Color photography with black and white film [Archive]

Incoherent Fool

3-May-2009, 21:31

So I was reading about Sergey Prokudin-Gorsky (http://en.wikipedia.org/wiki/Sergei_Mikhailovich_Prokudin-Gorskii), the father of color photography. He traveled all over Russia from 1909-1915 using a very early color photo process. He took three identical black and white photos, each with a red, green or blue filter. Then he would combine all three shots using red, green and blue light to produce a complete color photograph.

Upon reading this I realized the same could be done with modern black and white film using three filters and Photoshop. Below is my first attempt. I took three shots on TMX film (35mm, sorry!) using 25A, 58 and 47B filters, then aligned and merged them in Photoshop.

I figure this could come in handy. For one thing, it's a way to develop "color" film at home without using color chemicals or needing to send it to a lab. It might even be cheaper, although I haven't calculated it. It's also great if you're out in the field with only black and white film and you realize your shot would look better in color.

So does anybody have any experience with this? Is this more trouble than it's worth? Am I crazy?

Ron Marshall

3-May-2009, 21:52

You did a nice job!

If I had the filters I would be tempted to give it a try, just for fun if nothing more. Great if nothing is moving in the shot, although with the right subject one might get interesting effects with some movement.

Eirik Berger

4-May-2009, 00:56

That was really cool!
Are the 25A, 58 and 47B-filters the optimal combination?

Bill_1856

4-May-2009, 01:45

What a GREAT idea! A flash of GENIUS!

Joanna Carter

4-May-2009, 02:18

The most prolific publisher, I know, of "trichromie" is Henri Gaud.

http://trichromie.free.fr/trichromie/index.php?

His blog is all in French, but with a little help from Google Translate, it's surprising how much you can get out of it; and the images are inspiring.

eric t

4-May-2009, 02:45

if you don't speak french the blog might be a little off-putting, so many articles, you don't know where to start.

here is a good starting point, still in french though
http://www.galerie-photo.com/test-trichromie.html

Asher Kelman

4-May-2009, 03:41

So I was reading about Sergey Prokudin-Gorsky (http://en.wikipedia.org/wiki/Sergei_Mikhailovich_Prokudin-Gorskii), the father of color photography. He traveled all over Russia from 1909-1915 using a very early color photo process. He took three identical black and white photos, each with a red, green or blue filter. Then he would combine all three shots using red, green and blue light to produce a complete color photograph.

this post amazed me since I also admired Sergei's work. He did work for the Tsar. I think the first picture was a river view. I was shooting with a Mamiya 7II todat and thought how easy it would be for me to use B&W film for shooting architecture. Snaoshots with the Mamiya then careful shots with my 8x10 and scan and combine. This would allow the most area of data collection and the greatest devotion to color range as each color can be exposed separately and then color balanced in photoshop afterwards.

Asher

Asher Kelman

4-May-2009, 03:42

So I was reading about Sergey Prokudin-Gorsky (http://en.wikipedia.org/wiki/Sergei_Mikhailovich_Prokudin-Gorskii), the father of color photography. He traveled all over Russia from 1909-1915 using a very early color photo process. He took three identical black and white photos, each with a red, green or blue filter. Then he would combine all three shots using red, green and blue light to produce a complete color photograph.

This post amazed me since I also admired Sergei's work and was thinking of it today and planning to use film. He did work for the Tsar, BTW. I think the first picture was a river view. I was shooting with a Mamiya 7II todat and thought how easy it would be for me to use B&W film for shooting architecture. Careful shots and combine. Fewer shots are needed in architecture and, as mentioned, B&W is more easily processed.

Asher

Ken Lee

4-May-2009, 03:42

Eirik Berger

4-May-2009, 04:11

The detail comparison between Provia 100F and Tri-X (tricrom.) was kind of disturbing :eek:

I guess I will have to prepare myself for a increase in BW film consumption. I often use red filter anyway so I can spare two extra sheets now and then to try this out. Film is cheap and life is short.

here is a good starting point, still in french though
http://www.galerie-photo.com/test-trichromie.html

Incoherent Fool

4-May-2009, 08:18

That was really cool!
Are the 25A, 58 and 47B-filters the optimal combination?

These three filters are used for color separation so I knew they would work best. At first I tried 25A, X1 and 80A filters but my yellows were very washed out. The 80A doesn't filter enough red and green. I haven't tried the 25A, X1 and 47B but it might work well enough.

Eirik Berger

4-May-2009, 08:37

Looking in my filtercase (Hitech), I see that I have a red 25, green 58 and a blue 47 (not 47B). But after some quick googleing I think the plain 47 will work well too. I have never used the blue and green filters before, they just came along with I bought the filters a few years ago.

If just the snow could melt away so I can find some color to separate… :)

These three filters are used for color separation so I knew they would work best. At first I tried 25A, X1 and 80A filters but my yellows were very washed out. The 80A doesn't filter enough red and green. I haven't tried the 25A, X1 and 47B but it might work well enough.

Ron Marshall

4-May-2009, 08:49

Have you tried the stronger tricolor filters: 29-61-47?

Jeff Bannow

4-May-2009, 09:05

Silly question - how do you all differentiate which sheet is which color? Trial and error later, or are you marking them some how?

Gene McCluney

4-May-2009, 09:08

The Technicolor (http://en.wikipedia.org/wiki/Technicolor) process involved something quite similar - except that the Technicolor camera took all 3 exposures at the same time, by the use of semi-transparent mirrors. As a result, it was possible to control colors quite nicely, and Technicolor used very rich colors when producing their final blend.

Technicolor required a special camera, special operator, and special consultants. It was a bit like having an old IBM mainframe computer at your company with its required staff of engineers in blue suits: there was a certain "overhead" to doing business.

Kodak came along and invented a film that had 3 dyes embedded in it, which required no special camera and operators, consultants, etc. This allowed average people to shoot color movies, and lowered the cost to studios. The color was "just as good" - not. Soon, Technicolor was used only for high-end movies, and by the 1970's, it basically disappeared. As I recall, one of the Godfather movies was the last one ever made in Technicolor. (I may be wrong)

Because Technicolor was shot in b&w film, it is much more archival than modern color film, which fades over time. We can restore Technicolor films fairly easily. It's a "digital" process, after all.

Technicolor was only shot with the 3-strip b/w camera up thru the early 1950's. The Eastman Color negative then became the film of choice because of its simplicity in shooting. Technicolor continued to make prints using their dye-transfer process by making b/w separation matrix from the camera color negative film. The Godfather was the last Technicolor print before the revival but it was shot on Eastman Color Negative film, just the Technicolor print process was used for the release prints. Technicolor in the 1980s? or early 1990's briefly revived the Technicolor print process, but the expense of making the b/w separations and the printing process was too costly for most film productions, and Kodak improved their release print stock making it more resistant to fade, so Technicolor stopped their print-making process again, but the Technicolor lab remains as a processing and printing lab for Kodak films.

Vick Vickery

4-May-2009, 09:09

One of the Life photo book series included several shots of a photographer using a special camera to take a photo of Roy Rogers; the camera exposed three sheets of B&W film at one time with filters built in. This process was fairly common in the first half of the 20th Century,

Joanna Carter

4-May-2009, 09:11

... I see that I have a red 25, green 58 and a blue 47 (not 47B).
Those are certainly the colours that Henri Gaud uses.

To translate another part of his article on tests that he did, he recommends allowing the following ISO ratings for the three exposures, assuming you are using 100 ISO film

Red : 32 ISO
Green : 20 ISO
Blue : 20 ISO

Henri was kind enough to send me a set of Lee filters for trichromie and I have just got around to mounting them in holders and have started to play with a digital camera before burning some proper B&W film.

Richard M. Coda

4-May-2009, 09:39

Quick (dumb?) question. Are there any special processing differences for each color? (For example, do you have to do red at N+1 and green/blue at N-1?)

Joanna Carter

4-May-2009, 09:48

Quick (dumb?) question. Are there any special processing differences for each color? (For example, do you have to do red at N+1 and green/blue at N-1?)
No, treat all the films the same - N development. Don't forget, although the "effective speed" of each colour is a lot less, you are "adding" three exposures together to make the final print. Any area of white will be the sum of the "1/3" red, "1/3" green and "1/3" blue.

Incoherent Fool

4-May-2009, 09:56

Have you tried the stronger tricolor filters: 29-61-47?

No, but they would probably work. Maybe even better than the ones I've used.

Bruce Watson

4-May-2009, 10:07

Sounds like you are looking for an old Devin Tricolor (http://www.cameramanuals.org/booklets/devin_tricolor.pdf) camera. Purpose built for technicolor-like still photography, that is, three separate 6.5 x 9 cm glass plate negatives (had to use glass plates to ensure good registration between the three negatives), all the filters and mirrors built into the camera. See the diagram on the manual's page 8.

If I read the manual correctly they used Wratten A, B, and C5 filters. I think you'll find a number of interesting things in the manual. My personal favorite is that it took two different emulsions to get the wave lengths right, so the plates aren't all interchangeable. Very interesting -- like nearly all of photography, it's never as easy as it looks.

walter23

4-May-2009, 10:33

So does anybody have any experience with this? Is this more trouble than it's worth? Am I crazy?

I've been wanting to do this, but I have yet to find the right blue filter. It would be great for unusual film sizes, like my little 9x12 plate cameras. You can even print directly if you use three different coloured processes (like cyanotype layered with red & green gum bichromate prints). I wonder if anyone has done this with ULF negatives?

walter23

4-May-2009, 10:37

Silly question - how do you all differentiate which sheet is which color? Trial and error later, or are you marking them some how?

Take notes, same as with anything else.

Of course with only three colours to align it can't be that hard to sort out later anyway.

jnantz

4-May-2009, 10:43

Silly question - how do you all differentiate which sheet is which color? Trial and error later, or are you marking them some how?

you just remember what order you shot them in :)

http://photo-utopia.blogspot.com/2007/11/colour-images-from-b-film.html

walter23

4-May-2009, 10:43

No, treat all the films the same - N development. Don't forget, although the "effective speed" of each colour is a lot less, you are "adding" three exposures together to make the final print. Any area of white will be the sum of the "1/3" red, "1/3" green and "1/3" blue.

