a color space, so when you convert from LAB to High Paper White the result will be in RGB. No need to change mode to RGB, it already is in RGB.

Change mode to RBG is the same as Convert to defalut RGB working profile.

[bob carnie]

I will give this a go , generally I am working from supplied files or scanned negatives so I will do this with a scanned negative.
I will report back

Yes, you are missing something here. You are not meant to assign or convert your image to the print profile. You should be working in a wide gamut profile like ProPhotoRGB or, at least, AdobeRGB. Use the Proofing Setup to allow you to see how the final print profile will affect the image.


Your workflow for a camera should be to convert from the camera profile to something like ProPhotoRGB and work on the image, converting to and from LAB mode as necessary, but using the Proofing Setup to apply the printing profile.

Scanning is similar but, usually, you assign the scanner profile after opening the image in Photoshop but prior to converting to ProPhotoRGB. Then you must not convert to your printer profile, simply set the Proofing View.


You are getting fluctuating results on printing because you need to proof and print using a separate profile for each paper.


Which is why I suggest using PRoPhotoRGB as your default working profile, it is also a much wider gamut.


It is not converting to a profile that causes the contrast change, it is assigning a profile that causes the visual change but does not change the file.


Then you need to listen to what is being said ;)

Definitive workflow; try this at least once before you do anything else and let us know what happens:

1. Set your default working colour space to ProPhotoRGB

2. Use Image RAW to open an image from your camera.

3. If the import didn't convert to ProPhotoRGB, Convert to ProPhotoRGB.

4. On the View menu, choose Proof Setup|Custom… and choose your paper profile.

5. On the View menu select the Proof Colors menu item ticked

6. Make any edits that you don't want to do in LAB mode.

7. Convert to LAB mode.

8. Make any edits in LAB mode.

9. Convert back to RGB mode (the profile should, once more, be ProPhotoRGB.

10. When you are satisfied with your image, Print it ensuring that the Print dialog shows the Color Handling as "Photoshop Manages Colors" and select your printer profile.

[Joanna Carter]

I will give this a go , generally I am working from supplied files or scanned negatives so I will do this with a scanned negative.
I will report back
In that case, instead of stages 2 & 3, do this instead :

2. As long as your default working profile is set to ProPhotoRGB, simply scan your neg into Photoshop and check that the profile of the image is set to ProPhotoRGB. You can't use scanner profiles for colour negs, only transparencies, so don't try to do anything other than convert to ProPhotoRGB.

3. However, if you are scanning a transparency, you should assign the relevant scanner profile to the image before converting to the ProPhotoRGB profile.

Don't forget to setup the Proofing on the View menu :D

[Francisco Disilvestro]

I also prefer the workflow suggested by Joanna,
Just remember after step 9 (convert back to RGB) that you need to repeat step 5, select Proof Colors menu item ticked in the view menu, if you want to see the result in the paper profile.

[Joanna Carter]

Just remember after step 9 (convert back to RGB) that you need to repeat step 5, select Proof Colors menu item ticked in the view menu, if you want to see the result in the paper profile.
Thanks Francisco, I forgot about that bit ;)

[bob carnie]

Quick question for Stephen

Is there a way to profile papers in LAB?
If so how do they stick in both modes?
Is this why you recommend Lobster ?so to keep in RGB.

My original question refers to a very small contrast shift when back in RGB which to date I do a slight curve tweak to correct.
When in LAB I am mainly concerned with the L channel for most images. I am not using the A or B other than a few colour corrections.
I think this is what I am noticing with mode changes.

[Joanna Carter]

Is there a way to profile papers in LAB?
If so how do they stick in both modes?
Is this why you recommend Lobster ?so to keep in RGB.
No matter which mode your file is in, you should neither convert to or assign the paper profile to the image whilst editing. The correct way to see the anticipated results is to use the Proofing Setup which doesn't affect the file but does change the on-screen appearance of the image to match the paper.

But, as Francisco reminded me, after changing modes to or from LAB, you need to reset the View|Proof Colors menu item to be ticked so that you are seeing the effect of the print profile.

[Ellis Vener]

Are you assigning your RGB color space or converting to your RGB or LAB space? What color space are you starting from?

[Tyler Boley]

Bob, to be clear, I have no idea how you arrived at your workflow nor any desire to pick it apart, I was only addressing the one conversion step that did not meet your expectation.
As others are pointing out, the bigger picture is an entirely different and bigger can of worms....
Good luck,
Tyler

[bob carnie]

I am starting in Adobe 1998 colour space , starting in RGB mode, then I am changing modes to LAB mode to make adjustments, then back to RGB for final tweaks.

