Capture One, Argyll, CM Photo...and display vs. print

[RogerB]

Thank you for doing this...it is great to see that all the tones within the image fit within your display profile. I suspected this, and have done a gamut check on the image, but your plot shows this conclusively. Thanks again!

PP

Just occurred to me that your original must be in ProPhotoRGB and I had assumed sRGB. I say this because if I assign sRGB or AdobeRGB to your original the skin tones are very yellow, whereas if I assign ProPhotoRGB they look normal i.e redder. Doesn't make an enormous difference in the colours in numerical terms so the gamut plot is not very different; but, it does make the image look right. Your "proof" image on the other hand looks very saturated if I assign PPRGB and redder than the original after assigning PPRGB. Is there something odd about your colour spaces in different applications?

 

[ppmax]

Argh...You are correct. I exported the original image as TIF in the ProPhoto space, but then I converted to JPG and forgot to convert to sRGB. Doh!

The "proof" image was created from the orginal TIF/ProPhoto, which I then passed through tifficc to produce a soft proof. I also converted this to JPG and forgot to convert to sRGB before posting.

Suffice it to say that when I view the "proof" image it is a very good representation of what I see in the print...which has an abundance of yellow/tan in the skin tones. However, this may just be a coincidence due to lighting rather than any process related problem. I will hopefully be able to view this print under real daylight in about an hour.

thx
PP

 

[ppmax]

OK, having been blessed with some sun and natural light today I think I can safely say that my screen and print match...apparently the 5000K LED is pretty good for most tones but adds some amber/tan to the flesh tones. My $6 bulb is no substitute for a good light...but maybe it's good enough until I get a Solux or a booth

I've uploaded a "combined" version of the two images that I uploaded previously, now converted to sRGB. The process of producing a soft-proof with tifficc has also revealed a coincidence, whereby the manually generated soft-proof image (not in Capture One) closely matches the amber/tan cast introduced by the LED...as shown (rather, simulated) in the combined image below.

thanks again for all the help--
PP

 

[ppmax]

Out of curiosity, given a specific light/paper combination, is it possible to "flavor" a custom print profile for a specific lighting condition? In other words, how would I tune my print profile to the light emitted by the LED? Ideally, I'd want to dial up the amber/tan in my printer profile so that soft proofing in Capture One matches what will be my typical lighting scenario?

thx
PP

 

[RogerB]

Out of curiosity, given a specific light/paper combination, is it possible to "flavor" a custom print profile for a specific lighting condition? In other words, how would I tune my print profile to the light emitted by the LED? Ideally, I'd want to dial up the amber/tan in my printer profile so that soft proofing in Capture One matches what will be my typical lighting scenario?

thx
PP

You can do this with ArgylCMS. The -i parameter in colprof allows you to specify a custom illuminant (from actual measurement in your case) when calculating the profile. I've never tried it myself so can't say how effective it is. I prefer to avoid, wherever possible, illuminants with poor colour rendering and if I can't avoid them I just live with the consequences.

If you're looking for a different (better) viewing light you might like to consider the Soraa Vivid LEDs. Your 5000K "cooking" LED almost certainly doesn't emit any violet so doesn't excite the OBAs in your paper which will make it look a little bit more yellow. It's also rather deficient in the red which doesn't help skin tones. The Soraa Vivid LEDs emit violet and are much stronger in the red so their colour rendering is much better. Not as good as Solux, but they come in line-voltage GU10 versions so you have a very wide choice of inexpensive fittings, including desk lamps. Take a look at http://www.soraa.com/products/MR16-GU10 and have a read of some of their technical papers.

 

[mikling]

You can do this with ArgylCMS. The -i parameter in colprof allows you to specify a custom illuminant (from actual measurement in your case) when calculating the profile. I've never tried it myself so can't say how effective it is. I prefer to avoid, wherever possible, illuminants with poor colour rendering and if I can't avoid them I just live with the consequences.

If you're looking for a different (better) viewing light you might like to consider the Soraa Vivid LEDs. Your 5000K "cooking" LED almost certainly doesn't emit any violet so doesn't excite the OBAs in your paper which will make it look a little bit more yellow. It's also rather deficient in the red which doesn't help skin tones. The Soraa Vivid LEDs emit violet and are much stronger in the red so their colour rendering is much better. Not as good as Solux, but they come in line-voltage GU10 versions so you have a very wide choice of inexpensive fittings, including desk lamps. Take a look at http://www.soraa.com/products/MR16-GU10 and have a read of some of their technical papers.

The success of this will depend on how close the source is to the reference illuminant. Keep in mind as well, even if measured properly ( and you need a high end spectro to do this) if the source is very peaky, then it will create a resultant ICC that can be not very good when in a different light than the intended. If one goes to the trouble of meauring the actual light, then one should also measure the actual OBA influence of the paper itself as well. Worth it?..a lot more work in reality. As stated, it is better to just get a superior light source.

 

[ppmax]

If you're looking for a different (better) viewing light you might like to consider the Soraa Vivid LEDs. Your 5000K "cooking" LED almost certainly doesn't emit any violet so doesn't excite the OBAs in your paper which will make it look a little bit more yellow. It's also rather deficient in the red which doesn't help skin tones. The Soraa Vivid LEDs emit violet and are much stronger in the red so their colour rendering is much better. Not as good as Solux, but they come in line-voltage GU10 versions so you have a very wide choice of inexpensive fittings, including desk lamps. Take a look at http://www.soraa.com/products/MR16-GU10 and have a read of some of their technical papers.

Thanks for the referral--I'll check those out. All I want is a decent desk light for spot checking prints. Although I've sold a couple photos at an art show before, I don't think that will be a common occurrence!

FWIW the 5000K LED bulb I bought was a Philips 9W 5000K A19:
http://www.homedepot.com/p/Philips-...-A19-Dimmable-LED-Light-Bulb-455873/205818708

I couldn't find any real specs on it but thought I'd give it a whirl....

Worth it?..a lot more work in reality. As stated, it is better to just get a superior light source.

Indeed!

thanks again,
PP

 

[mikling]

The color temp is not the total spec either. The CRI is a "clue" not everything but LEDs today are typically "peaky" in the spectral content and leads to creating effects on certain colors depending on where the peak occurs.