Soft proofing Photoshop

[KTnovice]

I have a Benq sw2700pt monitor, I use the AdobeRGB monitor screen. that is calibrated with the Spyder5Pro and Palette master software. I calibrated using D65, L90 and gamma 2.2, 16bit LUT to edit for printing.

I edit in ProphotoRGB workspace, and so far have only prepared digital images for Photo club. The colours of these have been fine when projected or uploaded to the web. I now want to Print.

I have followed all the steps in the tutorial and have downloaded a paprer/printer profile from a commercial print Lab the most of my club use. When I proof the image there is a major tonal shift, and if I activate the OOG warning almost the whole image is OOG.
I also downloaded the Lab test image (sRGB profile), into the sRGB working space, and opened two copies side by side to compare. I proofed one version image (attached) all the magenta areas are considered OGG. This covers all 7 of the major colours. I am supposed to edit this version, to bring it close to the master copy, send the edited JPEG, sRGB to the Lab, get a photo back to then compare with my screen edit. Two hours later I am still no where with the editing, tried Curves, HSL individual colours, Bright/contrast......

I have also repeated this with 2 other similar test images, and different lab supplied ICC paper profiles similar results
Any help really would be appreciated.

[davidedric]

It's essential that either Photoshop or the printer are managing the colour, but not both. Have you set that?

Dave

[Shadowman]

Lightroom provides a good soft proofing tool to analyze out of gamut colors.

[James G]

In addition to Dave's question already posed....

Can you clarify the following:-

1) have you actually printed yet or are you still trying to resolve oog issues?

2) if you have printed, are the images too dark, too saturated etc?

3) I've also noted that you have Spyder 5. Pallete Master, as I understand it is a profiling tool that generates a profile on the monitor. Spyder creates a profile within your Operating System so are you effectively 'double' profiling?

4) You also say you have downloaded a paper/printer profile from the print lab, are you using paper supplied by them? (I would have expected you to download a profile from the paper supplier directly. eg from Epson, Canon, Canson etc)

Additionally, do you know what brightness value dyou use when calibrating the monitor with Spyder5. My experience is that the default/recommended 120cd/m2 is too bright for printing and results in disappointingly dark/dull prints. It becomes necessary to 'brighten' the image for print.
I usually calibrate my monitor for a value close to 100-105 cd/m2. I still find that I need to run test prints to determine optimal brightening factors for each of my favourite papers.

[Manfred M]

Have you converted the image you worked on from ProPhoto to sRGB before doing the comparison?

A major colour shift like that suggests a colour space issue.

[KTnovice]

In addition to Dave's question already posed....

Can you clarify the following:-

1) have you actually printed yet or are you still trying to resolve oog issues? "I had two prints back where there was a strong tonal shift in reds/greens and blue. Hence trying to resolve the OOG issues".

2) if you have printed, are the images too dark, too saturated etc? "Not overly dark autumn orange/red/browns over saturated. greens looked grey. sky blue hue shifted".
"had converted from ProPhoto RGB to sRGB for proofing, ready to save file as JPEG to upload".

3) I've also noted that you have Spyder 5. Pallete Master, as I understand it is a profiling tool that generates a profile on the monitor. Spyder creates a profile within your Operating System so are you effectively 'double' profiling? "Plug Spyder into monitor direct and do not use Spyder software, no alteration to OS, chosen correct profile for the display from Control Panel, no double profile conflict".

4) You also say you have downloaded a paper/printer profile from the print lab, are you using paper supplied by them? (I would have expected you to download a profile from the paper supplier directly. eg from Epson, Canon, Canson etc). "Yes the profile is for their printer and a specific paper they use http://dscolourlabs.co.uk/category/133-fuji-premium-large-format-photographic-prints".

Additionally, do you know what brightness value dyou use when calibrating the monitor with Spyder5. My experience is that the default/recommended 120cd/m2 is too bright for printing and results in disappointingly dark/dull prints. It becomes necessary to 'brighten' the image for print.
I usually calibrate my monitor for a value close to 100-105 cd/m2. I still find that I need to run test prints to determine optimal brightening factors for each of my favourite papers.

"I was using L 120cd/m2 for DPI images which was fine, have a second profile I can use for Print preparation using 90cd/m2"

Because of the colour issues I was proofing the sRGB profiled image (as instructed by lab), selecting the profile supplied, using perceptual, simulate paper. When I activated the OOG warning 7 of the colour swatches in the images I uploaded had gamut warning over them and the colour patches along the bottom of the test image supplied. (the screen shot did not capture the OOG warning colours).

I was not using the Print dialogue box so no colour mangement conflict, only creating JPEG to upload.
Thanks guys,
hope this has answered your points

[Manfred M]

" using perceptual, simulate paper.

I personally prefer using relative colormetric to perceptual as that rendering intent leaves all the colours other than the OOG (out of gamut) colours alone, giving you more "accurate" colours in the image. It forces those OOG colours to be in gamut.

What perceptual does is proportionally move all the colours that are OOG so all the colours are remapped, giving you the risk of some strange looking colours, especially in skin tones. This might be the reason you are seeing some strange colour casts. The other problem with perceptual is that every manufacturer uses different algorithms here as they try for a particular "look".

I tend to use relative colormetric about 99% of the time but every so often I find an image that looks better in perceptual.

[James G]

Like Manfred, I prefer relative colorimetric for rendering intent. I don't however have any experience of using a Print Lab. I print to an Epson SC P600 printer at home.
I process in ProPhoto RGB, but I convert to AdobeRGB not sRGB for print production. I only convert to sRGB for web presentation.
Not totally sure why the Printlab want the jpeg transmitted as sRGB, but.... you say you downloaded the test image into the sRGB working space. Did you get a profile mismatch warning when you first opened it in Photoshop and if so which option did you proceed with?
As Manfred suggested earlier the colour shift seems most likely to be caused by a colour space issue.

[KTnovice]

Hi,
the file supplied did not have an embedded profile so I assigned the sRGB profile.

I would like to thank all the people who have responded - and so quickly. I have spent over a week now trying to resolve this and have today decided to purchase the Epson P600 and some paper and work it out at home. (Last day of Photo shop so got a decent price).