Couldn't you, instead, compensate in scanning (assuming a digital post-capture workflow)? That way you could optimize your exposure for each of the three colours? I guess scanning underexposed film is easier anyway.

Joanna Carter

4-May-2009, 11:07

To translate another part of his article on tests that he did, he recommends allowing the following ISO ratings for the three exposures, assuming you are using 100 ISO film

Red : 32 ISO
Green : 20 ISO
Blue : 20 ISO

I've just read more of the "Test" article; Henri appears to go on to say that he finally settles on :

Red : 10 ISO
Green : 10 ISO
Blue : 16 ISO

Joanna Carter

4-May-2009, 11:13

Couldn't you, instead, compensate in scanning (assuming a digital post-capture workflow)? That way you could optimize your exposure for each of the three colours? I guess scanning underexposed film is easier anyway.
If you have exposed the films at an appropriate level, you should have straight scans, although, of course, you can alter whatever you want. If I know Henri well enough, his methodology is usually to do as much in the camera and as little digitally, as possible.

I thoroughly recommend trying to translate http://www.galerie-photo.com/test-trichromie.html, with Google or BabelFish, to see what Henri has to say what someone, who is acknowledged as an accomplished practician in his field, has to say.

Paul Kierstead

4-May-2009, 11:15

Ok, I am fascinated. I have a lee filter pack, might even have those filters. I'd take it that one would want to vary the shutter speed only to avoid strange aliasing issues with changing DoF (if accuracy was the desire, anyway). Assuming digital reconstruction, I am not sure I understand the effective ISO; outside of filter factors, I don't see the need. I expect analog reconstruction would be different, but one would think that less exposure would be wanted, not more. What am I being dumb about here?

sidmac

4-May-2009, 11:43

I did this in a traditional darkroom using a technique we called dye transfer. You would make three B&W negs through filters and then project each one on to a matrix film which held you dyes. After that it was like tricolor graphic arts printing. Took a group class in this and really only scratched the surface of it's potential.
Sid

Joanna Carter

4-May-2009, 11:49

I'd take it that one would want to vary the shutter speed only to avoid strange aliasing issues with changing DoF
Correct; you certainly don't want to change the aperture :o

Assuming digital reconstruction, I am not sure I understand the effective ISO; outside of filter factors, I don't see the need.
The effective ISO is due to the fact that, without filters, you would have uninterrupted passage of light. But, with the interposition of a filter, there is a loss of light transmission which has to be compensated for.

If we take the example of wanting to separate out the red light on a scene, without a filter, a sheet of film that is, theoretically, only sensitive to red light would respond to a theoretical 1/3 of the available light.

However, filters are not perfect transmitters of all the light that strikes them, thus interposing a red filter would attenuate, not only unwanted green and blue light, it would also attenuate some red light due to its very presence.

I expect analog reconstruction would be different, but one would think that less exposure would be wanted, not more. What am I being dumb about here?
Just as, if you were to use a red or orange filter for enhancing skies, you would need to increase your exposure by three or two stops, you still have to compensate for the presence of a trichromie filter. The filters really are quite dense to try and remove as much of the unwanted (parasitic) colours as possible.

Paul Kierstead

4-May-2009, 12:21

The effective ISO is due to the fact that, without filters, you would have uninterrupted passage of light. But, with the interposition of a filter, there is a loss of light transmission which has to be compensated for.

Ahh, many thanks for your reply. It is just the filter factor then; that is a lot of filter factor. They must be dense indeed. Still, I love the ... "flavour" ... of this process, conceptually anyway. I think it is something that will have to be tried once or twice at least, maybe a still life or the like (clearly moving subjects offer a lot of grief, as we can see from the clouds...)

eric t

4-May-2009, 13:41

Assuming you are using TRIX 320

Originally Posted by Joanna Carter View Post
To translate another part of his article on tests that he did, he recommends allowing the following ISO ratings for the three exposures, assuming you are using 100 ISO film

Red : 32 ISO
Green : 20 ISO
Blue : 20 ISO
I've just read more of the "Test" article; Henri appears to go on to say that he finally settles on :

Red : 10 ISO
Green : 10 ISO
Blue : 16 ISO

jb7

4-May-2009, 13:55

This is really interesting-
about a year ago I bought a 47 blue,
to go with my standard b&w red and green filters-
I've always meant to give it a go,
but never got around to it-

The numbers that are coming out are useful as a starting point,
though the last set seem counter intuitive-
Am I missing something?
Which film do the numbers refer to?

My school French might not get me that far...

I've got some Fomapan 200, must give that a go...

I wonder which it's closest to?

Struan Gray

4-May-2009, 13:59

Silly question - how do you all differentiate which sheet is which color? Trial and error later, or are you marking them some how?

If you've scanned the negs, just shuffle which one is assigned to which colour channel in assembling the final colour digital image until it looks right.

One of the surprising things this process turns up, is how often something looks right, even when it is completely wrong.

Henri Gaud has taken an idle musing of mine and turned it into an investigation of pure colour - and our brains' acceptance of almost random colourings once they get away from looking at flesh tones or nature scenics. Despite the number of times my name crops up in his pages, the credit is entirely his:

http://trichromie.free.fr/trichromie/index.php?post/2009/04/08/1118-le-point-sur-les-jeux-trichromes

Joanna Carter

4-May-2009, 14:26

Assuming you are using TRIX 320
Apologies; I had misread/translated some of the article :rolleyes:

Larry Gebhardt

4-May-2009, 16:23

There were also tri-color cameras that would take all three images at once. Then you had no registration issues. I've been thinking of building one, or trying to find a pristine older example.

sanking

4-May-2009, 18:40

In the past I did a lot of color separation work with B&W film, making three shots with a view camera. For in-camera separations the recommended filter set is 25 Red, 58 Green and 47 Blue. The 29, 61 and 47b set is recommended for making separations from transparencies. You mark the film holders as R, G or B, and before developing clip the R on one corner, the Green on two corners and leave the B as is. This helps to identify them later. Anyone who actually goes through this can combine the separation negative files, after scanning the negative, using the protocol in the attached document called MergeSeparations.

The best film for color separations is one that has a short toe and shoulder and a very long straight line. Either Tmax-100 or Tmax-400 work quite well. For best results you will need to expose and develop the three records individually, but if you have the luxury of curve correction in Photoshop just give +3 stops exposure for the Red and Green filters, and +2 stops for the Blue filter. Develop the Red and Green negatives for the same time, the Blue for about 50% more.

From the turn of the century until the early 1950s one-shot color cameras were manufactured that made three exposures at one time, one through the Red filter, the other through a Green filter, and a third through a Blue filter. This was achieved by means of pellicle beam splitters that transmitted a part of the light but reflected the other part.
These cameras were very expensive at the time. I am in the process of restoring a National Fotocolor 5X7 one-shot camera that sold for over $1200 in 1949. That would amount to something on the order of $25-35K in 2009 dollars. See attached file.

Sandy King

jb7

4-May-2009, 20:08

Thanks Sandy, very helpful...

sanking

4-May-2009, 21:28

Thanks. Hope you find the information useful.

Recently I have been scanning and merging in Photoshop some of the in-camera separations I did some two decades or more ago. Most have never been printed or seen as full color images.

The attached image is of Raven Cliff Falls in upper Greenvile County, SC. I made the original set of separations at about 7 am in very soft light, after a hike of two hours, following a drive of 1.5 hours to the site. The early morning shot was necessary because at this time of year the sun comes up almost directly behind the top of the falls. You can see the hot spot from sun illumination beginning at the top right of the image. 15 minutes later this shot would have been impossible because of contrast and flare.

The separations are on Ektapan 5X7, made through 25, 58 and 47 filters. I have printed in carbon other versions of this site, but this is the first time I have looked at the full color image of this separation set. The image at this point is pretty much direct from the separation set and could still use some corrections, but I think it gives an idea of the potential of LF in-camera separations with a regular view camera.

Sandy King

Thanks Sandy, very helpful...

D. Bryant

4-May-2009, 23:37

The attached image is of Raven Cliff Falls in upper Greenvile County, SC.
Sandy King

Sandy,

Have you ever photographed Cullasaja Falls outside of Highlands, NC? I'm sure you probably have but I have yet to make a stellar shot there. The shot of Raven Cliffs falls reminded me of Cullasaja Falls and Gorge. Water falls aren't usually my forte, but the falls around Highland are spectacular and challenging.

The premo spot I've found is to hike up the cliff across the highway from the spot that everyone stops at in the summer and fall. Setting up a tripod there is somewhat perilous but gives a nice elevated view of the scene. I would suspect local photographers have the good angles figured out by now but I've not seen anyone's work that I would consider outstanding of this particular scene. Another strategy is to climb sown from the highway edge but this is quite dangerous. I dropped my Nikon F2 there one day while attaching it to a tripod and watched it sail over the edge. Unbelievably I recovered the camera a few yards down with only a few minor scratches.

Don Bryant

jb7

5-May-2009, 00:47

looks great Sandy-
great colour-

Presumably, it was a still morning...

Using this technique, I suppose movement, as in leaves, or waves, or clouds,
might lead to coloured fringing...

sanking

5-May-2009, 06:21

Don,

I have photograhed Cullasaja Falls but only from the tourist spot and don't have anything close to spectacular of it. I rarely ever photograph waterfalls nowdays but at the time it was a lot of fun because of the experience of being in nature and the long, and sometimes very strenous, hikes.

Sandy

Sandy,

Have you ever photographed Cullasaja Falls outside of Highlands, NC? I'm sure you probably have but I have yet to make a stellar shot there. The shot of Raven Cliffs falls reminded me of Cullasaja Falls and Gorge. Water falls aren't usually my forte, but the falls around Highland are spectacular and challenging.

Don Bryant

sanking

5-May-2009, 06:33

My recollection is that the place was very, very still that morning, and the separations show that to be the case. You will indeed get rainbow and other color fringing effects with anything that moves while the exposures are being made. However, if the subject is still making separations on B&W film with a regular view camera is capable of excellent results, potentially even better than with color negative or transparency film of the same size.

