QUESTION FOR TOM KNOLL

[bob_gr...@adobeforums.com]

Hi Tom,

Under PS versions 1-3 you created a great little piece of software called Knoll Gamma. It was brilliant for fine tuning a monitor to a proof before you and the lads decided to make everyone adopt colour management.

I still use it today under 9.2.2 Mac software since most profiles only get you just in the ball park and adobe gamma setup is not accurate for press paper whites in my opinion and this is where Knoll Gamma comes in great.

So my question is any chance you could make it available for system x on the mac Squire. Im sure there are many pre press guys out there who would agree with me.

Kind regards to a legend (no that wasn't grovelling!)

Bob

[Jeff_...@adobeforums.com]

Well, I can't speak for Thomas, but that would be an extreme step backward. . .before Photoshop 4 you had little you could do to calibrate a monitor except for monitor setup files. Photoshop 5 pretty much changed that and now a well calibrated monitor should be the norm.

Wanking out your monitor to "look like your proof" is basakwards. . .you create accurate monitor and printer profiles so one can do soft proofing and have total control over the output.

but I'll let Thomas answer you if he's so inclined. . .

:-)

[prog...@adobeforums.com]

at least Knoll let you adjust the individual gamma and still control the overall...with adobe its impossible to get the colour balance and then swing the gamma back to where you want it.

The problem with your method Jeff is that you need an output profile.... what i cant seem to work out or find out is what happens if you just dont have one and cant get one...not everyone works in a profile sorted enviroment...and ever since v5 this has been a right nause.

[Bu...@adobeforums.com]

You can do the same thing as Adobe Gamma in OSX with Monitor control panel in system prefs.

[Veni...@adobeforums.com]

Bob,

I agree with the previous posts that you are working "backwards" and that your attempts to defeat color management will ultimately defeat you. You need to understand the principles of color management.

Using hardware or software to adjust your monitor to any given proof is a recipe for disaster. We used to work with printing houses that did that, (and nothing was ever predictable). But not in the last few years. Everything is color managed now and the printer's proofs and the final job match our proofs (and the appearance of the layout on our monitors and theirs).

That said, you might want to try SuperCal,

<http://www.bergdesign.com/supercal/>

As a "software" calibrator, it does a better job than the Apple or Adobe utilities and is more flexible. I have compared it to our Eyeone setups over the last month and have found the results as good (scary). Your mileage may vary.

VL

[prog...@adobeforums.com]

but what i cant follow is if what you do when you dont have a profile???? surely then you must calibrate to the proof you have?

[Ann_She...@adobeforums.com]

You can create quite useful semi-profiles to proof output from a particular press ("semi" because they work only one-way) by using the Custom features in Color Settings.

Choose Custom from the "Working Spaces/CMYK" pop-up.
Then Custom from the "Ink Colors" pop-up.
And set the swatches to match your existing hard proof.
Insert appropriate values in "Separation Options" and Save with a suitable name.

Then you just load your new "Profile" into your Proof Set-up/CMYK Profile and Save to a useful name. You will subsequently find your new Profile listed in your View menu's drop-down menu.

[prog...@adobeforums.com]

cool...

well the new kit is comming next week, so hopefully we'll nail this demon once and for all

[Welles_...@adobeforums.com]

Venecia L,

Thank you for the suggestion of trying SuperCal. My old Cinema Display looks better than it ever has before after creating a profile with that software. I paid for it within ten minutes of finishing the profile. It was that valuable.

It's funny...I've been a fan of BergDesign free filters for years but never tried their products. Go figure!

Cheers!
Welles

[Veni...@adobeforums.com]

Welles,

I'm glad someone else here has finally tried SuperCal. I posted a question a few weeks ago asking for others' experience with it. I wish others would download it and give some feedback (it's "honor ware" so it costs nothing if you don't like it.)

I'm almost afraid to say that it does as good a job as the Eye one. But I haven't found much difference in screen performance swithching between the hardware profile and that from SupeCal. And I can run a new SuperCal profile in a few seconds if needed.

VL

[bob_gr...@adobeforums.com]

ok

lets start with Geoff, a fellow scanner operator I believe.

Before PS 5 we used setup files with Knoll Gamma to get very close proof to monitor closeness. Without this"closeness" the best scanner operator in the world can not see contrast on a screen just by reading dot percentages because the highlight brightness was way tooooo bright. On film it was easier you could see the overall brightness of your separations and know straight away if the job was flat even without an eye glass.

My point here is that if I line up 6 different makes of screen monitors and apply adobe gamma then load a standard photoshop cmyk setup and load the same picture on all screens there is irregularities to every screen. They are not identical!!!
They are in the ball park but not 100% facsimile in all six monitors.

Now I can invest in colourblind software @ $A25000
if I had a boss willing to spend it but I don't.

The press profiles that ship with photoshop are not close for anyone else in the world except America maybe? because it is based on their inks. Photoshop never mentions the make of these inks so it allows us to modify the magentas and cyans via cmyk custom in PS 7 Now after several hours of twinking I can match a proof very close. One thing that always amazes me is when I move the white ink to 255 I get more highlight detail???
BUT.. Highlights and shadows are still incorrect for a accurate
proof to monitor visual this is where TOMS knoll gamma comes in.

It allows me to twink the monitor without effecting the profile and that is its beauty. I have extremely accurate monitor to press , monitor to proof, monitor to Indigo setups. They are that close I could almost do away with not reading the info palette.

It works, it works very well and is predictable. It is not a step backwards its a step towards quality when you do not have profile making facilities. The staff adopt colour management very well,
the bosses don't want to spend the money. Their attitude in too many cases ...they pay you to make sure the jobs right!

By the way TOM if you get a chance to nosey in on this any chance you can put the info box under the tool box the same width just allow it to run down the rest of the screen. Its forever in the way at some stage of retouching. Perfection then .

Sorry there's work to do

Keep the thread going it's getting interesting.

Bob

[Ann_She...@adobeforums.com]

Venicia:

I tried SuperCal too -- and I am also impressed.

[bob_gr...@adobeforums.com]

Yes your right Chris thanks

I can see that ...

Thread ended

Regards

Bob
Ps How about the info box idea though?

[Chri...@adobeforums.com]

Photoshop's Eurostandard profiles are build using Eurostandard inks, press conditions, etc. They should be very close as-is.

But, Knoll Gamma is dead -- leave it that way.

[g_ba...@adobeforums.com]

Bob,

Excuse me if you're seasoned with ink for blood...your #11 caught my attention.
I am just trying to get the theory right in my head.

> Before PS 5...the best scanner operator in the world cannot see contrast
on a screen

I thought Photoshop4 folks back then didn't put too much stock in their monitors, mostly going "by the numbers"?

> if I line up 6 different makes of screen monitors

The "theory" is if each monitor has an accurate, "calibrated" profile driving it, Photoshop's Adobe ACE will Convert the file's tagged colorspace into the custom monitor profile and display it accurately — and — each monitor will look as identical as possible.

> in PS 7 Now after several hours of twinking I can match a proof very
close

Are you building a target CMYK device profile/colorspace there, or are you tweaking a CMYK file for press?

>TOMS knoll gamma...allows me to twink the monitor without effecting the
profile

I think modern, ICC-savvy Photoshop workflows work the opposite:

1) Get the monitor right, and 2) THEN, get the target profile right.

Never go back and dink with an accurate "calibrated" monitor to compensate a bad target profile?

[Veni...@adobeforums.com]

Ann,

Do you have a hardware monitor profiler?

VL

[bob_gr...@adobeforums.com]

Mr Ballard

You are correct in your statements from a theory point of view. But in answer to your comments....

I thought Photoshop4 folks back then didn't put too much stock in their monitors, mostly going "by the numbers"?

Knoll gamma allowed me to set up for stocks... my comment was... even going by the numbers you are not good enough to interpret overall contrast of the scan by numbers you need to see something physical and if a monitor white point cannot match the paper white you are not going to get that ability from the numbers.

"Photoshop's Adobe ACE will Convert the file's tagged colorspace into the custom monitor profile and display it accurately — and — each monitor will look as identical as possible. "

Thats maybe the theory IF people inbed files. It is not happening in my example of 6 different monitors .

I think modern, ICC-savvy Photoshop workflows work the opposite:

1) Get the monitor right, and 2) THEN, get the target profile right.

