Building color profiles with frame glass compensation?

[narikin]

Has anyone built a color profile to compensate for frame glass?

I was looking at samples of expensive anti reflective frame glass, and the 'best' one - a laminated 4mm glass has an obvious color shift. The regular 3mm one has a clear light green to it also. Regular glass does too, for that matter, so how would one go about building a paper profile that included this factor into its readings.

my iSis cannot accept the thickness of paper + 4mm glass (I believe) so another route would have to be taken.
is there an expert out there who'd care to suggest a way?

thanks.

[John Nollendorfs]

If you use the Spyder Print system, you can fine tune your profiles to your heart's  content. It has sliders for all kinds of compensation possibilities. But I have not tried to compensate  for glass like you need. Maybe pose this question on the DataColor bulletin board on Yahoo groups?

[narikin]

Thanks, but I'm not looking to tweak a profile, I'm wanting to build it from scratch through the specific frame glass I'm using.

this also has the advantage that any UV/OBA reduction that the glass causes will also be accounted for in the profile.

[Pat Herold]

Oh my, that's a tough one.
It is certainly possible to measure media of that thickness on a table unit like a Barbieri or a SpectroScan, but having 4 mm of glass on top of the target means that the illumination light will be defused through that layer of glass and the resulting measurement will be too dark.

All I can think of is that there is some talk around the industry these days about a method of reworking a single profile to work on different papers by recalculating the profile with a different paper white.  I wish I knew more about it. This might be used in your case to take a non-glass profile and a spectral measurement of your white patch through the glass, and this could be used to build a new profile.  Even so, it might only provide compensation for the lighter colors, I'm not sure.

[narikin]

Oh my, that's a tough one.
It is certainly possible to measure media of that thickness on a table unit like a Barbieri or a SpectroScan, but having 4 mm of glass on top of the target means that the illumination light will be defused through that layer of glass and the resulting measurement will be too dark.

All I can think of is that there is some talk around the industry these days about a method of reworking a single profile to work on different papers by recalculating the profile with a different paper white.  I wish I knew more about it. This might be used in your case to take a non-glass profile and a spectral measurement of your white patch through the glass, and this could be used to build a new profile.  Even so, it might only provide compensation for the lighter colors, I'm not sure.

Thanks, its the lighter colors that concern me. they are most obviously affected by the light green-yellow tint of the glass.

so a iO table would not work through 3 or 4mm glass? damn.

[Ernst Dinkla]

Water white glass is one thing to start with and then eyeball the difference framing makes to one print while a bare copy hangs next to it. Edit the profile on brightness. I do not think you can make a profile of a target with 2-4 mm glass + 2 mm air between the target and the spectrometer. The optical geometry of a normal spectrometer is not made for that task.


met vriendelijke groeten, Ernst
Shareware too:
330+ paper white spectral plots:
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm

[smilem]

You need to edit the paper white with PM profile editor.

[framah]

Just a thought here... is anyone ever going to see a print right next to the original?

And if it is NOT a print from an original painting, so what if it looks a bit different when under glass. No one but you would ever know that.

I feel that maybe you are over thinking a problem that might not be that big of a problem.

But, then that's just me.

[Ethan_Hansen]

A possibility - not guaranteed to work, but there is hope - is to measure light coming through the glass and use this as a custom illuminant. I would either use a 5000K tungsten light source (Sol-Source or slightly overdriven Solux) or wait for a clear day, measure the horizon light until it is close to 5000K, then measure through the glass.

[narikin]

Interesting idea. I might give that a try!  thanks.

I guess one flaw is the light goes through the frame glass, reflects back of the print and back again through the glass to you, the viewer. Custom illuminant is measured 1x through glass, transmission rather than reflection reading, etc. But still, better than nought.

I have been building up a set of custom illuminant profiles for galleries here in NYC, and find more adhere closer to the 'Illuminant A' canned profile than D50 or D55.

[Eric Myrvaagnes]

I guess one flaw is the light goes through the frame glass, reflects back of the print and back again through the glass to you, the viewer. Custom illuminant is measured 1x through glass, transmission rather than reflection reading, etc. But still, better than nought.