Sandy

looks great Sandy-
great colour-

Presumably, it was a still morning...

Using this technique, I suppose movement, as in leaves, or waves, or clouds,
might lead to coloured fringing...

eric t

5-May-2009, 13:02

here you go

http://trichromie.free.fr/trichromie/index.php?post/2009/04/19/Finale-Rayol-Ooblik-2009

Using this technique, I suppose movement, as in leaves, or waves, or clouds,
might lead to coloured fringing...

jb7

5-May-2009, 13:08

yes, thank you Eric-

It is a little bit of a deal breaker-
I suppose subjects might just end up choosing themselves...

Whenever there's a nice bit of light in this country,
it's invariably accompanied by wind-

And it only goes still again when the sun goes behind a cloud...

Struan Gray

5-May-2009, 13:27

If you can live with the risk of parallax fringing, a multi-lens camera will eliminate motion fringes. With a stereo setup you can do bichromes, but there are four-lens cameras by Polaroid and Cambo for taking four passport-sized photos on 4x5 which would let you do three colours and black, or whatever you want. Sinar also made a lensboard with four 125 mm radionars, plus a septum to put at the focal plane.

sanking

5-May-2009, 13:35

That is very interesting. Could you point me to the four-lens Polaroid and Cambo cameras that do four shots on 4X5? I suspect that the merging software could overcome the slight difference in image size that would result from the use of different lenses. Not sure about the parallax fringing, though.

Do you know of anyone who has actually used one of these cameras for making separations?

Sandy

If you can live with the risk of parallax fringing, a multi-lens camera will eliminate motion fringes. With a stereo setup you can do bichromes, but there are four-lens cameras by Polaroid and Cambo for taking four passport-sized photos on 4x5 which would let you do three colours and black, or whatever you want. Sinar also made a lensboard with four 125 mm radionars, plus a septum to put at the focal plane.

Struan Gray

5-May-2009, 13:49

Both the Polaroid and Cambo models were called 'Miniportrait' cameras. Some were marked with both maker's names. For the lifetime of the auction you can see a picture of a typical model here:

http://www.budera.se/detail.asp?id=7578624

Most of them have a fixed focussing distance (they often have a built-in tape measure, and/or a fixed rangefinder) which limits their use for general photography. There were some with adjustable focus - I'm kicking myself for passing up on one for pennies last year.

I don't know of anyone using one for colour separations, but I intend to (when I get round to it :-). I think the parallax fringing will bother me less than motion fringes, and I can live with the tradeoffs. The idea of having all colour bands on one piece of 4x5 film is quite attractive too.

wclavey

7-May-2009, 05:17

sanking

9-May-2009, 20:28

A few days ago I posted an image of Raven Cliff Falls in Greenville County, SC. When I was working with the separation set I could not understand why the greens were so muted. Then, it hit me. In a flight of stupidity I combined Blue, Magenta and Yellow instead of Cyan, Magenta and Yellow. I am attaching a new file with the correct combination to show the difference.

Sandy King

Joanna Carter

10-May-2009, 01:35

Sandy, Which filters did you use and what exposure compensations did you make for each filter ?

sanking

10-May-2009, 08:43

I used Red (25), Green (58) and Blue (47). The exposure adjustment I made was +3 stops for the Red and Blue filter negatives, and +2 for the Blue filter negative. The film I used for this separation set, Ektapan, is no longer made, and every film will require slightly different compensation and development to produce a balanced separation set. However, even sets that are slightly out of balanced can be fixed in Photoshop by density and curve adjustments.

Sandy

Sandy, Which filters did you use and what exposure compensations did you make for each filter ?

D. Bryant

10-May-2009, 09:30

I used Red (25), Green (58) and Blue (47). The exposure adjustment I made was +3 stops for the Red and Blue filter negatives, and +2 for the Blue filter negative.

Sandy

As an off topic comment though related to the discussion, this thread reminded me of the Harris tri-color special effects shutter introduced and written about years ago by Kodak photographer Bob Harris. The technique and results are entirely different but maybe intersting for some.

Here is a link to a Flickr web page illustrating effects from the use of the Harris shutter:

http://www.flickr.com/groups/technique/discuss/72157602020995316/

By today's standards the results may look a bit hoaky but might be a fun diversion to those so inclined to try it.

Don Bryant

sanking

10-May-2009, 13:23

Here is another color image produced in Photoshop from in-camera color separations.

This one has some slight misalignment, probably from camera movement, and some color fringing from water on the plant which I sprayed just before making the negatives. Look at this in large magnification if you can because the fringing artifacts are quite beautiful.

Sandy King

ric_kb

10-May-2009, 14:31

the ideal film for this would be super XX...

Joanna Carter

10-May-2009, 17:48

And here is my first attempt at a "trichromie"

http://grandes-images.com/fr/Trichromie/DesBieres.jpg

Taken on a Mamiya RZ using Lee polyester filters, scanned, cleaned and merged in Photoshop.

Paul Kierstead

10-May-2009, 20:37

Very nice, Joanna. Possibly invfluenced by my fondness for S'Peters Ale (haven't tried Ruby Red, though). All of these have a unique look that is hard to put a finger on.

Bill_1856

10-May-2009, 20:50

IMO it's a frigging waste of time, energy, and film.

Joanna Carter

11-May-2009, 02:45

IMO it's a frigging waste of time, energy, and film.
Hmmm, interesting change of opinion from your first post to this thread :cool:

Joanna Carter

11-May-2009, 02:49

Very nice, Joanna. Possibly invfluenced by my fondness for S'Peters Ale (haven't tried Ruby Red, though).
I came across the Ruby Red, a few months ago, at our local Booths supermarket; it was going cheap, so we tried it, liked it and now buy it, even though it is no longer at the reduced price. It is slightly sweet but lighter than the Theakstons or Oyster stouts.

All of these have a unique look that is hard to put a finger on.
Are you talking about the images or the beers? ;)

sanking

11-May-2009, 05:05

Very nice work, Joanna. For your first "trichromie" I would call this a great success.

Sandy

And here is my first attempt at a "trichromie"

http://grandes-images.com/fr/Trichromie/DesBieres.jpg

Taken on a Mamiya RZ using Lee polyester filters, scanned, cleaned and merged in Photoshop.

sanking

11-May-2009, 05:07

Very nice work, Joanna. For your first "trichromie" I would call this a great success.

Did you expose and develop all three records on the same roll of film?

Sandy

Joanna Carter

11-May-2009, 06:03

Did you expose and develop all three records on the same roll of film?
Yes Sandy, I didn't want to burn big sheets to start with, so I used the RZ67. I took a series of shots of a colour chart to start with, to allow me to create a balancing curve that I could use with the subsequent series. So, I have taken three series: the colour chart, this one and some dried flowers in a vase, which I have yet to merge.

I shall be going to our regional monochrome group tonight and shall present this shot as a B&W image :cool: :eek: ;)

Bill_1856

11-May-2009, 06:18

Hmmm, interesting change of opinion from your first post to this thread :cool:

My first post was (unrecognized) sarcasm, as the process has been well known since Clark Maxwell demonstrated it in 1855. It is how the printing industry prints color, how they first made color movies, and anyone who ever did Dye Transfer 60 years ago did it routinely. It is an archaic process, of historical value only. It is, and was, also a PITA.

sanking

11-May-2009, 06:24

Joanna,

You were smart to include a color chart. Another option is to include a grayscale tablet somewhere in the image area and once you scan your separations you can just balance the separations in PS by adjusting the grayscale so that the densities match in the highlights, mid-tones and shadows. In the old days of assembly color printing the operators usually balanced separation sets by intensifying or reducing each of the individual records.

The use of PS to balance separation sets is a powerful tool and has the potential for some outstanding quality work in color that would not be possible with either color negative or color transparency film, assuming a static subject. With B&W film you can capture a huge dynamic range, far more than even color negative film, and with the appropriate choice of film and scanning technique you could make color prints of exceptional resolution and fine grain. Even assuming that you scan with a consumer flatbed like the Epson V700 or V750 it should be possible to capture some 2000-2400 spi of real resolution from a sheet of 4X5 film. That would give you a RGB file of some 80 - 120 mp of true detail.

Making color separations in outdoor scenes shares some of the same working problems as working with a scanning back, and final quality should be very similar. Look, for example, at some of the articles on digital field photography with scanning backs at http://www.betterlight.com/.

Sandy King

Yes Sandy, I didn't want to burn big sheets to start with, so I used the RZ67. I took a series of shots of a colour chart to start with, to allow me to create a balancing curve that I could use with the subsequent series. So, I have taken three series: the colour chart, this one and some dried flowers in a vase, which I have yet to merge.

I shall be going to our regional monochrome group tonight and shall present this shot as a B&W image :cool: :eek: ;)

Joanna Carter

11-May-2009, 06:33

My first post was (unrecognized) sarcasm, as the process has been well known since Clark Maxwell demonstrated it in 1855. It is how the printing industry prints color, how they first made color movies, and anyone who ever did Dye Transfer 60 years ago did it routinely. It is an archaic process, of historical value only. It is, and was, also a PITA.
I suppose it's all about the difference between those whose desire is to make pictures as an artisan process, as opposed to those who merely take pictures. You may not want to take any effort to create something different from the norm; I rather enjoy taking the time to experiment. As for archaic, I take it you don't class film as archaic as opposed to that digital stuff?

gevalia

11-May-2009, 06:33

My first post was (unrecognized) sarcasm, as the process has been well known since Clark Maxwell demonstrated it in 1855. It is how the printing industry prints color, how they first made color movies, and anyone who ever did Dye Transfer 60 years ago did it routinely. It is an archaic process, of historical value only. It is, and was, also a PITA.

Bill,

In reading your response, it looks like you're trying to shove your opinion down peoples throats. Now, I hope I'm reading this wrong and that that was not your intent. It sounds like you have some personal knowledge on the subject. Maybe you can give us more info on your experiences.

Paul Fitzgerald

11-May-2009, 07:09

"It is, and was, also a PITA."
"In reading your response, it looks like you're trying to shove your opinion down peoples throats."