Yes I am tweakng a standard photoshop file It is first a case of finding out the level of white brightness (or too much of it usually) and black shadow detail I can get from the profile before I need to twink the monitor to kill the brightness or open the shadows. Then I go back and adjust colours. It is a back and forward problem but If you think setting 5000 or 65000k as a white point is the whole standard for paper white matching then why doesn't it work with any photoshop profile? They are not in the ball park when it comes to the white end. Lets discuss it

"Never go back and dink with an accurate "calibrated" monitor to compensate a bad target profile?"

Then how do you get quality matching if thats all you have?

Good points thanks

Bob

[Chri...@adobeforums.com]

Bob - then something is wrong with your settings, or the person running the calibration software is color blind. Even with Adobe Gamma you should get CRT displays very close. LCDs are another story (and harder to deal with).

[bob_gr...@adobeforums.com]

Hi Rene,

."It seems to me you are throwing out the baby with the bathwater. " I am not....

My point here is setting up adobe gamma and loading a photoshop cmyk file will not get me accurate press results on a sheetfed press in CMYK. Unless I have a super duper calibration profiler with a spectro so I can record a profile from those inks I can only get on the outside edge of the pitch with standard profiles. Photoshop would put every profile maker out of business if they supplied that.

This is my point IF you do not have access to this PRICEY equipment you have to twink.

knoll gamma got me in the centre so I could see the whole pitch.

I do understand the need for continuity world wide and my ideas are probably not helping this forum since standardisation is what we need at a colour management level, and like RGB profiles we don't want to get into STD RGB versus Adobe RGB argument again do we?

So I bow out now with by baby little Tommie Knoll

Bob

[Rene_W...@adobeforums.com]

The "theory" is if each monitor has an accurate, "calibrated" profile

driving it, Photoshop's Adobe ACE will Convert the file's tagged colorspace

into the custom monitor profile and display it accurately — and — each
monitor will look as identical as possible.

And it works in practice, the room I am currently sitting on has four calibrated monitors and the same image on all four of them looks amazingly similar...

Thats maybe the theory IF people inbed files. It is not happening in my
example of 6 different monitors .

That's what educating your customers is all about. Besides, nine times out of ten all it yakes is a phone call asking for the person's colour setup to know what profile to apply and telling them to check the "embed profile" box when they save a file as it can't hurt and can only do some tremendous good..

Yes, there are still implementation problems insofar as not everyone uses profiles (yet), but we are getting there. It seems to me you are throwing out the baby with the bathwater.

[Ann_She...@adobeforums.com]

Venicia:

Currently I don't have a hardware monitor profiler.

I had a Colortron but it has stopped working.

However,I do have some reasonably well-trained eyes, and a very nice LaCie electron 22 blue III, and my output from press and in-house printers turns out as expected. Fortunately.

[g_ba...@adobeforums.com]

> IF people inbed files

If they don't tag the — CMYK — it really doesn't matter UNLESS we want to see it "accurately" on a "calibrated" monitor — or — we want to do any Conversions in Photoshop.

If they don't tag the — RGB — they screwed their own job.
In which case we Assign the profile that looks best on our "calibrated" monitor...and get it into our workflow.

If they deliver RGB — tagged or untagged — and we ignore the tag: We hose the color. This realy needs to be fully understood...

+++++

The best I understand — if the file is untagged, mysterymeat — and we need to see it accurately on screen or make any Conversion to any target device colorspace — we need to:

PS> Image> Mode> Assign Profile (that best renders on our "calibrated" monitor...

+++++

Then again, I work in a RGB workflow and balance my color 90% off my calibrated monitor — your milage may vary...

[g_ba...@adobeforums.com]

> So I bow out now with by baby little Tommie Knoll

Still, it _would_ be interesting to read Mr. Knoll's comments here on that statment:)

[Mike_O...@adobeforums.com]

Chris Cox - 06:07pm Nov 17, 2003 Pacific (#12 of 22) Photoshop's Eurostandard

profiles are build using Eurostandard inks, press conditions, etc. They
should be very close as-is. But, Knoll Gamma is dead -- leave it that
way. <

The Euro ink set is cleaner Cyans, the Magenta is about the same, the yellow is VERY dirty and the blacks are somewhat close.

And, no, Knoll Gamma.......IS.......very much alive and well in the commercial printing industry, but it's not being used as it was designed for.

[Jeff_...@adobeforums.com]

"And, no, Knoll Gamma.......IS.......very much alive and well in the commercial printing industry, but it's not being used as it was designed for. "

Which is one of the primary reasons the prepress and printing industries are so screwed up. . .

[Ann_She...@adobeforums.com]

"Euro" inks are not used only in Europe -- there are presses in the USA that use them too.

[Mike_O...@adobeforums.com]

No,

Because Adobe has designed Photoshop more towards photography and not print production.

The printing industry is screwed up due to the EPA, the stubbornness of those types of people in the industry, and the constant (EXPENSIVE) changes that are forced upon them.

I've changed Jeff......

and you?

[Mike_O...@adobeforums.com]

It's just a big pot of soup. Stick a fork in it and hope it stands up.

[prog...@adobeforums.com]

one thing i wonder about with profiles....what if your profiled system still doesnt give you either a) accurate screen preview of colours b) the colours look right on screen but the printer doesnt want to play?

what im trying to get at here is if your monitor varies too much to use a standard profile, then surely output devices can vary as well which either means that a new profile is required or the monitor needs adjusting. Also what happens if someone gets hold of your file and junks the profile somehow...? is there a way of getting a file to match an output without a profile, so the danger of having a profileless file is removed and that someone doesnt have 3 choices of how to treat the file when they bring it into their workflow...?

[prog...@adobeforums.com]

it seems to me that profiles are getting around incorrect devices rather than a) getting devices to work properly b) knowing what to do to a file to get it to print well. I realise that the profile idea is designed to solve point B, but you still have to adapt a file to get consistant colour if you change profiles anyway...all they seem to do to me is get inbetween a good calibrated monitor and a good output system...if you have both of those, why do u need profiles if you need to change the file per device anyway?

[Mike_O...@adobeforums.com]

one thing i wonder about with profiles....what if your profiled system
still doesnt give you either a) accurate screen preview of colours b)
the colours look right on screen but the printer doesnt want to play?<

The profiling software and or hardware sucks.

The monitor profile software and or hardware sucks.

The monitor sucks.

You suck...(the profile-eeee)

The proofer sucks and isn't stable.

what im trying to get at here is if your monitor varies too much to use
a standard profile, then surely output devices can vary as well which
either means that a new profile is required or the monitor needs adjusting.<

ANY device that has huge variations will not give you good results.

Also what happens if someone gets hold of your file and junks the profile<

Depends.

If it is a device that will accept ICC profiles AND is doing a conversion to some other color space, then yea, throwing away the profile that created that image is a bad idea.

[John_...@adobeforums.com]

ANY device that has huge variations will not give you good results.

Which is why the concept of profiling a printing press is flawed at best.

[g_ba...@adobeforums.com]

What kind of professional device has "huge" variations?

I suspect the "huge" variations are the smoke & mirrors of a broken, ill-understood workflow?

The best [I] have today is:

1) A calibrated monitor and good profile to proof my color on screen.

2) A good Color Management System — Adobe ACE — to do the Conversions to specific target device profiles and colorspaces, and

3) The interest in finding the weak link when the target-device proof doesn't match the calibrated/profiled good screen.

At the point Photoshop/AdobeACE cannot accurately CONVERT a tagged file into MonitorRGB, and Photoshop/AdobeACE cannot accurately CONVERT a tagged file into a target RGB/CMYK/Profile/ColorSpace — I believe we have reason to complain.

Here, I just don't see the problem.

Either we have a good, tagged file, a not-broken Photoshop, and good target profile/colorspace — or — we have a bad file, a broken Photoshop, or a bad target profile/colorspace.

How can it be otherwise???

(Duck and run for cover)...

[Mike_O...@adobeforums.com]

And what happens when you have a system that does not address black point compensation and file conversion is out of the question for that specific image and it needs to match a contract proof?

HMM?

What do you do?

HMM?

Tell the client you're screwed?

No, I'm sorry, you work around the limitations of the development of the software because not enough print professionals have had enough input into their own industry because traditionally, photographers have been steering the ship by looking at the stars and not at a compass.

hmm......

[John_...@adobeforums.com]

One man's "huge" is another man's 6 inches I guess.

How many offset presses have you dealt with on a daily basis?

[g_ba...@adobeforums.com]

> How many offset presses have you dealt with on a daily basis?

Zero, but that hits the heart of _my_ theory!

WHY can't press professionals with years of daily work experience be consistent?

Why do print professionals with years of daily work experience ignore tagged RGB profiles and hose our color on an hourly basis??

If the file is good, and it is properly Converted to the accurate press profile/colorspace — where lies the problem???

