Colour accuracy

A while ago I did a little project - it seemed straightforward but things never are as simple as they seem, and several EPZ members helped me. My wife's dad was a prisoner of war during the second world war, and when her mother passed away a few years ago (like so many, her dad didn't really get to see an old age), some papers came to light - like a newspaper, written in Japanese. I had it translated, which in itself wasn't straigtforward because the style of writing is now considered somewhat archaic, and I managed to create a mockup of the original, but with the text of the articles replaced with the translation. I owe a massive thanks to ViVla for his help. When I had it printed, the colour was off. I had realised it wouldn't be a perfect match, but the difference was greater than I'd appreciated. I used a professional lab and have no complaints about their service, but unfortunately they describe themselves as a self-service - er - service, and don't offer any advice. The printing wasn't cheap and I don't want to waste money on a trial-and-error scheme. Is there a way I can photograph the original and print side by side and somehow establish what alteration is needed to bring the mockup closer to the original so that I can apply that adjustment and print? I'd settle for a decent-to-good match, it needn't be perfect. I've tried a few local firms but none seem able to help, they're all geared towards graphic design and commercial printing. The thing with the original is that it's old, and prnited on cheap, thin, unbleached paper, and it's flimsy so I don't want to handle it any more than necessary. I scanned it at our local library and the light shone through it, and it seems to look different under every type of light (yes, I know things actually do look different, but noticeably so). Any help would be appreciated, folks. It means a lot because this, together with a couple of maps that came to light, are the only things of her dad's that my wife has, or has ever had. We'd like to get them framed and up on the wall.

A while ago I did a little project - it seemed straightforward but things never are as simple as they seem, and several EPZ members helped me.

My wife's dad was a prisoner of war during the second world war, and when her mother passed away a few years ago (like so many, her dad didn't really get to see an old age), some papers came to light - like a newspaper, written in Japanese.

I had it translated, which in itself wasn't straigtforward because the style of writing is now considered somewhat archaic, and I managed to create a mockup of the original, but with the text of the articles replaced with the translation. I owe a massive thanks to ViVla for his help.

When I had it printed, the colour was off. I had realised it wouldn't be a perfect match, but the difference was greater than I'd appreciated. I used a professional lab and have no complaints about their service, but unfortunately they describe themselves as a self-service - er - service, and don't offer any advice.

The printing wasn't cheap and I don't want to waste money on a trial-and-error scheme. Is there a way I can photograph the original and print side by side and somehow establish what alteration is needed to bring the mockup closer to the original so that I can apply that adjustment and print? I'd settle for a decent-to-good match, it needn't be perfect. I've tried a few local firms but none seem able to help, they're all geared towards graphic design and commercial printing.

The thing with the original is that it's old, and prnited on cheap, thin, unbleached paper, and it's flimsy so I don't want to handle it any more than necessary. I scanned it at our local library and the light shone through it, and it seems to look different under every type of light (yes, I know things actually do look different, but noticeably so).

Any help would be appreciated, folks. It means a lot because this, together with a couple of maps that came to light, are the only things of her dad's that my wife has, or has ever had. We'd like to get them framed and up on the wall.

Is the replica a digital file? Is your monitor calibrated? It should be possible in PhotoShop to use the color picker to sample the colour of the original. Although I wasn't directly involved, before I retired we did a lot of facsimile work of historic books, everything was photographed using colour reference charts, however that is slightly different to your problem. I would say you need to be viewing your replica side by side with the original, the viewing conditions are going to be problematical, ideally you should be viewing the original in the same conditions as your replica will be viewed, if that is not possible then as close to daylight as possible. (I use a couple of Ottlights) It is also worth noting that different methods of printing and the surface of the paper can make differences to the colour as well so it may be worth using a printer that issues ICC profiles and soft proofing with that.

Is the replica a digital file? Is your monitor calibrated?

It should be possible in PhotoShop to use the color picker to sample the colour of the original. Although I wasn't directly involved, before I retired we did a lot of facsimile work of historic books, everything was photographed using colour reference charts, however that is slightly different to your problem.