From reading the Kodak's instructions from the mid 40's, it WAS a PITA. Using the Tri-color filters to take the photo AND to print the photos does nothing for color balance SO besides different exposure/development for each filter you needed to make color-correction masks for each filter from the other 2 AND usually a highlight/shadow mask to reduce the dynamic range. Their suggestions would use an entire box of 11x14 for one 8x10 print. By the mid 60's Kodak did mention a different set of filters for printing, eliminating most of this expense.

Struan Gray

11-May-2009, 07:23

A local gallery had an exhibition a while back of work by a Danish illustrator who did a lot of lifestyle and advertising work in the 1950s and 60s, including Coca-Cola ads. I've forgotton his name (I'm not thinking of Haddon Sundblom), but I vividly remember one original drawn/painted illustration marked up for reproduction. The colours used were very different from the colours on the printed page - the tricolour separation process was optimised to a particular film with a particular set of contrast curves and it simplified the reproduction chain if the original was in the 'wrong' colours.

I.e. if a particular green was wanted in print, the painter would use a different hue which stimulated the seperation negatives by the right amount to produce green in the final combination. I assume this meant all the seperation negs could be processed in a standard way, perhaps in the same baths.

Ceramic artists have to perform a similar mental juggling trick with their colours, but it was interesting to see it used in paint.

sanking

11-May-2009, 08:04

There is the point. If our only interest in photography were merely "making a picture" there would be no logical reason to use anything other than a good DSLR and Epson printer. Many, however, are interested in the process as a valuable way to spend time in exploring the artistic potential of both contemporary and historical processes.

It is important to point out that while making color separations has its roots in photography's infancy, the ability to *easily* combine, balance and manipulate color separation sets with image software on a computer is of fairly recent origin. In the early 1990s this ability was only found at high end color printing establishments with scanning and output workstations costing in the hundreds of thousands of dollars. Today we can do the same or better work with equipment and software that virtually any of us can own and use.

Sandy King

I suppose it's all about the difference between those whose desire is to make pictures as an artisan process, as opposed to those who merely take pictures. You may not want to take any effort to create something different from the norm; I rather enjoy taking the time to experiment. As for archaic, I take it you don't class film as archaic as opposed to that digital stuff?

Joanna Carter

11-May-2009, 08:26

Today we can do the same or better work with equipment and software that virtually any of us can own and use.
Which is something I found quite astounding. After months of procrastination, leaving the filters tucked away, I finally set to work yesterday and, within about two hours, not including film drying time, I had the finished result.

What is more interesting is discovering that there is more than one way of combining the images; and that they produce a different feel to the final result.

I started off by applying CMY colours to the layers, using the "screen" mode and then converting the top two layers to "multiply" mode. After applying the curve from the colour chart, this produces a very accurate colour rendition and quite strong colouration.

I also applied RGB colours to the layers, using the "multiply" mode and then converting the top two layers to "screen" mode; which, after applying an "RGB" curve from the colour chart, it gave a softer, less accurate colouration which gave a certain "charm" to the image, marking it out, more clearly, as a different process.

Marko

11-May-2009, 10:17

Which is something I found quite astounding. After months of procrastination, leaving the filters tucked away, I finally set to work yesterday and, within about two hours, not including film drying time, I had the finished result.

What is more interesting is discovering that there is more than one way of combining the images; and that they produce a different feel to the final result.

I started off by applying CMY colours to the layers, using the "screen" mode and then converting the top two layers to "multiply" mode. After applying the curve from the colour chart, this produces a very accurate colour rendition and quite strong colouration.

I also applied RGB colours to the layers, using the "multiply" mode and then converting the top two layers to "screen" mode; which, after applying an "RGB" curve from the colour chart, it gave a softer, less accurate colouration which gave a certain "charm" to the image, marking it out, more clearly, as a different process.

There are also several ways to combine separations using channels instead of layers.

* After scanning, aligning and adjusting your separations, create three flattened grayscale images, containing one separation each.

* In one of the open images, switch from Layers Palette to Channels and select Channel Merge... from the palette options menu.

* After making your selections, you will end up with a color image in the mode you specified with all the open images converted to appropriate channels.

* It is very important that the starting grayscale images have identical pixel dimensions and be aligned and flattened.

* The number of grayscale images open at the time of using Channel Merge... command determines which color modes can be created. You can merge three images into an RGB color image and four into a CMYK one.

There are other methods for blending channels, such as using Calculations or Apply Image commands.

What is really amazing about this particular technique (creating color images using B&W film) is how many different results there are and how many different ways to get there, all starting from the same point!

For those who see no purpose in the exercise and who think their camera, processing and computer skills need no further improvement, there's always a TV with a sofa and a sixpack in front of it, no need to bother with this Internet thing either... ;)

Bill_1856

11-May-2009, 11:11

I have no problem with anyone who wants to use any process for the intrinsic artistic value, nor learning any process to salve their intellectual curiosity, but to resurrect printing color from sep negatives because there is a claim that it's a better or newer way to do it is out of their mind. Been there -- done that, (60 years ago when it was the only game in town.)
Just MY opinion, and worth exactly what you paid for it.

Paul Kierstead

11-May-2009, 11:23

... but to resurrect printing color from sep negatives because there is a claim that it's a better or newer way to do it is out of their mind.

Ok, you are entitled to your opinion, however I fail to see the basis of it; I don't see claims of better or newer. In fact, the very first post in this thread referenced work done 1909-1915. In fact, this whole discussion has been focused around the actual technique, and curiosity about it, not a pissing match of whose technique is better.

Its fine to not be an enthusiast, but I fail to see why it is necessary to introduce a straw man argument.

sanking

11-May-2009, 12:56

Bill,

If you have any of those old separation sets from 60 years ago it might be interesting to pull one of the best ones out, scan the negatives and merge and adjust the files in Photoshop as some of us are talking about. Until then the "been there done that" comment only applies to what you did back then, not to what we are doing today. I think you would be pretty shocked if you, or someone else, were to actually work on one your old separation sets with a current photo manipulation software. Just look what the Smithsonian was able to do with Prokudin-Gorsky's separatons, and I don't want to even imagine how complicated that work must have been.

Sandy King

I have no problem with anyone who wants to use any process for the intrinsic artistic value, nor learning any process to salve their intellectual curiosity, but to resurrect printing color from sep negatives because there is a claim that it's a better or newer way to do it is out of their mind. Been there -- done that, (60 years ago when it was the only game in town.)
Just MY opinion, and worth exactly what you paid for it.

Ben Syverson

11-May-2009, 21:59

This process may not be the newest, but it does have a claim as "best" for one scenario. For 100% still subjects (still lifes, art reproduction), a three-pass B&W trichromie will blow away any current color materials in terms of detail. Both color print and slide film max out at around 70 lp/mm, but with trichromie you could easily do twice or three times as good as that (provided your lens can keep up).

Kirk Fry

11-May-2009, 22:57

As the choices in large format color film go away this may be all we have left. And photoshop got here just in time to save us. It solves the two major problems of color separation: registration and color balancing. Three chunks of black and white film cost about the same as one of color and and you don't need expensive and toxic and hard to get chemicals to develop it. Almost all color printing is being done by computer anyway and while the color separation is a little more complicated it is not overly so. I see it more as 21 century method. All color photoshop images are just three (or 4) black and white layers.

KFry

Ben Syverson

12-May-2009, 12:09

Of course, there goes portraiture...

sanking

12-May-2009, 12:16

Not if you have a one-shot camera that exposes the Red, Green and Blue negatives at the same time!!

Sandy

Of course, there goes portraiture...

sanking

12-May-2009, 13:21

Marko

12-May-2009, 14:07

I have tried several methods of merging the Red, Green and Blue negatives in Photoshop. So far the best method appears to be Automate>Photomerge, then choose Layout = Perspective, and deselect Blend Images Together. Most separation sets, especially older ones, will have small differences in size becaue of slight difference in the focus of Red, Green and Blue light, and may also need to be rotated, warped or skewered due to scanning procedure. The Automate>Photomerge method has worked well for me with some older sets made with even older optics.

Sandy King

Yes, it is the best method, but I prefer to do it a bit more manually - I typically bring all (three) layers into the same image, select them all in the layer palette (they all have to be floating layers for this to work) and then run Edit/Auto Align Layers... Normally, I just set it to Auto and it takes care of all the individual discrepancies.

Photomerge actually uses the same function, only as a part of a wider script.

Before Photomerge was introduced, I would manually align the layers two at a time by setting the blending mode of the bottom one to Normal and the top one to Difference and then moving, resizing or scewing them until everything turns black. It was a royal PITN and a main reason why I didn't try the Tri-color separations (Trichromie?) more seriously.

sanking

12-May-2009, 19:45

I did that for many years in stitching 12X20" negatives scanned in two passes. I thought it worked very well and at prints of same size as original negative I never saw any artifacts from the lack of perfect alignment.

Then I started scanning 6X7 and 6X9 negatives in two passes at 5080 spi and printing them at 6X and larger. The lack of perfect alignment from manual stitching may not show on the screen, but it often shows in high magnification prints, or when you look at the image on the monitor at a higher magnification. Photomerge, when it can be used, gives pretty much perfect results that won't show any artifacts of stitching regardless of magnification.

Sandy

Before Photomerge was introduced, I would manually align the layers two at a time by setting the blending mode of the bottom one to Normal and the top one to Difference and then moving, resizing or scewing them until everything turns black. It was a royal PITN and a main reason why I didn't try the Tri-color separations (Trichromie?) more seriously.

sanking

12-May-2009, 20:23

Here is another image made from a fairly old separation negative set.

Site is a fairly small water fall in upper Greenvile count, SC, set behind a dogwood tree in fall foliage.

Does anyone else have any color separation sets made of landscape? Most the work of this type I have seen has been of still life arrangements.

Sandy King

Eirik Berger

13-May-2009, 00:58

I can´t wait to try this. If the surroundings just could get a little color.

Paul Kierstead

13-May-2009, 13:46

Here is another image made from a fairly old separation negative set.