[prog...@adobeforums.com]

g- i think the problem lies less with RGB and more with CYMK.

if you have a calibrated monitor then the RGB is less of an issue, but Knolls gamma allowed better control to software monitor calibration.

But with CYMK files its a different ball game. In earlier versions you could say the file that you had on screen had gone to print, but now you cant say wether that is the case, or that your profile is inaccurate, or that the profile has been lost, or that the profile has been changed...so effectively with a calibrated monitor accurate CYMK is now guess work unless you have the output device along side you and or are 100% certain of what happens to the file once it has left you. As i keep asking, what are you supposed to do if you dont know what device you are shooting too or how the file will be handled? Also, what do you do if your client proofs on one device then outputs on another?

All i want to do is make my file, convert to CYMK predictably, get the client to proof it, work with their colour comments, and pass it back so that their proof looks how they want it. Then the idea is they give the file and proof to a printer and say HIT THAT,...he can either hit it, fiddle with it or fail depending on the capability of the team and kit. But at the end of the day I want my C63y17m34k8 to hold its value and look like that, so if someone picks up the file later and moans about colour shift i can open the file and say its fine here and not worry about a profile screw up.

At the moment its a struggle to convert R128 G128 B128 to cymk. The monitor says its grey, it converts to CYMK and looks grey, but the runouts show it to be varying green on different devices. So now if the RGB is ok, the representation of CYMK is grey on screen, where is the error creeping in? Profile or printer? Do i find a different profile that fits (which surely cant be the way, cos thats back to guess work), i cant adjust the monitor cos the grey is neutral on screen or do i get the spanner out on the printer or do i profile the printer around a dubious print? Without profiles i could get a proof back and say simply that its a bad proof or my monitor is out...i can simply look up a colour mix to tell which is which...with profiles that seems to be out of the window, unless someone can elighten me.

Im so lost with this i cant even work out what happens if i dig up an old file to work against, if it contains a working space CYMK profile based on that machines settings what now if im working on a newer version with different settings?

[prog...@adobeforums.com]

I got it...stop everything...everyone, go back to v4, it all worked then.

Seriously i'd love to know what im supposed to do in this position...

[Bu...@adobeforums.com]

Its all Crayola's fault.

8-}

[Peter...@adobeforums.com]

Progress....

I'm sorry that you have to live in the dark ages. It sounds like your problem is that you don't have the correct output profile. No amount of Knoll Gamma is going to help you. This is going to sound harsh, but you need to quit making excuses for not having the right tools to do your job. Either get profile making hardware and software or have someone make you some custom profiles for your most frequently used proofing devices. If those are drifting, then nothing is going to help you, but if they are stable, you'll be amazed at how well everything works, and at how often you'll hit your proof on the first try - and that it looks remarkably like what you saw on your monitor. It will become the exception and not the norm when you run into problems. I haven't experienced the troubles you relate in years, and it's because everything is calibrated and profiled, and I have the correct output profiles for the devices I'm outputting to. Knoll Gamma is unnecessary.

[Rene_W...@adobeforums.com]

Its all Crayola's fault.

And those paint by the numbers kits :P

[prog...@adobeforums.com]

im not living in the dark ages, im living in the world where my artwork may hit one of 100 magazine covers across the globe, where it maybe printed on the cover, inside, on their cd, cd cover, actually appear on the cd show, on a billboard, TV (PAL + NTSC), web etc etc and thats without taking stock or coats into it. The artwork will be handled possibly by an agent, agency, bureau, client, art dept or all of the above. The file may need to be in RGB, but we give them out in CYMK so at least the colours are in CYMK gamut. We dont know the uses of the file at all, all we know its what it could be used for. I dont have to pick one, all i have to do is put a proof with the image and say "thats what it should look like". But now i have to worry if my profile is suitable, or remove it and be lost with regards to my proof anyway, or worry that some will change it or wont change it depending on the needs at the time. It would help if i even knew what output the client will proof on, but thats their choice...the files leave us cold, we can provide a laser but they're useless for proofing. We are image makers, not proofers, pre press or printers.

To put it in short we dont live in a profiled world. I can profile an output device, but we dont know if the device is accurate or if our profile is. Machines do vary depending on setup. A client of ours had his printer profile nuked after an engineer decided to update the firmware, it took 3 months before the manufacturer even admitted there was a problem. Another 2 to get new profiles...

I agree, in an ideal world, you have a calibrated monitor, and a profile for all devices, anyone who touches your file understands exactly what to do with the profile and when. BUT, that isnt reality.

Lets try one question then, assuming you have a profile for a device and a properly calibrated screen, when your output doesnt match your screen (ie theres an obvious colour shift) how do you determine wether the profile is incorrect or the device?
Its no good working up a profile around a workflow where you dont know if the profile isnt working properly isnt being handled properly or if the device is working properly. The only way to check is with a densiometer and even then all you can show is wether your output is indeed matching your input. You still dont know what could have happened to your profile.

Im not making excuses, im simply saying that before it was possible to calibrate around a proof in earlier version, as long as you were aware of the drawbacks, but now you cant even do that unless you are present 100% of the way until it slides out of whatever output device. This again, isnt my reality or a lot of peoples'.

If someone can explain how i can get with the program on this then fine, please do, or how i can work around it...but since v4 colour consistancy in the real world has gone out of the window despite good intentions.

Put it this way, when you send a type job off you send the fonts off, you dont want anyone slipping in their flavour of gillsans or the printer device choosing its linotype version, you just want your flavour. Colour profiles are no different. At the moment i cant see a way of knowing or stopping a file from being monkeyed around with, so it means any colour prediction is shot away. All i want to know is how to keep this unpredictable element out of the equation.

I guess that I want to be able to output a file, without profiles getting in the way and then change my CYMK representaion on screen without profiles to match the proof on a reliable output, and then lock the bugger down. I know that output devices cant match others, i dont need to see it or adapt the file, the match is the the skill of the printer or pre press, i dont want to make 50 versions of my file to just keep the sky blue like the client likes on the front cover the same as the inside.

[Steve_V...@adobeforums.com]

Do PAL and NTSC have different colour spaces as well as the resolution difference?

[Chri...@adobeforums.com]

Yet without the profile attached, the receiver has no idea how it should appear - and thus you've got even LESS chance of it coming out right.

Profiles were made for situations like yours.

[Chri...@adobeforums.com]

Steve - yes.

[Mike_O...@adobeforums.com]

WHY can't press professionals with years of daily work experience be consistent?<

G and P,

I'm gonna be straight and to the point with you.

Some if not most printers are arrogant, egotistical, SOB's that have seen their industry turn from a "craft" to a production process. Printing is and still will be a "craft" to some extent, BECAUSE the nature of a printing press is one of the most difficult devices to control on a regular basis and here's why.

First off, most printing presses are on the older side of worn out. Not all, but the smaller shops, are trying to get buy with the equipment that they have. They are expensive to maintain, repair, and control. Temperature, humidity, ink/water balance, dot trapping due to the grade or quality of the ink set, blanket packing, type(S) of blankets, thickness, pressure, paper stock plays a huge roll in the process as we all know and the fact that a press usually does not print or behave consistently from day to day.

Press rooms usually don't have controlled environments to help stabilize the behavior of the device. In a high, high end shop you'll see air conditioned rooms with humidifying nozzles that spit out water vapor every 10 minutes or so to keep the pressroom at a constant known and control able state. Those very high end shops are far and few between because of the cost associated with that kind of high tech investment to achieve a controlled final predictable print.

Your job will be consistent all right and your bill with be consistently high as well.

( You get what you pay for)

Another reason why color manglement is in the toilet in the real world is because of the cut rate that every shop has to offer to stay alive in this economic environment. It cost money, a lot of money to keep an expensive piece of equipment running in prime condition.

The market has basically kicked this technology (color mgmt.) to the curb before it has had a chance to catch on in the printing environment. Printers are basically forced to cut costs where they can to stay competitive so color mgmt. is one of those areas that get chopped because it's seen a feature or sales tag. The industry has been wild since it's inception so why spend a bunch of cash on something that they basically give away for free.........

Remember, all the color geeks out there that are affiliated with Adobe all say that the printer should profile their press and provide that file at no cost to the customer... I don't know what they have been smoking, but it ain't anything that ANY printer will do that has any sense of self preserverance. In color geek land theory, I think it's a good idea. From a business stand point, it's just stupid to give away your work. Remember, printers youst to charge for things like color separation. On the flip side of things, there are a fair amount of printers that think CMYK is a single set of numbers and a single color space if they even know what a color space is......

