Shooting chromes for artwork reproduction

[Uri A]

Hi people,

I have been offered a job shooting artwork for reproduction/archival purposes by a local gallery. I have offered them the cheapy digi option, or the 4x5 chrome & scan option (figuring that scanning chromes will result in a huge file and accurate colour rendition) I used to shoot this stuff on E100G waaaaay back in the 90's, but just checking in with the here & now:

(1) For chromes, what current emulsion would you use for purely color correct rendition under (decent studio calibrated) flash light, to capture maximum detail in all areas of the original? E100G is under the death warrant at B&H - would you shoot Provia 100? I haven't shot a chrome in 8 years, but back in my day it seemed pretty green & cool....
(2) back in the day, we used to make sure our 2 heads were flashing the same colour temp to make sure we weren't getting any gremlins in the color at scanning stage. For reasons that escape me, color temp meters (which I would've assumed went the way of the scrimshaw artist's carving tools) are still fetching $600+ on eBay. Does anybody have a method for checking the equality (and accuracy) of one's 2 heads short of splurging this amount of cash???

Your help is (as always) is appreciated!

[jb7]

I would assume that you could easily check that the two heads were outputting the same temperature (and power) using a DSLR, though I don't know how you could accurately determine the actual temperature or filtration. However, I would imagine that it should be possible, though it would have to be run in conjunction with clip tests.

[jb7]

I suppose it would also be necessary to attach the DSLR to your view camera, so that you were using the same lens...

[Uri A]

Thanks, jb7: fair point!

You're right that I should be able to use a DSLR to verify the color output/rendition of the 2 strobes by studying the histogram of the 2 images, flashing 1 head at a time against a neutral background (correct me if I am wrong - PLEASE!). I would think that this would work equally well assessing 2 shots from the DSLR with lens as mounting it to the 4x5 (for which I don't have the facility).

[jb7]

Yes, a grey card would be ideal, though any neutral tone should work.
For non-imaging purposes, you might be able to bodge something together, to make a lighting test exposure-
Something involving reversing rings or extension tubes and a sheet something with a hole in it springs to mind...

[Uri A]

Something involving reversing rings or extension tubes and a sheet something with a hole in it springs to mind...

Whaaa?

[linhofbiker]

I have shot 4X5 Fuji 64T with tungsten lights (at night) to copy water color paintings for 20+ years. The paintings are a maximum of 30x40 inches. The lens I use is a 135mm Componon-S in a compur shutter. The results are used as a record and can be scanned to make giclee prints of the original painting. I always shoot at 1/2 sec at f/22 since the same setup is used each time (if it is a small painting some adjustment is made for bellows extension)

[jb7]

To make up something like this, although you can buy an adapter on ebay to suit your setup-

http://img262.imageshack.us/img262/1672/dsc3782.jpg

[vinny]

To make up something like this, although you can buy an adapter on ebay to suit your setup-

http://img262.imageshack.us/img262/1672/dsc3782.jpg

Not sure how shooting through the same lens (let's say 135mm) on a 24mm x 36mm sensor is going to show any more than a fraction of the same field of view (135mm on 4x5 film) which would likely only show a fraction of the color information of the original. Yes a different lens may render color/contrast differently but you might as well shoot with the dslr instead of going through all that trouble.

[jb7]

Not sure how shooting through the same lens (let's say 135mm) on a 24mm x 36mm sensor is going to show any more than a fraction of the same field of view (135mm on 4x5 film) which would likely only show a fraction of the color information of the original. Yes a different lens may render color/contrast differently but you might as well shoot with the dslr instead of going through all that trouble.

For me, it would be no trouble at all. The purpose is not to photograph the scene, but to determine colour matching and intensity between two light sources. The transmission differences between the two lenses might be negligible, or a lot, but measuring through the taking lens would remove that variable.

[Robert Brummitt]

Well, i use to use Fuji Astia 4x5 & 8x10 with strobes. That gave me a very nice but rich color transparency. If the client wanted "pop" then Provia. And if some real strong colors and contrast? Velvia.
The one thing you need to be aware is oil paintings. They can give of glare where the brush strokes are. I then used polarizes on the lights and the lens. Lots of tests! Water colors, cryrons and pencil drawings were easy.
Now, I don't even know if Astia or Provia is still made. But they were the best! I did a comparison between Fuji to Kodak and my printers and I and the clients all agreed that Fuji had it down pat.
you may also check into Fuji's color negative. It was so easy to work with.
Boy, I miss mt color lab days!