I would say you need to be viewing your replica side by side with the original, the viewing conditions are going to be problematical, ideally you should be viewing the original in the same conditions as your replica will be viewed, if that is not possible then as close to daylight as possible. (I use a couple of Ottlights)

It is also worth noting that different methods of printing and the surface of the paper can make differences to the colour as well so it may be worth using a printer that issues ICC profiles and soft proofing with that.

Thanks Phil. It is indeed a digital file, and no, my monitor isn't calibrated. It's fairly new, though, and I don't think it's wildly off. You mention ICC profiles, and the printing service I used provides those; I didn't find they altered the on-screen image greatly, and I found the prints a good match for the screen. The problem, I find, is viewing the original next to the screen. It may be because the paper is dark/ fragile/ thin; I'm just not sure. I'm aware that my setup is basic; I've tried closing the blind and having diffuse daylight, closing the curtains, etc. Anyway. My idea was that I'd photograph each original alongside its corresponding print together by windowlight (not today because it's miserable outside!) so that conditions are identical and artificial lighting/ flash fall-off, colour cast from interior decor, etc., are all minimised. Then crop the photo to show the original, and copy & paste the replica into the digital file it came from so that the just-taken photo is the top layer, and the only one visible. Side by side, adjust the replica (all layers) to get as good a match as possible and when finished, discard the top layer. Hopefully then the underlying image, the one from which the replica print was made, will be a half-decent colour match.. Do you think this is feasible.. maybe..?

Thanks Phil.

It is indeed a digital file, and no, my monitor isn't calibrated. It's fairly new, though, and I don't think it's wildly off. You mention ICC profiles, and the printing service I used provides those; I didn't find they altered the on-screen image greatly, and I found the prints a good match for the screen. The problem, I find, is viewing the original next to the screen. It may be because the paper is dark/ fragile/ thin; I'm just not sure. I'm aware that my setup is basic; I've tried closing the blind and having diffuse daylight, closing the curtains, etc.

Anyway. My idea was that I'd photograph each original alongside its corresponding print together by windowlight (not today because it's miserable outside!) so that conditions are identical and artificial lighting/ flash fall-off, colour cast from interior decor, etc., are all minimised. Then crop the photo to show the original, and copy & paste the replica into the digital file it came from so that the just-taken photo is the top layer, and the only one visible. Side by side, adjust the replica (all layers) to get as good a match as possible and when finished, discard the top layer. Hopefully then the underlying image, the one from which the replica print was made, will be a half-decent colour match..

Do you think this is feasible.. maybe..?

May be worth a try, although I am not sure it would work, from what I can gather you are trying to match your replica to the original, the only way I can think of is to compare them side by side under consistent lighting. Overcast is good as the light is more diffuse... I think there could be lot of trial and error. I don't know who is producing the prints but to save costs you could just get prints from a small part of the file, duplicate the file, make a 6 x 4 crop and order a 6 x 4 print (you could produce several different prints with different adjustments, just note the adjustments down on the file so that they are visible on the prints)

May be worth a try, although I am not sure it would work, from what I can gather you are trying to match your replica to the original, the only way I can think of is to compare them side by side under consistent lighting. Overcast is good as the light is more diffuse... I think there could be lot of trial and error.
I don't know who is producing the prints but to save costs you could just get prints from a small part of the file, duplicate the file, make a 6 x 4 crop and order a 6 x 4 print (you could produce several different prints with different adjustments, just note the adjustments down on the file so that they are visible on the prints)

Thanks Phil. I think last time the mistake was on my part, although further investigation has revealed a noticeable difference in brightness between my monitor and the prints. Using a reference shot of each print alongside its corresponding original, I think I've reverse-engineered the difference and applied it to the files I sent for printing last time, so I'm hopeful that they will come out right this time. I'm going to order some test prints in the next few days/ weeks, so, fingers crossed. Cheers!

Thanks Phil. I think last time the mistake was on my part, although further investigation has revealed a noticeable difference in brightness between my monitor and the prints.