There is also the horrible and painful reality of dealing with direct to plate calibration and the high level of accuracy that it gives, which in return yields printing processes to be too precise and induces other issues such as scum dots or rosettes that don't trap correctly because the half tone dot is too sharp. Also, because of this high level of accuracy in "dot sharpness" any minor move, and I'm talking 1/3 of a % will yield a color shift, especially in pastels or highlight areas. In some ways the tradition way of creating press plates was much more forgiving. You can fudge things with a softer dot. And then there is sheet flair where as the trapping of the plates change as the paper is ran through each tower of the press. Sheet flaring happens due to blanks and/or cylinder pressure upon the specific stock that is being ruined. Cheep paper may save you money up front, but you'll have to accept what comes out the other end at press check time or change the stock. And changing the stock will shift the behavior of ink on paper. Plate registration due to sheet flaring will induce a perceptual shift in color.

So it's buyer beware. It's up to you, the customer to be informed what you are buying and know where and when to point fingers and why. (not a small order)

Dealing with large format printing is a whole other matter and has it's own set of rules to follow. ICC profiles play a much larger roll in this area because most, if not all files are cross rendered from the assigned profile of the hopefully color separated or captured RGB color space, and then converted to that proofers output color space. AND even then there are issues to be aware of when dealing with different ink sets, paper, RIP's and drying heaters, on and on.......

Try profiling a large format proofer that basically takes gasolineized ink and prints it on canvas. A press would be a dream to deal with compared to this device.

So basically, it's all good clean fun this color stuff. and then there are the bugs or limitations of softwares that throw you through a loop as well, leading to endless hours of research and tons of materials down the drain.

Then there is file conversions, people who slack in preping the files, behaviors and or good or bad habits....

It's a mess.

So you see, it's not that simple........

Not one bit.

So.............

Profile the proofing device and have the press operator hit the proof. Make sure that they have a proofing system that is based upon some kind of standards that is somewhat close to press behavior or SWOP standard. SWOP standards, by the way, is a pretty loose tollorance.

Oh, and don't get stupid by trying to make 1/2% moves on files, and spending a bunch of time justifying your job or some designers crack pipe dream, that will never reflect the final printed sheet.

[Mike_O...@adobeforums.com]

Yet without the profile attached, the receiver has no idea how it should
appear - and thus you've got even LESS chance of it coming out right.<

The reciever takes what you supplied and gives you back a proof.

You figure out WTF it should look like.....

[bob_gr...@adobeforums.com]

Wow Tom

you could have been enjoying a good thread here.

Well said Steve. Even here in Australia I'd have to agree except for the...."most printers are arrogant, egotistical, SOB's" naaa we don't have any of them here!

As you mentioned The Plotters (especially Agfas) are very accurate to the press inks now as Chromalins, Fujis and Matchprints become way to expensive to proof on for most run of the mill jobs.

If colour management is to succeed it is the profiles for the "proofing system" ie plotters, chromalins etc that should be given out. These are very stable and consistent because what's on a quality proof can be achieved on any press.

The printer could not give a toss about colour management he is only interested in a proof he can trust, he knows he can match something physical once its on the press .

Classic example I have just done a job for Singapore now we quoted $35 000 to print this Singapore quoted $18 000 so guess where it went. Asked for a profile none available asked for a dot gain told 14% That's good so are our presses so we scanned "as normal"the job was wet proofed in Singapore with FM screening came back 8% weaker than our screens predicted. 300 scans to colour correct!! No way. They told us to proof it our end send them the proofs and they would match it on the press!!! and that they did.....

"Its a long long road from which there is no......" Knoll gamma

Bob

[Mike_O...@adobeforums.com]

I don't want to make 50 versions of my file to just keep the sky blue

like the client likes on the front cover the same as the inside.<

Than tell Adobe to redesign the core functional workflow of managing color for CMYK files.......because it will be a cold day in hell when the printing industry could ever adopt a RGB workflow considering the call outs that designers make.

and that reminds me.....

I need to beat Bruce up after he gets back from Australia.

;o)

[Chri...@adobeforums.com]

Mike - why waste a proof?
Why not just attach a profile that ALREADY describes how the image should appear?

(because pressmen don't want to be held accountable.... I know)

[bob_gr...@adobeforums.com]

Come on Chris

Your too much in the theory world. Try running a press at 25000 copies an hour on paper that shrinks and stretches and keep register in all 4 colours then ask the press guy to run his ink density to such and such density and he'll be right because Colour management is alive and well. You wish..

The proof is there for more reasons than a colour profile
accuracy, Imposition, type mistakes, and a colour version for the customer to sign off that this IS what S/he wants. The pressman is accountable always ...to the proof.

Colour management has removed the need to scan/colour correct an image 10 times before the Art director says yep thats OK. It is not or should not be seen as replacing a proof. IMHO

[Rene_W...@adobeforums.com]

Lets try one question then, assuming you have a profile for a device and
a properly calibrated screen, when your output doesnt match your screen
(ie theres an obvious colour shift) how do you determine wether the profile
is incorrect or the device?

Well, here are some of the things we do:

Try to reproduce the same problem with a file we are familiar with (as opposed to the customer's file)

Check known images on the (calibrated) screen of several computers to see if there is a disrepancy between the monitors.

Check (and probebly redo) the calibration of the output device.

And, oh yes, since we have some really picky customers, we know that all profiles will do is put us in the right ballpark and some minor tweaking is sometimes needed, though a _lot_ less than before we implemented colour management..

Progress, this is what I tell customers who aren't very knowledgeable about CM and don't want to learn about it:Just embed it, it can't hurt and it can only do you good. If the person getting the file ignores the profile, your colours are hosed wether the profile is embedded or not, and if they respect the profile, but it's not there, your colours are hosed only if you did not include it.

[prog...@adobeforums.com]

bob & mike, your a breath of fresh air here...i know im not savvy on adobe CM, but i know theres real world problems with profiles.

I dont want to start a slanging match, so i'll try and take on board what people are saying and ask questions.

I should embed always. Fine. So basically all PS's in the studio at least should use the end profile. Ok, thats understood. But i'm correct in thinking theres no way to stop someone hosing the file?

I should find a profile, or tweak one to make a match to my proof or have one made, it doesnt matter which or what its called as long as it works yes?

If devices are hitting cromalin constancy (and colour accuracy?), then are we saying the devices output the same or that they output constistantly with a profile? ie flicking between profiles still show a difference between the files?

If my profile is changed after my file leaves me i'm screwed yes?

Rene, we make the files, theres no file comming in,...im checking old and new files with a new client and their new proofer. The old and new files neither match the new proofer or the repro output of the old file. So i guess the profile is screwed, so what do i tell the client? I need to predict a proof on my screen at a moment in time, and this time its lucky i see a proof before it hits wet press. So do i assume the profiles wrong and run, hoping the colour bias will shift back like it did with the old file when it hit press, or do i assume the machines out, or do i ask my client to profile their proof which will be like asking them what their serial no. of their washing machine is? Can i even assume that they havent hosed the colour?

Anyway, so when the client then gives me a file with another profile, do i adopt theirs, convert, or ignore? Which gives me the best chance of holding their colours in MY working colour area?

Im not trying to be confrontational, i really appreciate the help from you guys answering my questions. I would really like for some light to come on and for me to say "i now have confidence in Adobe's CM, everything will be fine"

[Mike_O...@adobeforums.com]

bob,

Chris Cox HAS been in a prepress environment in a past life, but as everyone knows, it's changed a lot.

And Chris, liability is only part of the issue of a pressman. Everyone in any field has to justify their job somehow (some not very well I might add).....

Pressman, would rather have their device more consistent than not and be in line with some proofing system. Once a file has been converted to CMYK, regardless of if it was the correct one or not, you can toss the profile, and that's where the problem is.. (extremely long topic better suited for Alpha discussions)

There is a trend towards profiling a press and cross rendering it to an Epson 10600 for proofing, but there are STILL clients, usually ad agencies who demand a halftone dot proof and won't accept anything else.

In this case, color mgmt. is a wash except for RGB files that are being converted to that color space. .........but everyone knows that the majority of files in commercial print are coverted before they know what device it shall be printed on and that's were color mgmt. has a big hole to drive a semi through.

Changing bad behaviors through education and years of wrong way workflows will have limited success.

Both Bruce and I have been on this merry go round time and time again.

Unfortunately, there will come a time when the engineers will have to deal with this subject and is the next logical step in the evolution of addressing a very messy topic. Messy is one of the reasons why no one really want's to touch it, but hey, I'm game........