[Uri A]

Thanks Robert!

I shoot Kodak neg all the time, but for color repro ...it's gotta be chromes: I don't wanna sweat over matching colors on a scan all day :)

But yes, Provia looks OK (and cheap, and reliably available) so I think I'll go the RDP route.

Many thanks for your (and everybody else's input!)

[Uri A]

Oh, and good call on the highlights in the oils!! That's EXACTLY what foiled me some years ago - good to be reminded!

[Drew Wiley]

For color accuracy my no. 1 choice would be Astia. No.2 would be E100G. Good luck finding
either of those in 8x10. Probably some 4x5 still around. A decade ago I would have used
tunsten-balanced Astia (CDUII), but any of that still around has probably expired. RTP
was also quite good for accuracy. Provia would be a last resort.

[vinny]

For me, it would be no trouble at all. The purpose is not to photograph the scene, but to determine colour matching and intensity between two light sources. The transmission differences between the two lenses might be negligible, or a lot, but measuring through the taking lens would remove that variable.

I don't think you're following me. Put a 135mm, 150mm or 210mm lens on your dslr at the same working distance as your 4x5 camera's film plane and you'll only see a small portion of the artwork, not enough to see potential errors. Kinda like shooting 4x5 polaroids on the back of a 16x20 camera.

[Brian C. Miller]

(1) For chromes, what current emulsion would you use for purely color correct rendition under (decent studio calibrated) flash light, to capture maximum detail in all areas of the original? E100G is under the death warrant at B&H - would you shoot Provia 100? I haven't shot a chrome in 8 years, but back in my day it seemed pretty green & cool....

E100G is history, so all that's left is Provia. Just profile the film using a target for scanning accuracy, and you should be fine. Or you can stockpile the remaining E100G. Another option is use negative material, and include a color target in each image. After all, this is large format, with plenty of room to spare, and the target can be cropped out at the end of the process.


(2) back in the day, we used to make sure our 2 heads were flashing the same colour temp to make sure we weren't getting any gremlins in the color at scanning stage. For reasons that escape me, color temp meters (which I would've assumed went the way of the scrimshaw artist's carving tools) are still fetching $600+ on eBay. Does anybody have a method for checking the equality (and accuracy) of one's 2 heads short of splurging this amount of cash???

Since you only want to check if the colors are equal, you can use your DSLR and a color target.

[jb7]

I don't think you're following me. Put a 135mm, 150mm or 210mm lens on your dslr at the same working distance as your 4x5 camera's film plane and you'll only see a small portion of the artwork, not enough to see potential errors. Kinda like shooting 4x5 polaroids on the back of a 16x20 camera.

Vinny, I'm probably not following you, but I was making a suggestion to try to help with the second question Uri posed-

"(2) back in the day, we used to make sure our 2 heads were flashing the same colour temp to make sure we weren't getting any gremlins in the color at scanning stage. For reasons that escape me, color temp meters (which I would've assumed went the way of the scrimshaw artist's carving tools) are still fetching $600+ on eBay. Does anybody have a method for checking the equality (and accuracy) of one's 2 heads short of splurging this amount of cash???"

The reason for using the taking lens, rather than the DSLR lens, was to take account of the colour rendition of the glass, which can vary between lenses. I never suggested that he used the DSLR to proof the entire picture. That was never the question...

[rdenney]

I just did a gig like this. I didn't use 4x5, but shot the film portion using a Pentax 6x7 with a 135mm Takumar macro lens. I used negative film, however, so that I would not be adding contrast to the paintings. I had Fuji 160C in the freezer, so that's what I used. I was afraid Reala and Ektar (my main stock in roll film) would be too saturated.

My lighting was two M11 heads on my Speedotron system, running 800WS each. That provided enough light to overpower the influence of any dim ambient lighting.

The paintings were oils, but with minimal linseed oil so the sheen was not excessive. I did get a touch of specular highlights on a couple of the paintings, but fortunately not enough to cause a problem. I didn't have to use (and test) the polarizers that had I brought along with me.