Site is a fairly small water fall in upper Greenvile count, SC, set behind a dogwood tree in fall foliage.
...
Sandy King

That is a really lovely image, Sandy. I am curious about something about your landscape images. The rocks, in particular, really look B&W; I mean they have something (and I really can't put my finger on it) that makes them look B&W. Have you processed this specific area in some way, or is it perhaps something to do with having been shot on B&W film? It is hard to even define what I am talking about; it is just that rocks, as show on typical colour film, have a "look" that is a little different then yours show. Or maybe I am imagining things.

Jeff Bannow

13-May-2009, 13:49

Any thoughts on what would happen if you used this technique with infrared film?

jb7

13-May-2009, 13:59

False colours-
whatever colours you like, I suppose...

sanking

13-May-2009, 14:26

Paul,

Thanks for your kind comment about the image.

And yes, there is some processing of specific areas in my trichrome images. The first thing I do after scanning the negatives is to adjust the B&W separations so that areas of the print that should be neutral tone have the same density and contrast. Granite rocks, which are the predominant rocks in the mountains of my area of the country, are quite neutral gray in tone so I adjust these areas of the separation negatives to match on all three records. After I convert the image to color I first adjust the color with Adjustment>Color Balance. Then I look at specific areas to see if the tones are as I would like them to be in terms of hue, saturation and lightness. There is a tool in PS for this called Hue/Saturation. This is a very useful took because with it you can saturate or de-saturate the entire image, or any primary or subtractive color in any part of the image. In the case of this image I selected a large part of the rock area on the right and de-saturated the offending colors, which were magenta and blue. One could also entirely de-saturate certain areas so that you would in efffect have a dual tone image, monochrome in one area and color in the other, but I did not go quite that far with this particular image.

Hope that answers your question.

I really hope others get interested in this type of work. It is really fascinating because it combines some of the very oldest techniques of photography with contemporary scanning and image manipulation software to produce imagery that is potentially highly unique.

Sandy

That is a really lovely image, Sandy. I am curious about something about your landscape images. The rocks, in particular, really look B&W; I mean they have something (and I really can't put my finger on it) that makes them look B&W. Have you processed this specific area in some way, or is it perhaps something to do with having been shot on B&W film? It is hard to even define what I am talking about; it is just that rocks, as show on typical colour film, have a "look" that is a little different then yours show. Or maybe I am imagining things.

Eirik Berger

13-May-2009, 14:35

I really hope others get interested in this type of work. It is really fascinating because it combines some of the very oldest techniques of photography with contemporary scanning and image manipulation software to produce imagery that is potentially highly unique.
Sandy

Yes, it is this combination of traditional tecniques and modern digital processing that fascinates me. We can pick and chosse the best from both worlds.

sanking

13-May-2009, 18:59

Erik,

Yes, the combination of old and new.

Also, the process involves many personal creative choices regarding color balance, saturation, tonal range, etc. No color profiles here, you establish all of this depending on what you want, within the limits of the original color separations. But it is definitely not false color, but what was contained on the record of the original three color separatioins.

I am attaching another image that I made today from a set of color separations made a long time ago. This is the first time I have ever seen this image, excepting what I saw when I was there nearly three decades ago. The color is very vivid, but it is what I recall seeing that day in the mountains of North Carolina in late fall 11/1982.

Sandy

Yes, it is this combination of traditional tecniques and modern digital processing that fascinates me. We can pick and chosse the best from both worlds.

Joanna Carter

16-May-2009, 11:09

At last, here is the other shot that I made at the same time as the beer bottles.

http://grandes-images.com/fr/Trichromie/FleursArtificielles.jpg

Paul Kierstead

16-May-2009, 14:21

Paul,

And yes, there is some processing of specific areas in my trichrome images

....<lots of good stuff not repeated>

Hope that answers your question.

It does indeed. The pictures were just throwing me a little; typically, the stones would have a cast (even a very very tiny one) and these were so neutral. The very fine tonal gradation further increased the effect. Thank you very much for the explanation.

Eirik Berger

16-May-2009, 15:24

It does indeed. The pictures were just throwing me a little; typically, the stones would have a cast (even a very very tiny one) and these were so neutral. The very fine tonal gradation further increased the effect. Thank you very much for the explanation.

I agree. The colors (in general) on the images in this thread has a certain quality. The rocks in Sandys images and the straws in Joannas image.

The colors are both neutral and saturated at the same time in a strange way. I really like the "look". I can´t wait to try it out myself. Or maybe it might just be me, I have had a few too many drinks after winnng the European song contest in Moscow tonight… :p

sanking

16-May-2009, 16:25

Joanna,

Beautiful saturated color and perfectly balanced grays. Really nice.

Sandy

Drew Wiley

16-May-2009, 19:38

Some of us still do it the old way - with a registration punch and film, and it's not just for nostalgia, despite what "Bull" thinks (misspelling intentional). If you've ever
looked at really good carbon or DT prints you might not be so thrilled with your
inkjets. But originally, tricolor seps were not even for prints, but for projecting thru
three different lantern slide projectors each itself fitted with a color filter. Never seen this done myself, but was once told by an old timer that the effect has never been equalled since, even by Technicolor.

cjbroadbent

20-May-2009, 05:59

...."IF WE ARE GOING TO HAVE COLOUR PHOTOGRAPHS, FOR HEAVEN'S SAKE LET'S HAVE A RIOT OF COLOUR, NONE OF YOUR WISHY-WASHY HAND-TINTED EFFECTS" - Madame Yevonde
Madame Yevonde (Yevonde Philone Cumbers, 1893 - 1975) was a photographer who pioneered the use of colour in portrait photography.
In 1911, she took an apprenticeship with Lallie Charles, the leading woman portrait photographer of the day. In 1914, having only taken one actual photograph, Madame Yevonde decided to set-up her own studio. Over the years she gained quite a unique and personal style, as well as a name for herself with London society, as a premier portrait photographer. In 1921, she started exhibiting her work at The Royal Photographic Society Annual Exhibition.....

Isn't that the trichrome camera Sandy mentioned elsewhere?

Joanna Carter

20-May-2009, 06:22

Isn't that the trichrome camera Sandy mentioned elsewhere?
Wow, that's some weight to hand hold !!!

Struan Gray

20-May-2009, 07:29

That was her Vivex camera. Imagine using it with three graphmatics for speed :-)

http://www.users.waitrose.com/~felice/index2.htm

http://www.npg.org.uk/collections/search/person.php?LinkID=mp06547&role=art&page=1

She obviously had a sense of humour, even if many of the colour portraits look downright hokey by today's standards.

Joanna Carter

20-May-2009, 08:01

That was her Vivex camera.
Wow, gotta get me one of those :eek:

She obviously had a sense of humour, even if many of the colour portraits look downright hokey by today's standards.
Heheh, it should be against the rules of LF to take yourself too seriously :p

BradS

21-May-2009, 19:35

At last, here is the other shot that I made at the same time as the beer bottles.

http://grandes-images.com/fr/Trichromie/FleursArtificielles.jpg

Fantastic!

Does anybody know how to do this in Gimp? I cannot afford Photoshop.

Tracy Storer

21-May-2009, 21:19

This has fascinated me for a long time, I was learning Photoshop at the same time I was doing a lot of gum-printing, mid '90s, and it just made sense. One of the first things I did in the computer lab at Mass College of Art (where I was working) was to use the B+W "eyeball" quick-cam and shoot through sep-filters and combine in PS. (PS3)

Eirik Berger

8-Sep-2009, 09:26

Here is my first attempt in tricolor work. T-max 400 using a 6x9 rollfilm holder and 25, 58, 47 filters.
It is of a walrus skull taken on Svalbard earlier this summer. This was great fun and I am happy with the result. And the subject haven´t moved in 400 years so that did not cause any problems at all.

http://farm3.static.flickr.com/2649/3900928436_3637acaae8_o.jpg

Joanna Carter

8-Sep-2009, 09:41

Here is my first attempt in tricolor work. T-max 400 using a 6x9 rollfilm holder and 25, 58, 47 filters.
It is of a walrus skull taken on Svalbard earlier this summer. This was great fun and I am happy with the result. And the subject haven´t moved in 400 years so that did not cause any problems at all.
I love the tiny little flowers in the lichen. Beautiful shot.

Robert Hughes

8-Sep-2009, 10:52

Marko

8-Sep-2009, 12:25

One of the historical reasons for shooting color separations was to achieve archivability of color images. Color film is notorious for color fade, whereas b+w film lasts as long as the base holds together.

I'm sure all those Photoshop files we generate nowadays will be readable in 100 years (just try reading any of your 20 year old files from disk today), but if you wanna have a backup with good color, separations will work long after the last OS X machine has turned off.

The historical reason for shooting color separations was to have color.

If you look at the work of Sergey Prokhudin-Gorsky, you will notice that most of his tri-color work was done during the first decade of the 20th century.

As for computer file readability, it all depends on the knowledge and skill of the operator. The GIGO concept applies there with a vengeance, just like in all other human endeavors.

Jeff Bannow

9-Sep-2009, 09:13

Robert Hughes

10-Sep-2009, 08:06

The historical reason for shooting color separations was to have color.
Of course the original reason was to have color. I'm referring to the efforts of the movie and still film industries in the mid and late 20th Century to preserve old 2- and 3-color films by making color separations of prints before they faded (original, mid-century Agfa and Eastman color prints are often unwatchable, having turned turned to a hideous orange hue).

GPS

10-Sep-2009, 08:38

sanking

10-Sep-2009, 14:48

Joanna Carter

10-Sep-2009, 15:04

What was your exposure time for the shots with the R, G and B filters? And how closely balanced were the separations after development?
The usual compensations I make are 3 stops for the red and green and 4 stops for the blue. I find, using Ilford Delta 100, the negs scan and combine very easily.

Rodney Polden

14-Sep-2009, 01:00

Fascinating and informative discussion. Thank you all for some very fine images.

Mike1234

14-Sep-2009, 10:10

...old Devin Tricolor (http://www.cameramanuals.org/booklets/devin_tricolor.pdf) camera. Purpose built for technicolor-like still photography...