Chris know's more about many things than he leads on about and I'll leave it at that.

and me........., confrontation gets you published!

;o)

[Peter...@adobeforums.com]

"im living in the world where my artwork may hit one of 100 magazine covers across the globe, where it maybe printed on the cover, inside, on their cd, cd cover, actually appear on the cd show, on a billboard, TV (PAL + NTSC), web etc etc ....We dont know the uses of the file at all, all"

In these situations, the best you can do is head for the greatest common denominator. Magazines printed in the U.S. are remarkably similar these days, so it shouldn't be any big deal to hit that market. Most billboard printer don't know it, but are printing SWOP files all day long and have adjusted their output to make those file look good.

The real key to your solution is communication with those who will be handling the files downline from you. They need to know what you did - to what standard the files were separated to so they can make an intelligent decision as to how to best procede.

You're correct that when you don't know the exact output, you don't know exactly how to prepare a file, but if you think that resorting to Knoll Gamma will help you, then it all falls apart. If you correct and separate a file to SWOP on a monitor that is tweaked, then you have no idea where you really are, and if you do happen to embed a profile, it means nothing.

I think it was mentioned somewhere earlier in the thread, but it bears repeating, that you can easily alter the CMYK preview by going into the classic non ICC cmyk setup and tweaking the dot gain curves and printing ink values to create something that forces the preview to match, all while maintaining your standard monitor calibration. Of course, it means seeing at least one proof beforehand in order to make the adjustment, and probably another to test it out -- but if you don't know what your output is going to be, then where ever you decided to make a proof wouldn't have any validity anyway.

"It would help if i even knew what output the client will proof on, but thats their choice...the files leave us cold"

Maybe the best you can do under the circumstances is to provide a calibrated RGB file with a guide print to go along with it, with instructions for whoever is responsible to make whatever separation or adjustment to get as close as possible to your guide. In order for this to work, you have to have an agreement with the vendor that this will be workable on their end too. I have done this many times with great success. It's a little backwards, but gets the job done with as little extra work as possible.

"Lets try one question then, assuming you have a profile for a device and a properly calibrated screen, when your output doesnt match your screen (ie theres an obvious colour shift) how do you determine wether the profile is incorrect or the device?"

Whenever I make a new profile, I always test it out. With a hardware calibrated Artisan or Barco, I'm pretty sure that the monitor is not changing, and I know the profile hasn't somehow morphed itself, so if there is a subsequent color shift, it has to be the output device.

ANY type of output, profiled or not is still dependent on the stability and repeatability of that device. Even if you decided that Knoll Gamma was your tool of choice, it still wouldn't predict any drift in your output device.

"Im not making excuses, im simply saying that before it was possible to calibrate around a proof in earlier version, as long as you were aware of the drawbacks, but now you cant even do that unless you are present 100% of the way until it slides out of whatever output device. This again, isnt my reality or a lot of peoples"

Modern ICC color management hasn't changed anything in regard to what you're saying. If you don't know the output, it's always a guess, but when you do know, then it's a no-brainer.

"If someone can explain how i can get with the program on this then fine, please do, or how i can work around it...but since v4 colour consistancy in the real world has gone out of the window despite good intentions."

For those who have taken the time to really learn, understand and implement, this simply is not the case. In fact the opposite is true. For most jobs there used to be an average of 3-4 round of proofing for every job. For those who successfully implement, that drops down significantly, in my experience to something slighly over one round of proofs for every job, saving both time and money.

It's possible, it works, but it requires a definite commitment to both learning and maintaining the workflow.

[g_ba...@adobeforums.com]

> The real key to your solution is communication with those who will be
handling the files downline

And if that doesn't work, I would think delivering RGB in ColorMatch or sRGB (a monitor-type colorspace), and US Web Coated SWOP v2 is the safest bet...?

[Mike_O...@adobeforums.com]

but if you think that resorting to Knoll Gamma will help you, then it
all falls apart.<

Peter,

Knoll Gamma is being used as a means to shift the gamma or the three channels independently towards a darker image, in order to get a better "temporary" view of pixels in the extreme highlight. AND, or show detail by lightening the image in darker areas.

It's being used as an evaluation tool and not as it was designed or intended for. It's not being used to dial in a correct preview, but as a tool for pixel evaluation in areas that are difficult to see under normal monitor calibrations.

WHY?

Because color adjustment layers such as Curves will shift the pixels which is not the intention of using Knoll Gamma.

When Adobe develops the next generation of color processing tools, this will be a MUST.

Curves, selective color, etc, will be a thing of the past. I just know it.....

"If someone can explain how i can get with the program on this then fine,

please do, or how i can work around it...but since v4 colour consistency

in the real world has gone out of the window despite good intentions."
<

I youst to think the same thing, but the it's a simple fact that color separation and monitor previews a KILLER and extremely accurate when set up correctly from PS 6 and up.

Times up, pencils down.

[Mike_O...@adobeforums.com]

Rick,

DUR!!!!!

Don't get me started.....

Not if you want to change the actual VALUE of those pixels. <

Only before a conversion is done....(Hense the Zen of color mgmt.)

You know what happens when the Big Guy "B" shows up....

Blam-O

Geeks come in all flavors, shapes and sizes.

The smell of a sound comes to mind.....

On a dime in time that is swept west ward with lime.....

[Rick_M...@adobeforums.com]

-- Because color adjustment layers such as Curves will shift the pixels
which is not the intention of using Knoll Gamma. --<

True. It changes the VIEW of those pixels without actually changing the VALUE of those pixels.

-- Curves, selective color, etc, will be a thing of the past. I just know
it..... --<

Not if you want to change the actual VALUE of those pixels.

----------------------

This is a great thread. Should be required reading for everyone who touches PS - and these days, that's pretty much everyone. (for better or for worse, 'til death do us part, etc, etc, etc...)

[Mike_O...@adobeforums.com]

In a tree you see, the geek of choice is glee..

All around with sound, that type has abound....

Um,

I think, therefore I MAC.

Just do it?

NO, just stupid............

[prog...@adobeforums.com]

yeh, i dont want to change the values of pixels, i want to change the view of them.

Ok, so here's what im going to do. Get a blueye lacie, calibrate, check with a printer that their proofs match somewhat their repro (ie theres no big bias or shift on good repro)...get a proof run.
Now before i run that proof do i attach any profile if i dont have one(or the printer dont) or do i run it off without one?

[Chri...@adobeforums.com]

Bob - the pressman is paid to match the proof. That's all he has to do, even for color management. It better if he can keep the press relatively stable (calibrated), but not required.

But color management means you can proof on a CRT (ok, cyans are still bad - but they're working on it, and the new AdobeRGB CRT comes a lot closer.), on a desktop printer or large format inkjet. You don't have to pull film and do laminate proofs for every f'ing page anymore. Yes, you'll still have to do some - but every one you don't do is money saved/earned.

[Chri...@adobeforums.com]

I know too many sites that still do them.

And if that shows my age, you should have seen the look I got the other day when I mentioned rubylith, amberlith, emeraldlith and swivel knives....

[Mike_O...@adobeforums.com]

actually they are still done from time to time when spot colors come into play.

Not all digital proofing devices can show traps.

I was just giving ya a jab as usual.....

[Mike_O...@adobeforums.com]

laminate proofs ?

Oh yea, Um, Chris, your age is showing......and you better check your fly.....

;o)

[Chri...@adobeforums.com]

I know...

[prog...@adobeforums.com]

its likely the proof wont match.

so do i send them an unprofiled file to proof and then build a profile around the result?

[g_ba...@adobeforums.com]

progress,

> here's what im going to do. Get a blueye lacie, calibrate, check with
a printer that their proofs match somewhat their repro

Agreed, in still other words:

Check with a printer whose proof matches your calibrated screen, whose proof matches their repro...

[g_ba...@adobeforums.com]

> You profile a device by sending it a known swatch target that the profiling
software understands when it's read in.

Is this what Ole No Moire.tif (UntaggedCMYK) is designed for, by chance?

[Mike_O...@adobeforums.com]

Yes and no.

That file (Ole No Moire.tif ) is just for a visual reference. (evaluation) A swatch target of lets say about 750 colors is much more accurate to draw conclusions from than a visual test file. That file is only part of the evaluation of device behavior and not the actual swatch target that would be used to scientifically read in values to create a profile.

Take a closer look at Andrew's web site.

[g_ba...@adobeforums.com]

> do i send them an unprofiled file to proof and then build a profile
around the result?

I half understand that theory, like the untagged files profile vendors have us print (with colormanagement off) for them to build the profile?