Are you assigning your RGB color space or converting to your RGB or LAB space? What color space are you starting from?

[bob carnie]

Tyler
I am not taking any offence to any reply's . my original question was basically based on an observation of mine that perplexed me regarding what happens after one moves from one mode to the other.
I am going to follow through with Joanna's and Fransciso's suggestions that may help clarify my original observations.
I do think the answer is basically that LAB will not retain a profile assigned.. simple as that.
Stephen has mentioned Lobster to me before and I have never followed through with his advice other than downloading the manual.
I am thinking by staying in RGB , some of what I see would not happen, but I love the flexibility of Lab for control of contrast and sharpening. *I do not like building contrast or sharpness on the colour channels but rather work on the L or K channel. I feel less artifacting and halo's are prevented this way.
We all come at this from different perspectives and learning curves, I am hoping that by asking the question maybe a light will click on that I may have missed.
My workflow may be different from others , but I am sure this could be said about anyone of us.
Just a side note,, ink printing is a very small part of my duties, I am mainly the file editing operator, then sending files to secondary computers to load up the printers. Therefore the actual mechanics of an inkjet machine is not a strong skill sets of mine, I do though operate the Lambda which is a 8bit RGB printing device.
I have thought about putting a 24inch ink printer in my room so I can proof , before going to the larger machine, but not there yet.

What I am seeing is a huge circle of steps that when connected truly wonderful prints are being made.
Working with multiple printers and medias just increase the amout of steps that need to be connected and fully understood.

My next big can of whoop ass is going to make an ICG scanner work in all of this workflow.


QUOTE=Tyler Boley;506331]Bob, to be clear, I have no idea how you arrived at your workflow nor any desire to pick it apart, I was only addressing the one conversion step that did not meet your expectation.
As others are pointing out, the bigger picture is an entirely different and bigger can of worms....
Good luck,
Tyler[/QUOTE]

[Francisco Disilvestro]

It is my understanding that LAB mode is device independent (the purpose of profiles is to make them device dependent) so you cannot have a specific profile in LAB.

If your output device is RGB then you should have a profile for it in the RGB space

According to some authors*, if you only edit the L channel, you should be able to get almost similar results with a curves adjustment layer with blending mode set to Luminosity

*Chavez, Conrad & Fraser

[Joanna Carter]

It is my understanding that LAB mode is device independent (the purpose of profiles is to make them device dependent) so you cannot have a specific profile in LAB.
I think Bob's problem was that he didn't understand what was happening when he changed from RGB to LAB and back again. As you correctly said, after changing mode, you need to re-apply the printing profile in the Proof Colors item of the View menu.

As long as Bob doesn't change the working space profile, switching from RGB to LAB and back again should leave him back in the same profile he had before changing to LAB mode.

Bob, changing to LAB mode doesn't "lose" a profile, it carefully converts from the ProPhotoRGB profile to the LAB profile; just as changing back from LAB to RGB will carefully convert from the LAB profile back to the ProPhotoRGB profile.

Just remember to switch the printing profile back on in the Proofing menu :D

As to you doing the editing and then passing on the file to others for printing, it is very important that you don't convert the file to any print profile; that is the job of the printing operator when they choose which paper to print to.

[bob carnie]

I think an old dog just learned a new trick.
going to the view menu and re-applying the profile after each mode change,seems to be the ticket, when I go back to RGB I needed to go to view menu and reapply the profile and my file looked the same as what it did in LAB .



I think Bob's problem was that he didn't understand what was happening when he changed from RGB to LAB and back again. As you correctly said, after changing mode, you need to re-apply the printing profile in the Proof Colors item of the View menu.

As long as Bob doesn't change the working space profile, switching from RGB to LAB and back again should leave him back in the same profile he had before changing to LAB mode.

Bob, changing to LAB mode doesn't "lose" a profile, it carefully converts from the ProPhotoRGB profile to the LAB profile; just as changing back from LAB to RGB will carefully convert from the LAB profile back to the ProPhotoRGB profile.

Just remember to switch the printing profile back on in the Proofing menu :D

As to you doing the editing and then passing on the file to others for printing, it is very important that you don't convert the file to any print profile; that is the job of the printing operator when they choose which paper to print to.

[Joanna Carter]

I think an old dog just learned a new trick.
going to the view menu and re-applying the profile after each mode change,seems to be the ticket, when I go back to RGB I needed to go to view menu and reapply the profile and my file looked the same as what it did in LAB .
Yaaayy!!! So, now you've got that sorted, how about not bothering to convert to the printing profile until the actual printing? :) :cool:

[bob carnie]

Well Now
please explain that one for me and I might consider.
to my thinking I want to see as I edit on my monitor what I will get on paper, I do not want to edit and find I have to fudge contrast or colour.
putting on the paper profile at the beginning seems to work fine for me if everyone else is doing it at the end I am willing to eat crow.. its not easy being humble.