Using a reference shot of each print alongside its corresponding original, I think I've reverse-engineered the difference and applied it to the files I sent for printing last time, so I'm hopeful that they will come out right this time. I'm going to order some test prints in the next few days/ weeks, so, fingers crossed.

Cheers!

[quote] and no, my monitor isn't calibrated. It's fairly new, though, and I don't think it's wildly off.[/quote] Unless you have a monitor that comes with a certificate that is to ICC standards for photography use, most monitors are too bright and too blue out of the box. They are intended for gaming. Although it is fairly new it could be well off.

Quote: and no, my monitor isn't calibrated. It's fairly new, though, and I don't think it's wildly off.


Unless you have a monitor that comes with a certificate that is to ICC standards for photography use, most monitors are too bright and too blue out of the box. They are intended for gaming. Although it is fairly new it could be well off.

[quote].. Although it is fairly new it could be well off.[/quote] Ah. Right. There's a camera club, not quite local to me but close enough, of which a work colleague is a member. He once mentioned them having a calibration device of some sort which members can borrow - might do that.

Quote:.. Although it is fairly new it could be well off.


Ah. Right. There's a camera club, not quite local to me but close enough, of which a work colleague is a member. He once mentioned them having a calibration device of some sort which members can borrow - might do that.

[quote]although further investigation has revealed a noticeable difference in brightness between my monitor and the prints.[/quote] As Railcam mentions most of the time monitors off the shelf are too bright etc. i.e. they are not set up for photography. You will be surprised at the amount you need to reduce the brightness. There is no set figure as the brightness will vary according to the ambient light and viewing conditions. One way is to hold a sheet of white paper next to the monitor and adjusting the brightness to match, it can often be very hit and miss but worth it in the end. Borrowing a calibration device would be a good idea but you do need to calibrate and profile on a regular basis as monitors will gradually change over time (not so much with LCD panels as with the old CRT monitors).

Quote:although further investigation has revealed a noticeable difference in brightness between my monitor and the prints.

As Railcam mentions most of the time monitors off the shelf are too bright etc. i.e. they are not set up for photography. You will be surprised at the amount you need to reduce the brightness. There is no set figure as the brightness will vary according to the ambient light and viewing conditions. One way is to hold a sheet of white paper next to the monitor and adjusting the brightness to match, it can often be very hit and miss but worth it in the end. Borrowing a calibration device would be a good idea but you do need to calibrate and profile on a regular basis as monitors will gradually change over time (not so much with LCD panels as with the old CRT monitors).

Okay, thanks again everyone. I ran the built-in Windows calibration wizard, which was vaguely familiar because I'd done it before.. now my monitor is a good stop, stop-and-a-half dimmer and is less blue. Next, enough of this guesswork! Looking at the printer's website - [link=https://www.theprintspace.co.uk/]The Print Space[/link] they mention calibration and the equipment they use and the standards they employ; notably Luminance: 120 ca/m2; gamma: 2.2; colour: 6500K; black level: min where possible; using the X-rite i1. Not knowing the first thing about spectrophotometers or any other monitor calibration whizz-bangs, can anyone just tell me what I need to do, please? I read somewhere that the Spyder 3 Pro not only does the necessary calibration, but also monitors ambient lighting conditions and constantly adjusts the display accordingly, which to me sounds awesome - but is it? Or would I be better off going for X-Rite? I don't print often, and don't own my own printer.

Okay, thanks again everyone. I ran the built-in Windows calibration wizard, which was vaguely familiar because I'd done it before.. now my monitor is a good stop, stop-and-a-half dimmer and is less blue.

Next, enough of this guesswork! Looking at the printer's website - The Print Space they mention calibration and the equipment they use and the standards they employ; notably Luminance: 120 ca/m2; gamma: 2.2; colour: 6500K; black level: min where possible; using the X-rite i1.

Not knowing the first thing about spectrophotometers or any other monitor calibration whizz-bangs, can anyone just tell me what I need to do, please? I read somewhere that the Spyder 3 Pro not only does the necessary calibration, but also monitors ambient lighting conditions and constantly adjusts the display accordingly, which to me sounds awesome - but is it? Or would I be better off going for X-Rite? I don't print often, and don't own my own printer.