+++++

More to my understanding of the theory:

Send them a known good file, a file with an embedded tag.
If they hose it, AdobeRGB will show the error.

Or CONVERT the known good file to whatever CMYK they recommend.

I prefer PDITarget.JPG for the skin tones and neutrals.
<http://www.gballard.net/nca.html#getagoodprinterprofile>

If the proof comes back hosed, ask WHY?
And HOW do we get on the same page??

[Mike_O...@adobeforums.com]

G,

You profile a device by sending it a known swatch target that the profiling software understands when it's read in. It's sent as if it were a live job as far as workflows go. CMYK targets for CMYK devices and RGB targets for RGB devices.

It's pretty simple....

By sending a calibration file such as images with flesh tones, and grays and perceptually known images that your youst to looking at, is the profile evaluation process. You can do it before and/or after you profile the device for a comparison.

Remember,

For RGB files, color mgmt. somewhat replaces device calibration so if you have a standing file that has been sepatarated with a known profile such as the US SWOP prepress coated v2, whatever, profile, and sent it to a CMYK device, and it comes out screwy, then you know that the device is not calibrated to TRO01 standards or anything close to a SWOP setting.

It's called a run around to check calibration of a device.

[Mike_O...@adobeforums.com]

If that's what they want, and that's not what I read from the original
poster, simply assigning a false profile temporarily with to the same
thing. <

Peter,

there is no control by doing that.

[Peter...@adobeforums.com]

"Knoll Gamma is being used as a means to shift the gamma or the three channels independently towards a darker image, in order to get a better
"temporary" view of pixels in the extreme highlight. AND, or show detail by lightening the image in darker areas. "

If that's what they want, and that's not what I read from the original poster, simply assigning a false profile temporarily with to the same thing.

The bigger problem here is that people are still using some generic, just convert it to CMYK conversion that isn't at all right for the output, and then blaming the profile, the monitor, or possibly the moon for their proofs...

Peter

[Peter...@adobeforums.com]

Mike,

You're sort of changing the rules here. The original poster was using K G to wack his screen into submission. Now you're saying that they're just using it to "see" pixels better. What is it? Maybe they need to start out with a good monitor calibration first, and then move on from there. I truly thought that K G was a thing of the distant past. The only people who still use it haven't quite grasped the ICC thing yet. It does take time to sink in.

[Mike_O...@adobeforums.com]

It was not my intention to take this thread somewhere else, nor was I trying to answer the original posters question.

I'm kind of bad about that stuff, I know.....

Even if you have a great profile set-up for the monitor, you still need, at times, to darken or lighten the image for specific evaluation(S). .....before edits are taken to the next step.

It's such a pain at times to show each channel on it's own, then throw a curve on a color adjustment layer to get the highlights to darken up to make sure you don't have a scum dot or a bad retouch job.

The accuracy of a profile, is a moot point in this situation.

[bob_gr...@adobeforums.com]

The real key to your solution is communication with those who will be
handling the files downline

This is the most important point EVERYONE.

If you wish to supply a proof to a printer make sure it is of an industry standard (your desktop printer is not going to be in this standard!! ) Even the major proof makers vary in their choice of cyan and magenta as anyone who has compared a FUJi red and green to a CHROMALIN red and green?

Ask the printer what their preference is in proofs, their costs if they supply a chemical or digital proof .

Most medium print jobs today include a press proof in their costs. They are wanted by the printer because he needs to know that this IS what you want. You need to proof check type and make sure your images look like your profile is telling you they look like. Because at the end of the day the PROFILE is nothing but a simulation it is the dot sizes on the paper that will determine the end result ,as many of you find out too late after trusting your profile.

If you cannot afford a proof then you are at the mercy of the printer
and it will be his "eye" and what he "feels" looks a pleasant picture to the eye I'm afraid. He'll run his densities of inks as close as he can but his "eye" will have the final say. That's the reality. No proof you have nothing to fall back on.

IF you do not know the destination printer and cannot afford a proof then expect to get your file hosed! because you will have to rely on their quality level. But tag your file for sure let someone know exactly what this files destination is, if they hose it , it is for a reason but at least they know what ball park you want to be in .
IE swop coated 20% dot gain , NEWSPAPER 230 GCR 35% dot gain this is important stuff so please tag your files.

As a pre press chappie who works in photoshop 10 hours a day since PS version 1.5 and you send a file to me I am really not interested in your super duper profile I need to know what it is but I will convert it into my colour space because I know what I am seeing on my monitor is what our presses will print . Then if your file does not match your proof on my screen I'll let you know.
But in order to give you that advise I need Knoll Gamma because even the printer profiles I have are not accurate when I calibrate my monitor with adobe gamma SOOOO I have had to twink them!! Making them useless to anyone else unless your monitor characteristics are identical to mine! HUMBUG. This is because I do not have a company that will invest in profile making equipment and why should they when I'm getting F'ing great quality from little tommy Knoll Gamma.

If you want to supply a file as Mr Ballard says...

"And if that doesn't work, I would think delivering RGB in ColorMatch or sRGB (a monitor-type colorspace)",

Be prepared to be charged for this conversion if its only one image I'm sure it will be done free of charge, but give them 10-100 images and that is alot of time which has to be payed for.
The problem as you all know is the brightness and vibrant colours you see in RGB are not going to be seen in many cases when converted to the smaller INK colour space CMYK . So what you see on your monitor is not what you will get unless you have a very good profile from this printer if you do you should convert to cmyk and save your money!

I do not agree with this comment...."and US Web Coated SWOP v2 is the safest bet...? 80-90% of presses world wide would be a sheetfed type press of some description Webfed presses are for newspapers and long run magazine work so ASK at the destination the dot gain between a web and sheetfed press is considerable!!

Bob

[prog...@adobeforums.com]

agree with bob and mike here on several points, i use KB for just that to affect my screen representation when the profile fails to deliver. And i agree its the proof which the printer needs to match beacuse this is what the client will sign off before it goes wet. Only when its wet can you see whats what, but the idea is that its close to the proof so it should be barring any horror a no brainer to sign off a wet if its close to the proof which is the printers domain. So we need to profile to the proof, and we need to do it in CYMK, becuase if an RGB file gets the profile nuked we are really in trouble with out of gamut colours not being represented + its usually better to work RGB colours to suit a CYMK space rather than letting them just be converted.

Anyway, back to my boat and the drive to the distant horizon of good CYMK representation.

1) we have some lacies comming with a blue eye, so i can assume RGB and gamma will be spot on ok?

2) we only send out files in CYMK, i dont want to go down the road of RGB with cymk profiles.

3) we know a printer who runs a "trueflow" system, and they say their hiedlebergs match it well (taking an average wieght and coat). But we dont think a PS profile is available....so what do we do with profiles?

4) they have a useful colour swatch book which comes off their press, so we can at least know what colours we should be getting...

so, with a CYMK file can someone tell me how to calibrate a profile to their setup, step by step idiot mode after a calibrated monitor from an existing CYMK file? Do we send profile free images to them or do we try and get one that looks close, send a profile image off and then tweak that. How do we tweak existing profiles? I also assume once we have this golden fleece of a profile and calibrated monitors all our machines can use this yes?

If these questions are in the manual then just let me know, but I often get confused by what is what in the manual because its program orientated rather than situation.

Thanks for the patience...think of it this way, one less problematic artist (in theory)

[Mike_O...@adobeforums.com]

First off, does there proofs, their own internal proofs match their presses?

If not, pray for your job.

If yes, than profile that proofing device by sending Andrew Rodney an e-mail with a check to profile their device for them and you.

<http://www.digitaldog.net>

Take your RGB files and convert it to that CMYK profile.

Take the Lacie software and hardware monitor bundle and do the step by step process of profiling the monitor. Load that profile into color sync. as you active system profile.

Set your CMYK working space to the profile Andrew created for you.

Assign Adobe RGB or sRGB or Colormatch RGB or any capture RGB color space and convert the file to CMYK.

Supply the printer with CMYK files with no ICC tags if they don't have a clue what to do with them.

..........done deal

[Rene_W...@adobeforums.com]

Hi Prog,

I have been super busy this week, and I don't think there's much I can add. What I will repeat is:

1) Talk to the people down the line
2) If they say they don't have a profile, get their colour setup and convert the files to that CMYK.

And remember, someday, your (colour calibrated) prints will come... :P

[prog...@adobeforums.com]

thanks mike...yep their proofs match the press. your advice makes sense...there is light at the end of the tunnel, although im a little perturbed as to why the user cant fiddle with profiles like one can with say gamma. ho hum.

so basically thats the score of what we have to do everytime we need to hit a target image on screen wether its a proof or repro or whatever?