Yaaayy!!! So, now you've got that sorted, how about not bothering to convert to the printing profile until the actual printing? :) :cool:

[Joanna Carter]

Well Now
please explain that one for me and I might consider.
to my thinking I want to see as I edit on my monitor what I will get on paper, I do not want to edit and find I have to fudge contrast or colour.
putting on the paper profile at the beginning seems to work fine for me if everyone else is doing it at the end I am willing to eat crow.. its not easy being humble.
If you are now using the printing profile in the Proofing view, you should not also convert to the printing profile, otherwise you will get the same offset from the profile twice, thus giving you a false idea of how the final print will look.

Take a look at any Photoshop teaching and you will rarely see the printing profile applied before or during the editing, as this then limits the file to only ever being printed on the paper that that profile is created for. The problem with editing in a printing profile is that it severely restricts the gamut of the file, thus making any other version with a wide gamut impossible without interpolating the missing colours. It also makes editing more difficult due to the same restrictions.

Please believe me, I am not out to humble you, simply to help you produce the best quality work you can ;)

[bob carnie]

good call on the double profile as I probably would have done that.

The reason that I apply the paper profile at the beginning is exactly the issue you bring up.
All my work is intended for a single use paper and if it is intended for another purpose then I rework the file from scratch.
When work reaches my desk it is usually dustbusted and kept as a master original for any future printing.
I never consider an image for anything else than the paper it is slated for.
My clients will come in with a test image and we will proof out on a variety of papers and a decision is made for style and paper surface.
Then the real job starts and I am only concerned with how the image will look on the paper we selected , therefore I would like its presence as I edit.
I am not concerned about how it will look on any other paper than the one going up on the walls.
I do understand that photoshop teachers tell us to apply at the end for multiple usages but I am not too concerned in that matter as I generally know where an image is going .

How about this one, all photoshop teachers will tell you to keep a folder for each adjustment so that you can go back and make minor adjustments.
Layer after layer after group wow the file looks very impressive.
I just flatten each layer as I go and am pretty confident with my moves. this really goes against the rules but, I guess its kind of like keeping notes in an analogue darkroom. I never keep notes just move forward and I do not revisit test strips or spend more than 10 seconds analyzing an image.

Where I am not cavalier is the calibration stage, nothing goes through our lambda or ink machines with a proper 21step grey scale balance, as well we are constantly keeping our profiles updated as we have found shifts in chemistry do happen and keeping this in balance is very critical.

So I still will need to have a reason why it would be better to apply a profile at the end rather than the beginning.http://www.largeformatphotography.info/forum/images/icons/icon12.gif





I must be crazy.



If you are now using the printing profile in the Proofing view, you should not also convert to the printing profile, otherwise you will get the same offset from the profile twice, thus giving you a false idea of how the final print will look.

Take a look at any Photoshop teaching and you will rarely see the printing profile applied before or during the editing, as this then limits the file to only ever being printed on the paper that that profile is created for. The problem with editing in a printing profile is that it severely restricts the gamut of the file, thus making any other version with a wide gamut impossible without interpolating the missing colours. It also makes editing more difficult due to the same restrictions.

Please believe me, I am not out to humble you, simply to help you produce the best quality work you can ;)

[Stephen Best]

Well Now
please explain that one for me and I might consider.
to my thinking I want to see as I edit on my monitor what I will get on paper, I do not want to edit and find I have to fudge contrast or colour.

You can soft-proof from Lab just the same as RGB. But be aware that you'll get different results by printing from Lab than by converting back to an RGB working space then printing as the mapping out of Lab is dependent on the space you convert to. How it renders the colours on screen in Lab mode will be different again. All this is something Margulis glosses over.

Rather than dissect your workflow I think you have to just aim for the best image on screen (irrespective of output), then make whatever changes are necessary for printing as a final step. With today's printers and profiling software there may be minimal change to what you see on the screen (assuming your screen is realistically calibrated). I'll typically build a number of profiles for each paper with different media settings, gamut mapping options etc and use soft-proofing to see which will give the least change on output with perceptual rendering. I'm finding that with my Epson 7900 I'm often getting away with Relcol. If you're continually having to tweak the image to get it to print with a good match, look at profiling solutions with better perceptual tables.