It would be intriquing to build something like this to shoot 6x12cm with. It would be a heavy/bulky monster though.

sanking

15-Sep-2009, 15:52

It would be intriquing to build something like this to shoot 6x12cm with. It would be a heavy/bulky monster though.

I am in the end game of restoration of a big cousin (5X7 National Fotocolor) of the 6X9 Devin one-shot. Everything is finished but I am having a devil of a time aligning the pellicle beam splitters. The one to the green filter negative fell into place perfectly on the second try, but the one to the blue filter negative is giving me fits.

You can actually get much better results with a regular view camera making three separate exposures through Red, Green and Blue filters, but as in stitching multiple shots with a DSLR you must have a subject that will cooperate and lie still while you make the three exposures.

Sandy King

Mike1234

15-Sep-2009, 17:57

sanking

16-Sep-2009, 06:14

Sandy,

It doesn't surprise me you're having alignment issues... must be a real bugger. However, once it's pretty close you can scan the films individually and align in PS. The advantage of the splitter system is that the exposure is made to all colors silutaneously which eliminates the artifacts caused by subject movement.

Yes, if you can get the alignment close enough PHotoshop will merge them. At this point I am ok with the Red and Green records but the Blue is still too far off. As you say, the advantage is that all three shots are made at one time. The disadvantage is that a 5X7 one-shot camera is way larger than a typical 5X7 view camera, and you have to compose through the filters and pellicles, which costs more than four stops of light light loss (three for the filters, one+ for splitting the light into three parts.

But still, this is all quite interesting.

Sandy

jb7

16-Sep-2009, 06:22

Applying the same movements to all three must be a bit of a pain...

I was thinking that making a turret for the filters (I only have the screw in type)
might help speed things up a little-

Still mean to try this-

Arthur Nichols

16-Sep-2009, 06:37

I would think that using the same approach as described above in this post one could photograph a scene uising color film (or direct digital capture), scan it make all adjustsment is PS and split it into the three colors and then using a film recorder output each color to a separate black and white negative. I know this sounds cumbersome but it would provide a very stable long term storgae medium for color images. I personally don't like digital for long term storage. I am an engineer for my day job and sometimes have to access old CAD drawings. There can be problems opening them, they contain artifacts and sections which cannot be edited. I have had enough problems trying to access old digital information that when I project these issues out to 20 or more years it looks like big problems and lots of lost information.
I hope others feel that this is a relevant post and not an attempt to start another digital vs film dialogue.

Mike1234

16-Sep-2009, 07:07

Arthur,

Your post is definitely relevent.

Yes, one could use the R, G, and B channels from a direct digital capture to output to separation negs but the quality would be better just leaving the image digital. Creating separations for your color negs is a different story. But since I'm into hybrid photography anyway I would just scan mine. Regards to the file format becoming obsolete... since most digital photographic files are universal rather than proprietary to certain software there really are no worries regarding inability to open them long after they we made them. If an issue comes up with a particular format we'll have plenty of time to deal with it. That said, there certainly are proprietary image formats such as Adobe PSD. Just make sure your images are all converted to up-to-date formats. Batch processing should make this a quick/painless task.

Joanna Carter

16-Sep-2009, 07:10

Everything is finished but I am having a devil of a time aligning the pellicle beam splitters. The one to the green filter negative fell into place perfectly on the second try, but the one to the blue filter negative is giving me fits.
Sandy, have you thought of using dichroic mirrors to split the light based on colour? then you would not need the filters as well.

sanking

16-Sep-2009, 11:21

Sandy, have you thought of using dichroic mirrors to split the light based on colour? then you would not need the filters as well.

I have not thought of that and have no information about the subject. Can you direct me to a source for these type of mirrors?

Sandy

Joanna Carter

16-Sep-2009, 12:34

I have not thought of that and have no information about the subject. Can you direct me to a source for these type of mirrors?
The only way I found out about them was to Google "dichroic mirrors". I really don't have a source for them yet.

dh003i

17-Sep-2009, 00:51

I wonder if there is an alternative way to do something like this. The goal is to get the increased tonality, reduced grain, and increased sharpness of a B&W, while adding color.

So, here's my idea for an alternative:

1. Take 1 exposure with your favorite color film (I like transparencies).
2. Take the same exposure with your favorite B&W.
3. Scan both in.
4. Since all B&W photography could thought of as photographing the brightness of something, use the B&W file as a template for how bright pixels in the color file should be.

The result: information about tonality/brightness would have the characteristics of the B&W film, while the colored information would have the color film feel.

I'm not sure how one would go about doing this in Photoshop or GIMP (I use GIMP).

Mike1234

17-Sep-2009, 03:30

Drew Wiley

17-Sep-2009, 13:43

Couple of thoughts here. If one was inventing a tricolor camera from scratch the
best way would be a beamsplitter prism which could be specially coated. It would
weigh quite a bit but solve a lot of alignment issues and be far more precise than
pellicles (all hypothetical, since no one will want to put out the machinst money).
The second issue is that yes, you can more easily make tricolor images digitally, and
a few people do, but digital capture still does not have the dynamic range of certain
black and white films, so is no real substitute at this stage of the game. For distant
shots without parallax issues, you could simply gang three MF slr's together, exactly
aligned, and trigger them all at once atop a tripod built like a tank. Minor parallax
adjustment could be realigned in PS. In the studio it's a lot simpler. A precision
registered back from an old process camera and a filter turret from an early Sinar
digital camera, old Durst enlarger, or homemade would be just the ticket. People did
this routinely at one time with still lifes for dye tranfer and carbro printing.

dh003i

17-Sep-2009, 14:24

Maybe if you replace the Lightness Channel with your B&W scan after converting the color scan to Lab Color. There will still be noise in Channels A & B... purhaps a little Gaussian Blur in the color channels to address that. You're not going to approach the quality possible with tri-color shooting but it might be a fun/interesting thing to try.

Thanks. To test out this method, I went into GIMP and picked up a photo that I'd taken with my DSLR. I decomposed into L, A, B components, then did a huge gaussian blur (40 pixels) on the A and B channels (30 pixels). I did this to try to simulate the loss of resolution you get going from B&W 330 l/mm film (say Rollei Ortho 25) to Kodak E100VS. I then recomposed.

The results aren't bad. In images without a lot of different contrasting colors, it works very well, and it is difficult to tell the difference. Most of the information that we see, that determines sharpness to our eyes, it seems, is in the luminosity.

However, in images with a lot of different colors -- like a macro shot I have of a leaf during the fall -- you can see the effect when you look at it 1:1. It is sort of like a chromatic aberration + soft focus effect. Details regarding brightness are there, but at the edges of things -- particularly fine detail -- you see a kind of "color spill". You also see some loss of "color contrast".

See this comparison of the macro shot of the leaf (actually, a crop of that shot). (http://www.tabblo.com/studio/stories/shared/32823/ypgqr8mecx3fw90)

PS: Kodak E100VS has a resolution of 120 l/mm, Rollei Ortho 25 has 330 l/mm, so that B&W film is 2.75x as sharp as the E100VS. A pixel blur of 3 would probably have been more appropriate.

In any event, at that level of blur, the color bleed would probably hardly even be noticeable, as it was with my 40-px blur. But then it'd have to be blurred again, I suppose, to account for the additional grain, as you say. However, that probably wouldn't be too much more blur (I doubt it'd be equivalent to a 40-px blur).

Mike1234

17-Sep-2009, 15:16

dh003i

17-Sep-2009, 15:27

^^^ The "possible gains" will be in taking a separate B&W high rez image in addition to the color one and using that B&W image to replace the Luminosity Channel. The gaussian blur should not be so great... just enough to hide the noise from the color scan.

Again though... I don't know if there are any "gains" to be had with this procedure.

Yes, I understand that. I was just doing a little test on a digital capture from my DSLR that I already had. I wanted to see, in the worst case, how much of the information that constitutes "sharpness" is in the luminosity channel vs. the A and B channels. It seems that most of the information is in he L channel.

Thus, regarding the procedure you suggested -- a B&W negative and a color slide of the same thing, replacing the L channels from the color slide with the B&W negative -- it seems that this wouldn't result in much loss of IQ vs. using three color-masked B&W negatives and merging them to make a color photo.

Does that make sense or am I off base?

PS: My point in decomposing an already complete image, and Guassian-blurring the A & B channels wasn't for the same reason you suggested. It was to simulate the lower resolution of the color film, and then recombine that with he luminosity channel (which would be the B&W film). (it was actually overly aggressive in this context, as I think only a 3-6 pixel blur would be necessary to simulate the loss in resolution in going from 330 l/mm to 120 l/mm).

Don Hutton

17-Sep-2009, 18:10

Don't confuse the theoretical resolution of emulsions with what you could expect to put onto film under real world shooting conditions - you are almost never going to exceed 200 l/mm (100 lp/mm) - and you're simply never getting there with large format and with the very best medium format optics only very rarely with perfect conditions/technique and perfect/ideal everything else. You can get there with 35mm with the right gear/technique and some good luck, but the spectral sensitivity of the film you'd be using is hardly ideal.

Secondly, you're going to have to get that information back off the film too - you will lose something even with the finest drum scanner ever made. Work out the potential maximum resolution a perfect scanner could extract at 4000SPI - and the 100% efficient scanner does not exist - and you're still getting back down towards the resolving power of the best transparencies.... If you really want it all, just shoot 8x10 transparency and have it drum scanned - I don't believe that there is a better solution.

dh003i

17-Sep-2009, 20:34

sanking

18-Sep-2009, 10:12

Let's put this in some perspective.

The main reasons IMO one might be interested in making color separations on B&W films are.

1) Increased color fidelity in that the dyes of color negative and color transparency film are bypassed.

2) Greater sharpness and finer grain for the same size format compared to color negative and color transparency material.

3) Much greater dynamic range than available either with digital or with color negative or color transparency film.

4 Just for the fun of it as the methodology puts us in touch with the earliest pioneers of color photography.

Obviously making color separations on B&W film is only practical when the subject is static and one has the time to change the filters and film between the Red, Green and Blue filter exposures. However, this type of photography is every bit as feasible and practical as making making multiple exposures for stitching as many are doing with a DSLR, or for making long exposures with scan backs.