I assume i need to send them the proof and the untagged file for them to create a profile? I suspect i need to find someone this side of the pond in the uk then.

So once i have a profile for that particular system i dont need to send out embedded profiles to them anymore?

So am i right in assuming that profiles purely change the percieved colour of a file on a system rather than affecting the file's colour itself. If so why do people strip profiles to avoid colours wandering?

[Rick_M...@adobeforums.com]

This of course assumes that the offset house can create a match between the proof and the press when they take into account the stock the job is being printed on.

I would ask the printer how they account for different stock, i.e., printing on a bright blue-white stock vs. printing on a cream stock. If their proofer is a Kodak Approval, then they can run the proof on the actual stock. If this is the case, then you need to create a different profile for each different stock you might be using.

If, however, they are using some sort of inkjet/RIP proofing system, the question is not so clearly answered. MO et al - What do you guys do in that case to account for different stocks?

[prog...@adobeforums.com]

Bear with me guys.

just to clarify...when i send out the CYMK file to the house that i want to make a proof i should not embed any profile and make sure they dont. Then i then send this "clean" file and the print off to get profiled?

[Ann_She...@adobeforums.com]

The Profile only works its magic at the actual moment that you convert your RGB file into CMYK. The profile that you use governs the way that the Black plate is generated; the way that your RGB colours are mapped to CMYK inks; and the amount of Dot Gain that you build into the file. That is all that it does.

If it is a properly made two-way profile it then maps those CMYK values back into an accurate replication (in RGB) which is what you are seeing on your monitor when you look at a "CMYK" file.

[John_...@adobeforums.com]

I should think that the presense or absense of a profile in this CMYK test file would make no difference, unless their workflow somehow performs CMYK>CMYK conversions in their print stream.

Otherwise, they more than likely just print whatever numbers are in your file.

Profiles give meaning to numbers, but without a good description of the color space you're headed for, sending a known set of numbers to see what color they produce would be key toward defining that color space and returning an accurate profile that will convert the RGB you see in a softproof into the correct CMYK numbers for that device/printer.

[g_ba...@adobeforums.com]

It makes sense to [me] to embed the CMYK profile for delivery (unless the downstream software has a known problem with embedded profiles (after it gets out of Photoshop)).

If they ignore the tagged, embedded CMYK profile/colorspace — so what (if they don't mess with it)?

But — if they have colormanagement not broken, they will know the source exactly AND they will be able to softproof it accurately on screen, AND they will be able to make any behind-the-scenes Conversions...

Without the CMYK tag, they will not be able to see the color accurately on a calibrated monitor, and they will not be able to make any accurate Conversions — unless — your file happened to be last saved in their WorkingCMYK.

[John_...@adobeforums.com]

Isn't the point of these test swatches used for profile making to, in essense, send mystery meat then see what flavor you get back?

[Mike_O...@adobeforums.com]

so basically that's the score of what we have to do every time we need

to hit a target image on screen wether its a proof or repro or whatever?<

Output profile, + monitor profile = accurate previews. (granted that those profiles were generated correctly) and that both your monitor and output devices are stable enough.

I assume i need to send them the proof and the untagged file for them
to create a profile? I suspect i need to find someone this side of the
pond in the uk then.<

Get the correct profiling target, run the file out of your output device as if it were a live job, read in the target into a profile creation software, build the profile, stick it into Photoshop.

So once i have a profile for that particular system i don't need to send

out embedded profiles to them anymore?<

It depends upon your workflow and processes.

If it's just a simple process of separation to CMYK to a proofer that will not do additional cross generations to other color spaces, than yea you can chuck the tag. If you need someone at the other end of the food chain to do things like color correction and edits, whereas an accurate preview is needed, than don't toss the ICC tag. If you are getting other proofs besides a halftone contract proof, YOU NEED THE PROFILE to cross render to something like an Epson, or Fuji, etc.....

So am i right in assuming that profiles purely change the perceived colour

of a file on a system rather than affecting the file's colour itself.<

Profiles give meaning to the numbers in the file. Those numbers will reproduce differently between devices. That's why profiles were developed.

Correct previews are only part of the equation within a profile. The profile describes file behavior upon conversion. (Source color space)

If so why do people strip profiles to avoid colours wandering?<

because there are some systems that take that profile and convert it to some other color space and that's where the confusion lies. You can have a profile within an image without a conversion at the RIP. It just depends upon how the receiving applications and processors deal with that information.

To do a good separation and than chuck the tag is one workflow. The color geeks don't like this ideology, and I don't either, but it stops the brain farts down the line if someone who is profile illiterate.

Research your service providers to find out what you should really do. I con only give you so much rope unless you are my client.

just to clarify...when i send out the CYMK file to the house that i want

to make a proof i should not embed any profile and make sure they don't.<

You have to speak to them and find out what they know or don't know about the subject. Don't come back until you get that information.

;o)

It's up to you to find out the stuff on there end, and I'll guide you the rest of the way.

[prog...@adobeforums.com]

just a last couple of questions

1) if we just get some generic file in with another profile should we convert or strip it? ( i assume "ignore" strips it...?)

2) if we get in a file with another profile but we need to maintain the clients precise colour breakdown ie c50y21m32m5 do we convert or maintain it if we wish to put it in one of our files and work around it but still run proofs to our profile?

thanks again mike...and everyone else...im going to nip outside...i may be sometime ;)

case of beer at the tent door ;)

[Peter...@adobeforums.com]

"...printer profiles I have are not accurate when I calibrate my monitor with adobe gamma SOOOO I have had to twink them!! "

Bob, this may be the key to part of your soft proofing problems. Adobe Gamma is notoriously inconsistent because it relies on the the trained or not so trained eye of the user to make key decisions about white point, black point, gamma etc, often resulting in a far less than perfect calibration. There is no reason to not have a good hardware calibration procedure in place in any kind of professional environment where people are responsible for color, and taking that a step further, the few hundred dollars extra spent on something like an Artisan quickly pays for itself in a few proofs.

You've been bitching about shitty output profiles, but haven't really given it a good shot yet. You have a questionable at best monitor calibration and profile and no real accurate output profile, so I'm not surprised in the least at your results. I can pretty much guarantee that if you got yourself an Artisan and a good custom profile, you will be amazed at how well it works. I KNOW how well it works. If Andrew can't make the profile for you, I certainly can.

[John_...@adobeforums.com]

2) if we get in a file with another profile but we need to maintain the
clients precise colour breakdown ie c50y21m32m5 do we convert or maintain
it if we wish to put it in one of our files and work around it but still
run proofs to our profile?

That precise color breakdown, which I'm guessing is basically a grayish blue (that is 5% black yes?), is not all that precise. It's just CMYK numbers.

If the embedded profile defines a color space that is different enough from your own destination space, then converting is the logical answer, as it should in theory maintain the author's intent.

However, in the printing biz, you get into a whole realm of liability when you change a client's file. In one sense you aren't changing anything, but in another you are.

The real point is, if you are being sent CMYK files separated to some left-field color space, somebody somewhere don't have a clue what they are doing. Color management should end at the RGB>CMYK conversion.

[Mike_O...@adobeforums.com]

1) if we just get some generic file in with another profile should we
convert or strip it? ( i assume "ignore" strips it...?)<

It depends on what you are trying to do!

Look,

Contract proofs are as follows.

You give me a file, you separated it with whatever profile or color table.

I have a "contract proofer" A halftone dot proofer. A contract proofer is "supposed" to be set up to SWOP standards.

Based upon my calibration, I give you back a proof whereas the file has not been converted from what you have given me. IT'S what can be called a dot for dot proof. (which by the way is a bad saying, but for argument sake, what you give me is what I give you back based upon the calibration of my device. Color managements only roll here is upstream in your hands when the RGB to CMYK conversion was done.

2) if we get in a file with another profile but we need to maintain the clients precise colour breakdown ie c50y21m32m5 do we convert or
maintain it if we wish to put it in one of our files and work around it but still run proofs to our profile?

thanks again mike...and everyone else...im going to nip outside...i may be sometime

case of beer at the tent door

[Mike_O...@adobeforums.com]

Mr. Progress.

Stop beating yourself up and just do this on your next job.

Create a single color black swatch. 100% black after the RGB to CMYK conversion.

If your proof comes back and that swatch is 4 color, you know the monkey on the other end hosed your file...

[prog...@adobeforums.com]

:) gonna go hose armed monkey hunting :D

cheers :D

[bob_gr...@adobeforums.com]

Peter

your comments are a bit below the belt old boy

"You've been bitching about shitty output profiles, but haven't really given it a good shot yet.