Note that if you're working with Lab you're going to end up with wild colours (even with just lightness changes) which will be harder to manage. Keep in mind that state of the art printers don't cover sRGB in places. Lab just isn't a good tool for finessing images.

[Joanna Carter]

good call on the double profile as I probably would have done that.

The reason that I apply the paper profile at the beginning is exactly the issue you bring up.
All my work is intended for a single use paper and if it is intended for another purpose then I rework the file from scratch.
Then you are obviously happy doing more work than necessary. Have you never thought that, if you keep the file in a state that can be reprinted to a different paper without further work, you can still charge the same rate and get on with another job quicker?


Then the real job starts and I am only concerned with how the image will look on the paper we selected , therefore I would like its presence as I edit.
But that is what the Proofing is for; so you can see what it will look like without having to affect the file.


How about this one, all photoshop teachers will tell you to keep a folder for each adjustment so that you can go back and make minor adjustments.
Layer after layer after group wow the file looks very impressive.
I just flatten each layer as I go and am pretty confident with my moves. this really goes against the rules but, I guess its kind of like keeping notes in an analogue darkroom. I never keep notes just move forward and I do not revisit test strips or spend more than 10 seconds analyzing an image.
I just love the way I can go back to a file, alter or remove layers, and improve the print without having to start from scratch. But then my images are for sale in galleries that are more concerned with artistic printing than just a quick print job :cool:


So I still will need to have a reason why it would be better to apply a profile at the end rather than the beginning.
Of course, you are entitled to work however you want, but that doesn't stop us from arguing with you :D :rolleyes:


I must be crazy.
I'm sure I couldn't comment on that statement :p

[Peter De Smidt]

I agree with Joanna. Printer spaces are not meant to be good editing spaces. They are not necessarily very uniform. Converting to a specific profile, especially one that's small and irregular, can cause a lose of tonal information. For example, suppose that there are two colors that differ in Adobe RGB. They might nonetheless get lumped together in the printer color space. Once that information is lost, no global color correction will get it back. If, on the other hand, you do your editing in Adobe 98 using soft proofing, perhaps with a Dan-the-Man move to LAB and back, you'll minimize tonal distortion and loss, and you'll have more control over how the colors get converted by using rendering intents.

[rdenney]

One of the things I learned early on is the difference between correction and targeting. As loathe as I am to offer my meager skills and experience to the posters in this thread, I think that distinction is relevant here.

Bob seems to have a single focus--taking a scanned or provided file and making it look good on one paper/printer combination. He is trying to do that in one step.

The problem is when the paper/printer combination changes. Then, the file has stuff in it that will have to be undone.

My process is to make corrections to create a file in a wide-gamut color space (I use ProPhotoRGB, but Adobe is pretty good, too) and work that file until looks the way I want it to look on my calibrated monitor.

Then, I turn on Proof Colors, with the proofing profile set to match the paper/printer combination.

I then (on different adjustment layers or in a copy of the file) adjust the image as necessary to look the way I want it to look on that output device. That may include resampling, and it will almost always include an additional sharpening process at the output resolution. That is the targeting.

Since many of my images might end up on the web, or printed by a lab in addition to my own prints, I can individually target my basic corrected image to each of those output devices, without having to repeat my basic corrections. In Bob's case, where his shop uses both a Lambda and inkjets, having a corrected file that could be targeted for each should save time.

That distinction also makes it easier to remember not to apply targeting profiles to correction processes, which is Joanna's concern. The printer profile for a particular paper is something the software uses to make the image look like it will look when printed. It is a narrow color space, because the prints have narrow color spaces. Editing in that color space will make it harder to restore zip to the file for viewing, say, on the web.

For me, I make corrections in a wide-gamut color space, and then target for narrower-gamut color spaces. This is especially true if the output space is 8-bit versus 16-bit.

I have also tried converting to LAB color for doing sharpening on the luminosity layer instead of in RGB, but it has always annoyed me that I have to flatten my image before doing so. I'm not as confident as Bob, heh, heh. In any case, that seems a different issue from correction vs. targeting--I've done LAB corrections in both processes.

I used to try to adjust images directly from the scanner to the printer profile, though I still used the printer proofing feature. My work flow got a lot more efficient when I started separating correction and targeting processes, and I had a lot fewer instances of trying to remember what moves I had made when trying to print to a different device. I hate getting just what I want and then waking up in a new world and having to do it all over again. Going from one device profile to another is a lot harder than going from a corrected file in a wide-gamut space to a new device. And there always seem to be new devices.