[quote]constantly adjusts the display accordingly,[/quote] Not a good thing, although not easy to achieve, particularly in a domestic situation you need things to be constant including the ambient illumination. You should ignore any recommendations for the luminance (brightness) of the monitor, that all depends on the ambient light where the monitor is positioned and the viewing conditions for the print, 120cd/m2 is just a general purpose figure, for instance, I work with my monitors at 80cd/m2. Even if you don't print very often having a calibrated and profiled monitor has advantages, depending on budget have a look at the X-Rite i1 display (a quick edit - it appears that both X-Rite and Datacolor no longer appear to supply sub £100 gear)

Quote:constantly adjusts the display accordingly,

Not a good thing, although not easy to achieve, particularly in a domestic situation you need things to be constant including the ambient illumination.

You should ignore any recommendations for the luminance (brightness) of the monitor, that all depends on the ambient light where the monitor is positioned and the viewing conditions for the print, 120cd/m2 is just a general purpose figure, for instance, I work with my monitors at 80cd/m2.

Even if you don't print very often having a calibrated and profiled monitor has advantages, depending on budget have a look at the X-Rite i1 display (a quick edit - it appears that both X-Rite and Datacolor no longer appear to supply sub £100 gear)

I use an old Xrite Colormunki Photo to calibrate my monitor and produce printer profiles. Prints match my screen (allowing for the lighting differences - monitor projected light, prints reflected light). The current version of the Colormunki is the Xrite i1 Studio. Not cheap but should solve all your colour problems. As Philh04 says, 120cd/m2 is a brightnes guide only. I also have my monitor at 80cd/m2 for a perfect brightness match with my prints. Printer profiles produced by paper manufacturers are good, producing your own bespoke ones for a particular batch of paper will give spot on results. As a minimum you should calibrate your monitor. The Colormunki Display is now the i1 Display. I like the videos produced by Joe Brady, he is a working photographer and talks a lot of sense. Try watching: (Monitor Calibration) https://www.youtube.com/watch?v=Wsq9kIW_dLg (Printer Profiling) https://www.youtube.com/watch?v=KsPnM7-LM_Y I hope this is of help to you. Keith

I use an old Xrite Colormunki Photo to calibrate my monitor and produce printer profiles. Prints match my screen (allowing for the lighting differences - monitor projected light, prints reflected light).

The current version of the Colormunki is the Xrite i1 Studio. Not cheap but should solve all your colour problems.

As Philh04 says, 120cd/m2 is a brightnes guide only. I also have my monitor at 80cd/m2 for a perfect brightness match with my prints. Printer profiles produced by paper manufacturers are good, producing your own bespoke ones for a particular batch of paper will give spot on results.

As a minimum you should calibrate your monitor. The Colormunki Display is now the i1 Display.

I like the videos produced by Joe Brady, he is a working photographer and talks a lot of sense. Try watching:
(Monitor Calibration) https://www.youtube.com/watch?v=Wsq9kIW_dLg
(Printer Profiling) https://www.youtube.com/watch?v=KsPnM7-LM_Y

I hope this is of help to you.
Keith

Thanks again. I just had a quick look at my reverse-engineered adjusted files, and I think I was homing in on a good result, but I'm hoping to hold out for a calibrator. Failing that I'll order test prints using these files and a series of +/- adjustments and pick the best of the bunch to go ahead with full size prints. I've been eyeing up Spyder and X-rite i1 models, but I may be able to borrow one (don't know which) from a local club. SWMBO frowned at the prospect of buying one, so if they come through for me then that'll be a life saver.

Thanks again. I just had a quick look at my reverse-engineered adjusted files, and I think I was homing in on a good result, but I'm hoping to hold out for a calibrator. Failing that I'll order test prints using these files and a series of +/- adjustments and pick the best of the bunch to go ahead with full size prints.

I've been eyeing up Spyder and X-rite i1 models, but I may be able to borrow one (don't know which) from a local club. SWMBO frowned at the prospect of buying one, so if they come through for me then that'll be a life saver.