I'm not "bitching" about the profiles I'm saying Adobe can't supply the world with 4 photoshop profiles and expect the world to all be able to use them so when PS 5 came out we had to find away around it and Knoll Gamma did that. Back then profile making facilities were only just emerging!!! This thread has got abit of track if you remember my original request.

I am not colour management illiterate. I understand the theory
I'm just using the tools I have because most print houses are owned by Ebonezer Scrooge's.

"You have a questionable at best monitor calibration and profile"

That is your opinion Peter IT WORKS just like it did before PS5 and CM were we all * ankers before PS5 no we used the info box and knoll gamma that's why he made it!!! and if that upsets you sorry but don't ridicule something you haven't seen.

"and no real accurate output profile, "

Very few do Peter Unless your profiling from a 1200 colour patch like Matchprint and very few have that ability. Why don't you send me one of yours and I'll assess your profile making ability .

"so I'm not surprised in the least at your results."

This is the one that got to me what results have I mentioned?
I have a very accurate monitor setup (it has taken many hours granted) but it is accurate and I don't ever rescan.

I'm sorry if I see no need to follow Adobes ideas when I have the skills to know what colour will print if I see 25y 15m 8c I just need to see picture contrast no one has that skill, for the rest of the world who is not trained to understand dot percentages these are the people who need Colour Management so help them don't attack me or the trade.

The truth is by the time PS3 came out people were really talking about PS and its great ability to let every Tom Dick and Harry become his own boss the trouble was everyone was complaining to Adobe that when they sent their job to print it looked like Doogy Doo so the boys @ PS decided the world needed a standard and they would introduce it and tell everyone
they needed a profile from their printer throwing the ball back in the printers court and taking the heat off of Adobe so to speak.

Help the people who need it I don't and you missed the gist of my thread.

Bob

[John_...@adobeforums.com]

Just for the record, I use Knoll Gamma as well, and have a pretty good monitor to proof agreement.

-Bob Cratchett

[Mike_O...@adobeforums.com]

Under PS versions 1-3 you created a great little piece of software called
Knoll Gamma. It was brilliant for fine tuning a monitor to a proof before
you and the lads decided to make everyone adopt colour management.<

For the record, everyone has been using color mgmt. since the creation of the application, well, at least when Tom made it color. Users are mistaken when they say they don't use color mgmt.

I still use it today under 9.2.2 Mac software since most profiles only
get you just in the ball park and adobe gamma setup is not accurate for
press paper whites in my opinion and this is where Knoll Gamma comes in
great.<

That's because you haven't read up on the latest technology to do it the correct way. Knoll Gamma was a manual way to address the lack of control that was ingrained at the time of the development of those older applications. (A work around) The current build of Photoshop gives you an extreme amount of flexibility that was poorly missing in those old versions. Knoll Gamma was the bridge between the problems. THAT'S why it was developed back then and that's why it will never be updated for the particular use it was designed for.

So forget about Tom re-releasing a bridgeable applet utility for something that has been replaced with a higher level of accuracy.

So my question is any chance you could make it available for system x
on the mac Squire. Im sure there are many pre press guys out there who
would agree with me.<

Not a chance in hell dude and you can forget about this prepress guy wanting to go back to monitor adjustment via Knoll Gamma!

May monkies rain down on you!

;o)

[Rene_W...@adobeforums.com]

That is your opinion Peter IT WORKS just like it did before PS5 and CM
were we all * ankers before PS5 no we used the info box and knoll gamma
that's why he made it!!! and if that upsets you sorry but don't ridicule
something you haven't seen.

A T-square, triangles and ruling pens also worked, so why don't you go back to that?

Sure it worked way back in the old analog days, cause that's all we had, but these days we have better tools. Yes, these tools need us to get more training to use them that triangles and ruling pen, but they are also more powerful.

The same applies to colour management.

You should always try to use the best of the tech available to you and for precise colour reproduction ICC profiles is it.

[Peter...@adobeforums.com]

"That is your opinion Peter IT WORKS just like it did before PS5 and CM were we all * ankers before PS5 no we used the info box and knoll gamma that's why he made it!!! and if that upsets you sorry but don't ridicule something you haven't seen. "

It's not just my opinion, it's the opinion of virtually anyone who's taken the time to implement. And yes, I have used Knoll Gamma, but not since I got my first PressView back about the time PS 4 came out. Even then, you could have a calibrated and profiled monitor and load either custom tables or even ColorSync profiles to make your conversions and provide soft proofing. What worked best then was to calibrate and then load the Lino profiles, which provided great soft proofing for most of what was available then. Now there are even better options that are easily within reach of most professionals.

I'm not ridiculing anyone. I have seen and used KG and gave it up when something better came out, but it seems you are the one who haven't given the newest tools a try.

[bob_gr...@adobeforums.com]

Mike,

Your comments to date have been sound but why do you keep bringing up what I have already agreed with Chris Cox on way back at #13 and again at #19????? all I have now done is defend my use of it which has annoyed you ,

Maybe because you don't work in a CMYK environment full time or maybe like Chris you used to when Rubylith was the norm. I don't know. For myself
I have been drum scanning for over 25 years. I have worked in web offset and sheetfed print company's and pre press bureaus over the years. I have worked on Dianippon, Hell and Crosfield scanners currently we have 2 crosfield scanners and Fujis Lanovia flatbed and Heidleburg * colour presses with an Indigo digital machine. We scan RGB and CMYK working a 16 hour days 5 days a week, we turn over thousands of CMYK scans a month and a knockback is rare. We run test proofs of the plotters and Indigo weekly to check for drift. We know our job and we do it very well and our method IS colour management. Not the type you want but it WORKS .

KNOLL GAMMA works with a twinked standard run of the mill Photoshop Profile on all our monitors. It WORKS very well probably beyond Toms and your understanding of it's original use, especially when you want to mimmick cast papers whites & newsprint in the white and shadow end and dot gain.

If you can't believe it works then like I said you are not a CMYK specialist but a colour theory geek.

I accept knoll Gamma is DEAD. On the few occasions Tom has visited the forum he has shown to be a nice chap and for a man with his knowledge I was hoping he would answer my question himself (and then I wouldn't be here 96 threads later) and say yer " Thanks for the compliment I'll see what I can do" BUT he can't because Colour Management must become the norm so you don't really have the right to speak his thoughts ..

... "Not a chance in hell dude and you can forget about this prepress guy wanting to go back to monitor adjustment via Knoll Gamma!" Touchy words. Why do you want to speak for everybody?

The TRUTH is APPLE will kill Knoll Gamma because to move to system X
will cause that ....HENCE my request!!!!!! So sleep soundly tonight the Ghost of Xmas past will not return to haunt you and the ghost of Xmas future will visit me. I'm not afraid.

"May monkies rain down on you!

I'm sorry I do not understand this American?? comment...

" If it's along the lines of may the fleas of a thousand camels infest your armpits." Like wise sir.

FOR ANYONE joining this thread this late who is interested in getting accurate CMYK from their RGB file Mikes thread is spot on, remember...

"The proof of an accurate profile/monitor calibration is the Proof." Soooo

First off check , does their proofs, their own internal proofs match their presses?
If not, pray for your job. (He's a touchy fellow)

If yes, than profile that proofing device by sending Andrew Rodney an e-mail with a check (Cheque for those who speak the Queens English) to profile their device for them and you.
<http://www.digitaldog.net>

Take your RGB files and convert it to that CMYK profile.

Take the software and hardware monitor bundle and do the step by step process of profiling the monitor. Load that profile into color sync. as you active system profile.

Set your CMYK working space to the profile Andrew created for you.

Assign Adobe RGB or sRGB or Colourmatch RGB or any capture RGB color space and convert the file to CMYK.
Supply the printer with CMYK files with no ICC tags if they don't have a clue what to do with them.

In this example it's not necessary to tag the file because your profile is matching their proof system. If you are sending the file without a proof and no idea of the press it is going to or have a profile to convert to CMYK , my advise is embed the file your stuffed really because it could be a web or sheetfed press big differences in dot gain so try and deal locally. sorry Mike my 3 cents.

99% of the files I deal with that have not turned off embedding and when I open them have the good old PS standard 20% swop coated profile with GCR !!!! that's when I hose it. Because I know this person is not ICC savy. Why because GCR printing is only used on newspaper web presses over here.
So another big point/problem ASK if they prefer GCR or UCR

All good stuff when it works but then you can read thread #44 take it away Mike.

Bob from Oz