In addition to the film, I also shot it all on my Canon 5D, using the 50mm Compact Macro lens.

But here's the key: I made the first photo with each technology of a painting that I owned, and included in that picture a reflective IT8 target (5x7"--came with the Monaco EZ-Color with my Epson 750) and a gray card. That gave me ground truth for matching color and brightness with the negative. Frankly, that was no more difficult than overcoming a slide film's natural color cast, especially given the paucity of transparency emulsions currently avaiable.

The digital images turned out fine. I spent about an hour matching color, and then applied the same matching strategy (programmed as an action) to all the images. And the machine prints from the film look spot-on, so I must have hit the center of the target reasonably well. I have not yet scanned the film, but I will be able to profile the scanner in Vuescan for that film and lighting using the IT8 target in the test shot.

The film scans will allow the artist to make reproductions of original size (ranging from 8x10 to maybe 24x28), but for the notecards, books, and website targets, the digital files will be fine. I won't scan the film until she needs something bigger, except maybe as a test.

Rick "happy with the results" Denney

[Lenny Eiger]

I've been shooting flat art since I was a kid. For watercolor use chrome, for other types of paintings use neg if you can. Bracket the exposures on the chromes a third of a stop or two up and down. With neg maybe a half a stop each way. Choose the best exposure and scan it.

I would suggest a drum scanner, of course. If you don't have one, just make the client pay for it. If you are sending a neg, send them a digital shot of the image so they can see where the colors should lay...

When you get to the photoshop part learn how to use channel masking. Remember that film can't reproduce all the colors exactly and you will have to do some work - that's normal. Profiling won't help, the amount of tuning it will save you is minimal. Finally, I find including a set of color swatches or a grey card is entirely and totally useless.

Don't make this too hard... it's actually pretty easy. Just expect the masking and the extra test prints to get this right...


Lenny

[Uri A]

I find including a set of color swatches or a grey card is entirely and totally useless.

Lenny
Thanks Lenny, but if you shoot neg and have no swatches and don't have the artwork in front of you when you scan, how will you get the color right?

[Uri A]

Thanks Rick for your explanations!

Yeah, I'll shoot digi for web as well, but the brief is for chromes, so... still trying to find the best emulsion. Wouldve gone with E100G, but it's all over for Kodak, so looks like it'll have to be Provia with very slight magenta filter perhaps (from memory it's a touch green). I'll test it out.

Thanks everybody

[rdenney]

When you get to the photoshop part learn how to use channel masking. Remember that film can't reproduce all the colors exactly and you will have to do some work - that's normal. Profiling won't help, the amount of tuning it will save you is minimal. Finally, I find including a set of color swatches or a grey card is entirely and totally useless.

Jeez, Lenny, your color memory must be pretty accurate to remember the hues in a painting after it's gone to its new owner. And your digital camera must be pretty accurate to provide the person making the scan an accurate notion of the color. Must be a better digital camera than mine--I carefully and accurately dialed in the color balance on mine but still had to make significant adjustments to align the colors.

And when I got the colors looking right on the IT8 target, shazzam!, they also lined up on the painting I used for a reference. Just like magic. Don't know how I would have lined up colors for paintings no longer in my possession without having included the color chart. But, as I said, maybe your color memory is better than mine.

Rick "it's always easy to guess" Denney

[rdenney]

Man, you guys think painters have more money than photographers. Renting a digital back, laser scanning each image? My last gig would have cost me $1500-2000 in my costs to do either of these.

Artist: "Rick, can you photograph my artwork so I can make reproductions in the future?"

Rick: "Sure. My out-of-pocket costs are $2000, and my price to you is $2800."

Artist: "That's more than I grossed from the paintings that sold at the show!"

Rick: "Yes, but that way you'll be able to make reproductions four times the size of the actual painting, and be able to count the fibers on the canvas."

Artist: "Could I take these pictures myself with my D5000?"

The approach I took with my client limited my out-of-pocket costs to two rolls of Fuji 160C, which I bought five years ago and put in the freezer, plus lab processing, and it added up to maybe $40.