Rick "who has had to throw away a lot of scan files made early on" Denney

[bob carnie]

Joanna
In my experience once a paper has been chosen the client and I usually stick with it and it may be years before a different paper would be used.
You can go to the Royal Ontario Museum and click on .Vanitey Fair Exhibit > Nigel Dickson portraits. Just finished this 35 piece quick print job, on fuji - flex . Without our custom profile present at editing stages I am afraid I would have not been able to complete the job to his or my satisification.
Fuji does not give a profile for this and I can say that it has some colour nuances that need to be seen while working on it. By creating our own profile and applying it at the begining I am still wondering why this would be bad practice.

I actually love the practice of working quickly on many files , over and over, and love making mistakes, I have found that this is the very thing that has kept me in business all these years.

We never practice on clients work, but I can assure you in the darkroom and on PS we are willing to try every trick in the book to please our clients.

Making a final tweak at the lambda workstation or on PS should be easy and I basically see absolutely no need to keep a layer history. I guess if it's important to some then why not, but for me just a waste of keystrokes.

I do see the need if I was working with multiple images, type or photomerges, but that is not what I do. My working partner is a professional photographer and his skill sets with Creative Suite completely blows away my knowledge, He does keep layers , upon layers, as there is always changes that need to be attended to . I never, never, never want to work on montage images, I did enough by hand in the 80's to satisfy me for life.
I work strictly on single files slated for gallery shows ,simple files where my skill set of dodge and burn come into play. To date I have done enough shows to feel confidence as I edit and any changes are minor and easily accomplished with a minor adjustment. Therefore my lack of attention to layer stacking.
With a good balanced monitor, good profile and a calibrated printing device 95% of the work is done on screen.
As I said at the beginning of this thread, the real proof for me is that first test print which today is a pretty dead nuts match to my monitor.

When you say affect the file by applying a profile are you implying that adding the profile hurts the file, or are you meaning that it just makes any adjustments I am doing stick with the file and are of no use for other papers?
If so,I do want to make the file specific to the paper. I do not want it to be used for other purposes.
If we can agree that adding a profile at the beginning does not hurt the file then yes I am happy redoing the file if it is going to a different stock and yes I would recharge the client.
I can assure you that operating a art print' framing lab today will make you crazy. Keeping up with all the small details is a never ending learning curve. Its these little details that you learn like view >profile that make me happy. and crazy
regards
Bob






Then you are obviously happy doing more work than necessary. Have you never thought that, if you keep the file in a state that can be reprinted to a different paper without further work, you can still charge the same rate and get on with another job quicker?


But that is what the Proofing is for; so you can see what it will look like without having to affect the file.


I just love the way I can go back to a file, alter or remove layers, and improve the print without having to start from scratch. But then my images are for sale in galleries that are more concerned with artistic printing than just a quick print job :cool:


Of course, you are entitled to work however you want, but that doesn't stop us from arguing with you :D :rolleyes:


I'm sure I couldn't comment on that statement :p

[Peter De Smidt]

If we can agree that adding a profile at the beginning does not hurt the file then yes I am happy redoing the file if it is going to a different stock and yes I would recharge the client.

Bob

As I explained above, we don't agree on that. Is the loss of tonal information caused by editing the file in a printer color space big enough to be a problem? That's a subjective call.

[Joanna Carter]

As I explained above, we don't agree on that. Is the loss of tonal information caused by editing the file in a printer color space big enough to be a problem? That's a subjective call.
And I would also agree with you Peter.

[percepts]

When you say affect the file by applying a profile are you implying that adding the profile hurts the file, or are you meaning that it just makes any adjustments I am doing stick with the file and are of no use for other papers?
If so,I do want to make the file specific to the paper. I do not want it to be used for other purposes.


in 8 bit mode the maximum value of any colour channel is 255 or in 16bit mode 32768. But different profiles can't utilise all of that value. Doing a convert of a file physically alters the pixel values downwards in the file to fit within the limits set by the profile.
When converting to a profile which has wider limits it does not maximise the pixel values to fit depending on the conversion options you set.

So converting to a printer profile will physically reduce pixel values from an adobe rgb profile to an 8bit profile.

But you are doing other stuff which is not so good.

Firstly you assign Adobe RGB. You have not told us what profile the acquisition device is using. If outputs in sRGb then you should convert to Adobe RGB as its a bigger space and zero or negligible clipping would occur. Next you convert to print profile which will significantly clip image file data. Then you goto to LaB which has a potentially bigger space than the print profile. Your alterations may put the file space outside the range of your print profile. So why wouldn't you expect there to be a change when you come out of Lab. Whilst your workflow may give you images that are highly acceptable to you, that doesn't mean they are as good as they could be.
So you would be better to work in Adobe RGB, forget about Lab, and then proof with paper profile.
But, and this is critcal, when you created or acquired the print profile it should have done working from a file which was created and fully 100% Adobe RGB otherwise the profile would have been created to optimise a lesser colour space. You haven't told us about that.
If the profile is/was created from an Adobe RGB file, then when you goto print proof using the print profile, there should be no difference from working in Adobe RGB. And that way you absolutely minimise any clipping (data loss) from your original file acquisition. Therefore you maximise the potential of the image.