In fact, I think one could make a good argument that making huge files for large color prints is every bit as practical with a view camera making three exposures on B&W film as exposing the scene with a Better Light scanning back. For example, the Better Light Model 6000 E-HS, requires a minimum time of 35 seconds to produce a 144 mp file (10200 X 13600). In practice one could make three exposures with a 5X7 view camera through R, G and B filters on B&W film in less time, and scan the negatives ate 2540 spi to produce a 225 mp file. Merging the files in Photoshop to make a full color image is fairly routine so a film work flow is not at all unreasonable. In fact, with a static subject one might be able to get as much as 3000-4000 ppi from a sheet of fine grain 5X7 B&W film like Acros or Tmax-100, which in turn would give a image file of 315 mp - 560 mp. With static images and big vistas these kind of figures are quite realistic.

Sandy King

Joanna Carter

18-Sep-2009, 10:41

...lots of good stuff...
With which I totally agree :D

Paul Kierstead

18-Sep-2009, 12:07

In fact, I think one could make a good argument that making huge files for large color prints is every bit as practical with a view camera making three exposures on B&W film as exposing the scene with a Better Light scanning back.

I'd also suggest it is as practical as doing a stitched shot, which lots of people do all the time.

Actually, with something like a Grafmatic, you could have those 3 shots done *really* quickly. Still won't help with wind and trees, though.

dh003i

18-Oct-2009, 20:42

These three filters are used for color separation so I knew they would work best. At first I tried 25A, X1 and 80A filters but my yellows were very washed out. The 80A doesn't filter enough red and green. I haven't tried the 25A, X1 and 47B but it might work well enough.

Since I would prefer to use the Cokin creative filter system and slide the filters into the slots, what filters would be best from the Cokin creative filter system (http://www.cokin.com/filtres.html?=#004)?

They have

Red; Ref. 003
Green; Ref. 004
Blues:

Normal Blues (darker to lighter)

Blue (80A); Ref. 020
Blue (80B); Ref. 021
Blue (80C); Ref. 022

Dark Blue

Blue (82B); Ref. 024

Marko

18-Oct-2009, 21:08

Since I would prefer to use the Cokin creative filter system and slide the filters into the slots, what filters would be best from the Cokin creative filter system (http://www.cokin.com/filtres.html?=#004)?

They have

Red; Ref. 003
Green; Ref. 004
Blues:

Normal Blues (darker to lighter)

Blue (80A); Ref. 020
Blue (80B); Ref. 021
Blue (80C); Ref. 022

Dark Blue

Blue (82B); Ref. 024

Red and Green are OK, but they don't make blue filter suitable for color separation. But Hitech makes the entire color separation set in Cokin P size (http://www.bhphotovideo.com/c/search?ci=4032&N=4294540732+4291417993+4294954292) - you can order them from B&H (http://www.bhphotovideo.com/c/search?ci=4032&N=4294540732+4291417993+4294954292). I already had Coking Red and Green so I bought only blue, but if I were staring again, I'd get the entire set.

Hope this helps.

Marko

dh003i

18-Oct-2009, 21:30

Red and Green are OK, but they don't make blue filter suitable for color separation. But Hitech makes the entire color separation set in Cokin P size (http://www.bhphotovideo.com/c/search?ci=4032&N=4294540732+4291417993+4294954292) - you can order them from B&H (http://www.bhphotovideo.com/c/search?ci=4032&N=4294540732+4291417993+4294954292). I already had Coking Red and Green so I bought only blue, but if I were staring again, I'd get the entire set.

Hope this helps.

Marko

Thank you very much for your suggestion. This will work, I think, for my 135mm, 203mm, and 305mm lenses. However, I also have a wide-angle 90mm lens, with lots of coverage; so I'm worried the Cokin P's may vignette my Nikkor-SW 90/4.5.

dh003i

19-Oct-2009, 22:14

The usual compensations I make are 3 stops for the red and green and 4 stops for the blue. I find, using Ilford Delta 100, the negs scan and combine very easily.

Let me see if I'm understanding this correctly...

Lets assume we're using Fuji Neopan Acros 100, which doesn't require reciprocity compensation until 128-1000 seconds (then it only needs +1/2 stop, or a 1.414x increase in exposure time).

So, lets assume that our large-format shot requires 1 second at ISO 100 for a regular exposure.

Hence, the RGB-filtered exposures would require +3 stops for red and green (2^3 = 8x) and +4 stops for blue (2^4 = 16).

That results in exposures of 8s for red, 8s for green, and 16s for blue. So what would have been a 1s exposure without filters takes at least 32s when split into 3 exposures with filters. This is correct?

Now, if we use a film that requires reciprocity corrections beyond 1/2 second, those times greatly increase. Adox Pan 25 requires +1 for readings of 1s (so 1s => 2s), +1.5 for readings of 10s (so 10s => 28s), and +2 for readings of 100s (so 100s => 400s). I've done a regression, and this works out to exposure compensation (as a multiple factor) of 2^[1.1 * (t - 0.5)^0.13].

So, if we take that 1 second exposure reading at ISO 100, it becomes 4 seconds at ISO 25. Adding +3 stops for red and green, that becomes 32 seconds (4s * 2^3); adding +4 stops for blue, that becomes 64 seconds (4s * 2^4). But then reciprocity failure would have to be accounted for, so the 32s becomes 105s [2^(1.1 * 31.5^0.13)], and 64s becomes 237s.

Don Hutton

20-Oct-2009, 13:55

Check this guy's tricolor work out... http://www.loc.gov/exhibits/empire/

jon.oman

20-Oct-2009, 14:38

sanking

20-Oct-2009, 14:45

Lee Red Filter (http://www.bhphotovideo.com/bnh/controller/home?O=workaround.jsp&A=details&Q=&sku=102724&is=REG)

Lee Green Filter (http://www.bhphotovideo.com/bnh/controller/home?O=workaround.jsp&A=details&Q=&sku=102728&is=REG)

Lee Blue Filter (http://www.bhphotovideo.com/bnh/controller/home?O=workaround.jsp&A=details&Q=&sku=102726&is=REG)

I assume these three 3x3 filters will work?

Jon

Check the numbers. You should have a Red (25 or 29), Green (58 or 61) or Blue (47 or 47b). In the past 25, 58 and 47 was considered the correct set for in-camera separations from liffe, 29, 61 and 47b the set for making separations from transparencies. The difference between the two sets is that the first (25,58 and 47) are broader band filters. Lee does not appear to have a 47 in their line, only a 47b.

Sandy King

jon.oman

20-Oct-2009, 16:17

Check the numbers. You should have a Red (25 or 29), Green (58 or 61) or Blue (47 or 47b). In the past 25, 58 and 47 was considered the correct set for in-camera separations from liffe, 29, 61 and 47b the set for making separations from transparencies. The difference between the two sets is that the first (25,58 and 47) are broader band filters. Lee does not appear to have a 47 in their line, only a 47b.

Sandy King

Thanks for the info. These are 25, 58, and 47b. The cost is not bad either.

Jon

Don Hutton

20-Oct-2009, 16:37

Have a look around for Sinar filters - they used to make a 25, 58 and 47 in 100mm square resin filters which conveniently fit Lee holders. I bought a set on Ebay a while back which were unused.

Eirik Berger

26-Dec-2009, 14:40

[QUOTE=Eirik Berger;505596]Here is my first attempt in tricolor work. T-max 400 using a 6x9 rollfilm holder and 25, 58, 47 filters.
It is of a walrus skull taken on Svalbard earlier this summer. This was great fun and I am happy with the result. And the subject haven´t moved in 400 years so that did not cause any problems at all.

Eirik,

Great work.

What was your exposure time for the shots with the R, G and B filters? And how closely balanced were the separations after development?

I am also thinking of doing some separation work with a 6X9 camera so your technical notes would be appreciated.

Sandy King

Sorry I haven´t been around the forum for a while. Since this was my very first attempt and I did not remember exactly how much to compensate the green and blue filter (the red one I am familiar with :) ), I went for a three stop compensation for all three filters. The plan was to evaluate the negatives when I came home. Both red and green looked ok, but the blue negative was underexposed. When merging the scans I needed to make a pretty steep curve for the blue-filtered negative. I have not had the opportunity to refine the technique since we entered the dark season in october. The sun is returning in march, so I will work more with this way of imagemaking then.

I guess I will do some testing with grey patches (I use this one a lot with digital photography http://www.qpcard.se/BizPart.aspx?tabId=31&prod=1&catId=1) to adjust the compensation factor for each filter. I will try to get a color balanced set of negatives right out of the jobo-drum :)

I made a large (ca 60x90 cm) inkjet print of this image, and the texture and colors look great. I should have worked more with aligning the layers though. Color "fringing" is quite noticeable around the edges of the scull. I will do some more work on the digital image and make another print. It is amazing to see the typical BW-texture and sharpness in a color print. It sure beats scanning trannies, which I normally have done when making color prints.

I liked working with the 6x9 back, because it is never any doubt which negative is with what filter. The first is unfiltered, and the three following negatives are with red, green and blue filters. And with 8 exposures on each film you get 2 images on each roll. Neat.

David Hedley

12-Jan-2010, 05:06

This looks fascinating, and I'm going to give it a go with a box of TMax400 that needs to be used. Can I ask a couple of basic questions;

- how are the negatives scanned? In other words, are the three negatives scanned as greyscale, or is the Red negative scanned to the R channel, the Blue Negative to the B channel, etc?
- if scanned as greyscale, how are colours then derived?
- if scanned as R, G and B, are the three separate scans simply combined onto a 'rough' colour image which can then be adjusted using normal photoshop tools?

Sorry for the basic questions, but I know relatively little about photoshop and colour, as most of my work is in normal monochrome photography.