Measure illuminant through two thicknesses of the glass. That should come close, even though neither is a reflection reading.

[Ethan_Hansen]

My guess is that capturing the illuminant more correctly should compensate for most of the effect. It wouldn't hurt to measure through two thicknesses of glass. My presumption was that the illuminant would be akin to D50. If not, by all means measure under the actual lighting conditions. Unless fluorescent is being used for viewing the actual prints, however, you need to ensure a full-spectrum source is used to build the light profile.

We have made a few profiles for weird lighting conditions. Measuring the actual illuminant was a necessary first step. Editing the profile tables to compensate for color distortion in the light reflected from the print may still be necessary.

[alain]

Hi

A similar question :

I do spray most of my prints with a protective spray  (Talens 680 protective spray). 
Should I make profiles after the spraying?

[Ernst Dinkla]

Hi

A similar question :

I do spray most of my prints with a protective spray  (Talens 680 protective spray). 
Should I make profiles after the spraying?

On matte papers?  I use it too sometimes but the main reason to use it is that it, to my eyes, does not change the image on matte papers. Not the velvet blacks on matte papers, not the colors. There is no build up on the surface. Which is for me a reason not to make other profiles. With matte canvas and protecting layers of satin and gloss Lascaux varnish there are good reasons to make profiles with varnish applied on the target.


On measuring the light through two glass sheets: seems a good way to use the changed illuminant spectrum in profile creation but could one also include a compensation for the light reduction based on a measurement?

The two glass sheets method should work for framed non-OBA papers, the OBA paper framed will have more complicated filtering.


met vriendelijke groeten, Ernst

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

[alain]

On matte papers?  I use it too sometimes but the main reason to use it is that it, to my eyes, does not change the image on matte papers. Not the velvet blacks on matte papers, not the colors. There is no build up on the surface. Which is for me a reason not to make other profiles. With matte canvas and protecting layers of satin and gloss Lascaux varnish there are good reasons to make profiles with varnish applied on the target.


On measuring the light through two glass sheets: seems a good way to use the changed illuminant spectrum in profile creation but could one also include a compensation for the light reduction based on a measurement?

The two glass sheets method should work for framed non-OBA papers, the OBA paper framed will have more complicated filtering.


met vriendelijke groeten, Ernst

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

Hi Ernst

I do this on a baryta paper : harman gloss AB.   It gives back the look when I see when the paper is just printed.  Deeper blacks and "better" colors" and more scratch resistant.
I don't know if the colors itself change.

[Ernst Dinkla]

Hi Ernst

I do this on a baryta paper : harman gloss AB.   It gives back the look when I see when the paper is just printed.  Deeper blacks and "better" colors" and more scratch resistant.
I don't know if the colors itself change.

In that case you will see a higher Dmax, a wider dynamic range and most likely a wider gamut. You should create a profile from a varnished target.

met vriendelijke groeten, Ernst
Shareware too:
330+ paper white spectral plots:
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm

[JohnTodd]

I was thinking about treating this as an LCC problem - shooting an LCC frame through both an LCC diffuser and a sheet of the target glass - but I'm out of my depth in this company. Would that answer the single/double-thickness issue?

[elolaugesen]

Hi:  Having organised a few exhibitions in various environments I think that you are looking for something that is a dream.
If the art work is 6ft any person looking at the art must be quite a distance from the art work or else they will not see what they are looking at. 
At that point all the lighting and the environment come into play and I challenge any human eyes to see the real colours or even the detail then challenge what you have created.

Having said that glass makes a huge difference and the main artist I work with (my wife and partner) will only use anti reflective/waterwhite/museum type glass.

cheers..    elo

[Dinarius]

If the glass creates the degree of shift you claim it does, and if you compensate fir that shift, won't the viewer of the reproduced work (I'm assuming you're going to all this trouble for repro) be seeing something different to what's hanging on the wall? If so, won't the exercise be self defeating?

I shoot a lot of artwork, and my clients (mostly galleries) want what the prospective buyer sees documented, not what was in the artist's studio prior to framing.

Just my tuppence worth.

D.