And everything must be done in 16bit until final conversion to printfile profile for output. Otherwise along the way you clip down from what Adobe RGB is capable of.

p.s. moving from 16 bit Adobe RGB to a print file profile may show some differences because it will clip. But that is what print proofing is for. A chance to finally tweak the the file. If you want you could work in 8bit throughout the process but that kind of makes Adobe RGB pointless as you can't fit adobe rgb into 8bit so you always get a false interpretation on screen.

[bob carnie]

Percepts
All scans are in adobe 1998, all profiles are in adobe 1998.
All work is in 16bit till the very end, for inkjet profile - 16 bit, but for lambda the profile is in 8bit * which would be a hindrence if moves are out of gamut*? Never thought about this aspect and really worth listening too.
Most moves in LAB is to adjust contrast and sharpening.. very small colour adjustments.
The change I am seeing I think Joanna hit on was that when in LAB I was losing the profile , and I am about to test this over the weekend .

My question to you is about using RGB only, how are you adjusting contrast and density?
also how are you sharpening your image?
I have did countless tests and outputed these tests , which show that sharpening and contrast in LAB totally blows away working only in RGB.
but once again I am really open to trying different methods that I have not tried.
For me the proof is on the print.
Curve changes and sharpening on the three colour channels seems to be problematic, but to my eyes at least doing this in the L channel of LAB have very strong advantages.




in 8 bit mode the maximum value of any colour channel is 255 or in 16bit mode 32768. But different profiles can't utilise all of that value. Doing a convert of a file physically alters the pixel values downwards in the file to fit within the limits set by the profile.
When converting to a profile which has wider limits it does not maximise the pixel values to fit depending on the conversion options you set.

So converting to a printer profile will physically reduce pixel values from an adobe rgb profile to an 8bit profile.

But you are doing other stuff which is not so good.

Firstly you assign Adobe RGB. You have not told us what profile the acquisition device is using. If outputs in sRGb then you should convert to Adobe RGB as its a bigger space and zero or negligible clipping would occur. Next you convert to print profile which will significantly clip image file data. Then you go to to LAB which has a potentially bigger space than the print profile. Your alterations may put the file space outside the range of your print profile. So why wouldn't you expect there to be a change when you come out of Lab. Whilst your workflow may give you images that are highly acceptable to you, that doesn't mean they are as good as they could be.
So you would be better to work in Adobe RGB, forget about Lab, and then proof with paper profile.
But, and this is critcal, when you created or acquired the print profile it should have done working from a file which was created and fully 100% Adobe RGB otherwise the profile would have been created to optimise a lesser colour space. You haven't told us about that.
If the profile is/was created from an Adobe RGB file, then when you goto print proof using the print profile, there should be no difference from working in Adobe RGB. And that way you absolutely minimise any clipping (data loss) from your original file acquisition. Therefore you maximise the potential of the image.

And everything must be done in 16bit until final conversion to printfile profile for output. Otherwise along the way you clip down from what Adobe RGB is capable of.

[bob carnie]

Joanna and Peter
could you expand on your reasonings that this is destructive.

And I would also agree with you Peter.

[Peter De Smidt]

Joanna, Percepts and I have already explained why cramming an Adobe 98 files into a small and irregular printer space for editing will result in a loss of tonal information. I don't know what more we can add.

I do have a question, though. You say your scanned file is in Adobe 98, but you have to tag the image when importing into photoshop. Why is that? The standard color-managed scanning workflow is to create a scanner icc profile for each film type that you'll be scanning. The file should be tagged with the scanner profile by the scanning software. If it isn't, when importing into Photshop, you assign the scanner profile to the file and then convert the file to an editing space, such as Adobe 98.

Regarding sharpening, sure, using Lab is fine, but is it really all that different from unsharp masking (or smart sharpen) in RGB, where after sharpening one immediately goes to edit>fade sharpening>luminosity mode?

Regarding image contrast, here's what Dan Margulis currently recommends:

1. Duplicate Layer.
2. Use Curves or channel blending to get better image contrast.
3. Change layer’s blending mode to luminosity.
4. Use Shadow/Highlight if criteria are met.