Rick "just sayin'" Denney

[rdenney]

There are always those who have the scratch to pay high prices for high-end work, and those with the skills to earn that work. But those who can figure out how to fulfill the requirements of people without that scratch will find plenty to do, also. Artists are generally much poorer than museums, given that they actually have to sell their work instead of living on donations and grants, but even the starving among them have needs, too.

Rick "providing a good quality alternative to the artist standing the painting against a tree on an overcast day" Denney

[Drew Wiley]

There are always reproduction errors of some sort. Polarization introduces a shift in values
between different elements of the scene simply because it is polarization. Great for destroying impasto. Guess that's why most Rembrandt paintings look horrible in books.
But working without polarization is an even bigger evil. I built a huge copy table system
with fixed hot lights and capable of multiple camera formats. It was very predictable and good casual money for awhile, esp when tungsten or Astia LF film was routinely avail. But I can understand why a specialist nowadays would invest in any expensive scanning back rig and plug an obscenely expensive Apo El Nikkor on it, esp for museum documentation.
Be a nice specialty niche.

[bob carnie]

We do a fair amount of repro work here. Our kit is a phase back with the use of polarized light and we do use macbeth cards to white balance , grey balance and set end points for black and white..
The phase is tethered to a mac workstation and we can control the repro pretty well.
I have done a lot of repro work in the past using film but frankly do not see the point with what is available these days.

We are considering the Betterlight scanning back for this purpose and are looking into the Uberscanner which is touted to be pretty shit hot.
I am looking for a way of taking CieLab numbers on original art and then matching them in photoshop. I think by knowing the colour values of the original it goes a long
way to laying down the same values on paper.

I must concur with what Mr Denny is implying that there is no such thing as colour memory.

[Jim Michael]

I can see the value in including a color swatch chart and grey card in each batch of film. You are being careful with your process control in exposure & lighting, but every color process in the lab has some variation, even when that process is "in control".

[bob carnie]

very true.. I can run 4 step wedges a day on my lambda and read them .. the software will tell me I am in control on each one, but my eyes are telling me they are different...

I can see the value in including a color swatch chart and grey card in each batch of film. You are being careful with your process control in exposure & lighting, but every color process in the lab has some variation, even when that process is "in control".

[Lenny Eiger]

Jeez, Lenny, your color memory must be pretty accurate to remember the hues in a painting after it's gone to its new owner. And your digital camera must be pretty accurate to provide the person making the scan an accurate notion of the color. Must be a better digital camera than mine--I carefully and accurately dialed in the color balance on mine but still had to make significant adjustments to align the colors.

And when I got the colors looking right on the IT8 target, shazzam!, they also lined up on the painting I used for a reference. Just like magic. Don't know how I would have lined up colors for paintings no longer in my possession without having included the color chart. But, as I said, maybe your color memory is better than mine.

Rick "it's always easy to guess" Denney

Well, my color memory is excellent, it always has been. When I was printing color in the darkroom, in 1973, I used to put my hands on the dials of the old Beseler color head and dial it until I saw what I wanted at on the easel. I could easily beat the analyzer to the first good print. I also have somewhat of a photographic memory when it comes to images. I can usually tell you by whom and when a photograph was taken, for any photographic book I have read through.

I am also informed by Art History. While getting my masters in Photography, I had to take 6 years of Art History. I start by understanding what the artist was trying to accomplish. This is a large part of it.

And to be fair, I do my best to hold on to the painting for as long as I can to let the colors seep into my brain, and whenever possible, keep it until I am done printing. (Or borrow it when I am ready to do the final.)

Lenny

[Drew Wiley]

Then there's a whole other skill level ... the guys who translate the photograpic image into
the world of halftone reproduction. I've got some of those early Eliot Porter books - first
editions with the varnished pages (now somewhat yellowed). Contrary to what many people think, those reproductions of Glen Can etc weren't made from dye transfer prints
but from the chromes themselves. They'd polish and etch the plates in such as way ... wow, what a skill level, all pre-scanning.

[rdenney]

I have one of those early Porter editions ("In Wildness...") and it's everything you suggest. It's by far the best book of color photography I've seen from before about 1990.

Rick "who first saw Porter in a showing of his dye transfer prints at the University of New Mexico in the early 80's" Denney