Notice that he's not using Lab to adjust contrast.

[percepts]

My question to you is about using RGB only, how are you adjusting contrast and density?
also how are you sharpening your image?
I have did countless tests and outputed these tests , which show that sharpening and contrast in LAB totally blows away working only in RGB.
but once again I am really open to trying different methods that I have not tried.
For me the proof is on the print.


If you are working in Adobe 1998 and goto lab and then back to adobe 1998, then you will likely not see a change or it will be so subtle you wouldn't notice it. So yes Lab is OK. But that isn't what you have described. You have already converted to print profile (8bit) before going to Lab and increasing contrast IS a colour change by another name and that can and very likely will put your file out of gamut when going back to print profile.

[Joanna Carter]

Regarding image contrast, here's what Dan Margulis currently recommends:

1. Duplicate Layer.
2. Use Curves or channel blending to get better image contrast.
3. Change layer’s blending mode to luminosity.
4. Use Shadow/Highlight if criteria are met.

Notice that he's not using Lab to adjust contrast.
I simply use a Curve adjustment layer rather than playing with copying an image layer. that means I can go back to the layer and adjust if other layers don't work with the curve I have applied.

What I will also do is to create a Curve adjustment layer, set it's blending mode to luminosity and then adjust the curve; so it only affects the luminosity, not the colours.

[Peter De Smidt]

As long as you don't want to do channel blending, an adjustment layer set to luminosity is fine.

I recently started channel blending, though, and it a very powerful technique to increase detail and contrast, and it isn't very hard.

If anyone is a member of Kelbytraining.com, check out Dan Margulis's 3 part Picture Postcard Workflow. It's not just about happy bright colors, i.e. picture postcard type pictures, although it can certainly give you those if desired.

The workflow is:
1. Get rid of color casts in RGB.
2. Optimize image contrast in RGB.
3. Pump up colors, if desired, in Lab.

[bob carnie]

Peter
Patience, what is the use of a paper profile at end other than showing one how files will change once exposure hits paper.
I really do not see how having an embeded profile of the stock you are about to use being a detriment while one is editing .

One thing you can add is your definition of a profile, and how it works within your workflow.
Are you making one standard file then applying profiles at the end and seeing dead nuts reproductions on your first test on different stocks from the same file?
When changing stocks how you make your file adjustments to compansate for all the different surfaces or is your attachment of a profile at the end doing this for you.?
Do you change stocks daily? Do you make your own profiles for the stocks you are using?
This thread has got me wondering , and curious minds need to be satisfied.


Re Dan Margulis
I am definately not qualified to speak for him, but

I believe what you are referring regarding image contrast is what he refers to as image weight which he does recommend channel blending and curve adjustment in RGb , but I assure you he does move to LAB to then set the L channel for contrast and density. In fact as of Jan he recommends Shadow Highight in LAB on the l channel.
He uses the L channel or the K channel for density/contrast and sharpening.
I have spent enough time with him to know this is part of his workflow.If he has changed his workflow it is news to me as of Jan this year.

once again what I gleaned from him and not his words

RGB curves nuetralize image, RGB channels to set image weight, LAB for highlight Shadow if appropriate, LAB for density contrast in the L channel, Dan does a lot of colour work in the AB channels which I do not.. then back to RGB for minor colour tweaks if he is using a RGB printer device. He goes back to CMYK for most of his teachings and there he uses the K channel for density/contrast and sharpening
I should say that he has been my main PS teacher over the last few years , I have spent 4 sessions with him and I really like his style and methods. I have on numerous times tried an all RGB workflow and after a few thousand images seem to like his methods.

But I can assure you that I am open to other ideas of methods. This whole thread started with a observation I had regarding mode changes, it seems to be resolved, now it seems I need to reinvent my workflow . I am willing to do so but would like some detail's to digest and try.




Joanna, Percepts and I have already explained why cramming an Adobe 98 files into a small and irregular printer space for editing will result in a loss of tonal information. I don't know what more we can add.

I do have a question, though. You say your scanned file is in Adobe 98, but you have to tag the image when importing into photoshop. Why is that? The standard color-managed scanning workflow is to create a scanner icc profile for each film type that you'll be scanning. The file should be tagged with the scanner profile by the scanning software. If it isn't, when importing into Photshop, you assign the scanner profile to the file and then convert the file to an editing space, such as Adobe 98.

Regarding sharpening, sure, using Lab is fine, but is it really all that different from unsharp masking (or smart sharpen) in RGB, where after sharpening one immediately goes to edit>fade sharpening>luminosity mode?

Regarding image contrast, here's what Dan Margulis currently recommends:

1. Duplicate Layer.
2. Use Curves or channel blending to get better image contrast.
3. Change layer’s blending mode to luminosity.
4. Use Shadow/Highlight if criteria are met.

Notice that he's not using Lab to adjust contrast.

[Peter De Smidt]

Go to http://www.kelbytraining.com/instructors/dan-margulis.html. Watch the first couple of lessons in the Picture Postcard Workflow Part 1, which you can do for free. That's where Dan explains his current workflow recommendations, including why he used to recommend other methods.

You might want to check into Real World Color Management by Bruce Fraser, Chris Murphy, and Fred Bunting.

[Peter De Smidt]

From RWCM,

The components of Color Management.

All ICC-based color management systems use four basic components:
>PCs. The profile connection space allows us to give a color an unambiguous numerical value in CIE XYZ or CIE LAB that doesn't depend on the quirks of the various devices we use to reproduce that color, but instead defines the color as we actually see it.

>Profiles. A profile describes the relationship between a devices's RGB or CMYK control signals and the actual color that those signals produce. Specifically, it defines the CIE XYZ or CIE LAB values taht correspond to a given set of REG or CMYK numbers.

>CMM THe CMM (Color Management Module), often called the engine, is the piee of software that performs al the calculations needed to convert the RGB or CMYK values. THE CMM works with the color data contained in the profiles.

>Rendering Intents. The ICC specification includes four different rendering intents, which are simply different ways of dealing with "out-of-gamut" colors--colors that are present in the source space that the output device is physically incapable of reproducing.

The PCS is the yardstick we use to measure and define color...The key feature of both CIE XYZ and CIE LAB is that the represent perceived color."

"It's this property that makes it possible for color management systems to use CIE XYZ and CIE LAB as teh "hub" through which all color conversions travel. When a color is defined by XYZ or LAB values, we know how humans with normal color vision see it."

...

"A profile can describe a single device, such as an individual scanner, monitor, or printer; a class of devices, such as Apple Cinema Displays..., or an abstract color space, such as Adobe RGB 98...But no matter what it describes, a profile is essential a look-up table, with one set of entries that contains device control signal values--RGB or CMYK numbers-- and another set that contains the actual colors, expressed int he PCS, that those control signals produce."

[Peter De Smidt]

Are you making one standard file then applying profiles at the end and seeing dead nuts reproductions on your first test on different stocks from the same file?
When changing stocks how you make your file adjustments to compensate for all the different surfaces or is your attachment of a profile at the end doing this for you.?
Do you change stocks daily? Do you make your own profiles for the stocks you are using?


I start with an input profile. This is either a scanner ICC file, or a camera specific file in the case of a digital camera. With a scanner file, I assign it to the file in Photoshop and convert to a working space, usually Adobe 1998. If it's a digital camera file, Lightroom takes care of the conversion to Adobe 1998 or Pro Photo RGB. All Photoshop editing is done in these color spaces, with a possible mode change to LAB, but I always end up with an RGB file. This file is not converted to a printer output space. Rather, I use soft-proofing, small corrections and test prints to optimize the file for a specific output. I print from Photoshop for color prints, specifying the printer profile in the appropriate area of the Photoshop print dialog. The working file itself is never converted to the output profile, as this would damage the file.

Images on a monitor and on a piece of paper are not the same. They will never match exactly. Small test prints are always necessary no matter how much color management you use. If possible test prints should be viewed under display conditions.

I don't print on very many surfaces. Yes, I do make my own output profiles. No, I don't change stock daily.

Well, that's enough free consulting for one day.

[bob carnie]

Just consider it the betterment of another soul.

I start with an input profile. This is either a scanner ICC file, or a camera specific file in the case of a digital camera. With a scanner file, I assign it to the file in Photoshop and convert to a working space, usually Adobe 1998. If it's a digital camera file, Lightroom takes care of the conversion to Adobe 1998 or Pro Photo RGB. All Photoshop editing is done in these color spaces, with a possible mode change to LAB, but I always end up with an RGB file. This file is not converted to a printer output space. Rather, I use soft-proofing, small corrections and test prints to optimize the file for a specific output. I print from Photoshop for color prints, specifying the printer profile in the appropriate area of the Photoshop print dialog. The working file itself is never converted to the output profile, as this would damage the file.

Images on a monitor and on a piece of paper are not the same. They will never match exactly. Small test prints are always necessary no matter how much color management you use. If possible test prints should be viewed under display conditions.

I don't print on very many surfaces. Yes, I do make my own output profiles. No, I don't change stock daily.

Well, that's enough free consulting for one day.