Noeyedear

12-Jan-2010, 07:34

Emmanuel BIGLER

13-Jan-2010, 01:34

Can I ask a couple of basic questions;

Hello from France !
May be you'll find answers to your question in Henri Gaud's blog
Henri Gaud lives near Paris and is one of the proselytes for this strange technique ;)
This is the google translation from French, I hope it will be useful

About combining the 3 B&W scans into a single coulour image
http://translate.google.com/translate?hl=fr&sl=fr&tl=en&u=http%3A%2F%2Ftrichromie.free.fr%2Ftrichromie%2Findex.php%3Fpost%2F2008%2F02%2F21%2F624-assemblage-avec-la-mthode-dite-des-calques

Henri Gaud's blog translated by google, with many examples of images & links to other aficionados of this technique.

http://translate.google.com/translate?hl=fr&sl=fr&tl=en&u=http%3A%2F%2Ftrichromie.free.fr%2Ftrichromie

Note that you do not need any computer if you have access to a RA-4 coulour paper processing machine.
This is nothing new and was alredy used in the past in the printing industry.
After several years doing tricolor images by computer, Henri Gaud has recently re-played the old game of coulour additive printing with an enlarger! 0% digital ! 100% photo-chemical and optical !!
And it seems easier than expected !

http://translate.google.com/translate?hl=fr&sl=fr&tl=en&u=http%3A%2F%2Ftrichromie.free.fr%2Ftrichromie%2Findex.php%3Fpost%2F2010%2F01%2F03%2FTrichroR4A

http://translate.google.com/translate?hl=fr&sl=fr&tl=en&u=http%3A%2F%2Ftrichromie.free.fr%2Ftrichromie%2Findex.php%3Fpost%2F2010%2F01%2F03%2FTrichromies-et-R4A-Chapitre-III

It the google translation, simply replace the funny "drawing" by "printing"
In French, "tirer" means both "to draw" like a drawer and to print like "tirer une image"

David Hedley

13-Jan-2010, 04:24

Alors, Emmanuel, un grand merci pour votre réponse. Les liens vous avez fournis sont très utiles, et il ya encore un plus qui définit une approche que je vais essayer d'abord, après que je reçoive un filtre bleu (on l'espère dans quelques jours!);

http://photo.guigue.free.fr/1_new_07_2007/trichromie/trichromie.htm

stephenhunter57

13-Jan-2010, 09:21

Yep works great but it is a lot of messing about, we did it in college amonst lots of other really interesting stuff

r_a_feldman

1-Feb-2010, 11:48

Just got this link through the Internet Directory of Camera Collectors list. It has some nice pictures of the 1929-vintage german "Bermpohl Naturfarbenkameras" and a discussion of printing the B/W negatives: http://www.vintagephoto.tv/color1.shtml. The owner of the site also has started a Yahoo group on the subject, at http://tech.groups.yahoo.com/group/history_of_color_photography/.

Bob

Paul Kierstead

22-Feb-2010, 19:26

Just tried my first one (shh, 6x9, not LF, but close enough for fooling around). I was unable to align the three layers well enough; clearly the camera moved enough to introduce distortion, since when one part, for example, was aligned perfectly something higher up would not be. Rotation could also not correct it. Aperture was the same for all shots.

I think the problem is my filter holder; I am using polyester filters by Lee, in holders, slid into a holder. They are a pretty tight fit and a little bit of a battle. On my lightweight CF tripod, it seems destined to move the rig (I suppose I could weight it down).

Do you just hold the filter in front? They were long exposures, I forsee it waving about a little. Not sure if that matters. Or should I just weigh down that tripod, lock the head really well and slide those filters in an out?

If I am using film holders especially, I could see that easily introducing very small movements too.

Emmanuel BIGLER

23-Feb-2010, 02:33

Just tried my first one ....
Do you just hold the filter in front?

Hello from France !
Yes Paul in the 3-color technique the main limitation is unwanted displacements of the camera between the 3 shots.
So it is easier to achieve with a rollfilm holder and a camera solidly mounted on a good tripod. Hewever in his first attempts Henri Gaud started by the 8"x10" format ! Outch ! But he used a very heavy thripod :)

However you do not need to attach the filters to the lens.

A good friend of mine, Philippe Domingos from Nancy, France...
http://www.philippedo.eu/wysiwyg/
...uses the same gelatin filters as advertised by Henri Gaud on his blog. Size is about 10cm (4") square
Philippe has made a kind of a cardboard-mounted "booklet" of filters and he simply holds the "booklet" open to proper the Red, Green or Blue "page" in front of the lens, simply hand-held, no screw, no support of any kind.
In fact gelatin filters are very thin and they do not need to be perfectly positioned perpendicular to the optical axis. In fact thin gelatin filters can be used slanted without damaging the image sharpness too much.
By not touching the camera when switching filters you avoid 3 possibilities of displacing the camera during the exposures. However you still have to wind the film on and touch the back... except with a motor-driven rollfilm back.

About long exposure times, some interesting portraits have been achieved by the 3-color-on-B&W-film process but with a medium format camera.
By H. Gaud with a motor-driven Rollei 600x: http://trichromie.free.fr/trichromie/index.php?post/2007/02/22/114-enfin-un-portrait

By L. Askienazy: http://trichromie.free.fr/trichromie/index.php?post/2009/06/29/Trichromie-by-Laurent-Askienazy
(for this image, some digital post-processing was required for a fine registration of the 3 layers : the "making of" is here: http://trichromie.free.fr/trichromie/index.php?post/2009/07/01/L-Affaire-Askienazy)

Philippe Do

23-Feb-2010, 06:42

Hello,
Thank’s Emmanuel for the description.
You will find here two picture for illustration.

It’s easy to use. :-)

http://www.philippedo.eu (http://www.philippedo.eu/nimp/DSC00092.JPG)
http://www.philippedo.eu (http://www.philippedo.eu/nimp/DSC00093.JPG)

Paul Kierstead

23-Feb-2010, 19:28

Well I was able to align them enough for some practice to get a flavour of it. I did find out that the blue filter definitely needs a little more exposure. Very ugly, but a start!:

http://homepage.mac.com/paulkierstead/Sites/Gallery/trichrome-test.jpg

(excuse the MF image, but it is certainly in the spirit of the site) Alignment is still less than perfect, especially around the flowers (almost to the extent that I swear they moved). I still need to work on the correction. But it definitely gives a different, interesting look.

gth

11-Jun-2010, 22:40

Where can I find separation filters that fit into the filter slot of a 6 In Goertz Artar?
This was originally used to make color separation films, right?

Alan S

21-Jun-2010, 19:05

Thank you so much for the interesting thread. I really would like to try this but as I have PS elements 6, I have been having difficulty getting it easily done. Anybody tried this using elements?

Alan

Henri Gaud

21-Jun-2010, 23:24

Where can I find separation filters that fit into the filter slot of a 6 In Goertz Artar?
This was originally used to make color separation films, right?

You are right.

You can go here : http://www.leefiltersusa.com/lighting/products/colors/

Dr Klaus Schmitt

24-Jul-2010, 17:18

I use a similar technique, but don't restrict myself to the visible spectrum. Here blue = ultraviolet, green = visible, red = infrared.

http://www.pbase.com/kds315/image/93179123/original.jpg

gbogatko

24-Jul-2010, 21:53

Here's my first successful attempt.
Cheapo that I am, it's on 120 film.
I really like the result -- especially the extended tonal range.

George

numnutz

25-Jul-2010, 05:10

just to go a bit further with this process

- I wonder if you take pictures through the red and green filters only, you could simulate the blue exposure by adding together the red and green exposures. Then make the blue layer white, and then subtracting the red/green away from the blue (exposed white). making a duo-chromie...

I hope I have explained it satisfactorily what I see is being done is synthesizing the blue exposure.

nn :)

gbogatko

25-Jul-2010, 09:12

Early technicolor was a two-film process.
What colors did that early one use?

gbogatko

25-Jul-2010, 09:15

just looked it up on wikipedia -- first process was red and green.

Henri Gaud

25-Jul-2010, 09:38

just looked it up on wikipedia -- first process was red and green.

Red and green,
It was for anaglyphe.

The first process was trichromy by Ducos du Hauron,
And the filters are Green, violet and orange, in 1868.

gbogatko

25-Jul-2010, 15:09

I love this discussion. Just fascinating.

jnantz

25-Jul-2010, 15:21

sorry for my ignorance
but can this be done with orthochromatic
film ( or paper negatives ) or is this a strictly
panchromatic film sort of thing ?

thanks!

john

dh003i

25-Jul-2010, 23:08

I think that orthochromatic film doesn't capture enough of the red spectrum. You need panchromatic film. Check out these two threads

Efke vs. Adox films (http://www.largeformatphotography.info/forum/showthread.php?t=53190&highlight=adox)
Efke PL 25 M datasheet (http://www.largeformatphotography.info/forum/showthread.php?t=53654&highlight=adox)

The upshot of it is that if you want an old-style high-resolution film with great tonality, which works with tri-color, try Adox Pan 25 (http://www.freestylephoto.biz/12945-Adox-Pan-25-iso-25-4x5-25-sheets?sc=24100). It is widely available. (You could also do Adox Pan 100 or Efke 100).

Henri Gaud

25-Jul-2010, 23:11

sorry for my ignorance
but can this be done with orthochromatic
film ( or paper negatives ) or is this a strictly
panchromatic film sort of thing ?

thanks!

john

If you want to be realist, you take tricolor with filter Red, Green, Blue, and panchromatic film
If you want make your picture, you can take orthochromatic film and filter green and blue.
If you want make bichromy, you can take panchromatic film and orange and cyan filter.

You have many explain here :

With google english : http://translate.google.com/translate?hl=fr&langpair=fr|en&u=http://trichromie.free.fr/trichromie/&prev=/translate_s%3Fhl%3Dfr%26q%3Dhttp://trichromie.free.fr/trichromie/%26sl%3Den%26tl%3Dfr

http://trichromie.free.fr/trichromie/index.php?category/In-english

In french : http://trichromie.free.fr/trichromie/

You can ask me for any thing about trichromy

http://trichromie.free.fr/trichromie/index.php?post/2007/01/31/89-proposition-de-travail

Have a good day ...

gbogatko

26-Jul-2010, 07:16

Three-strip Technicolor is explained here:

http://en.wikipedia.org/wiki/Technicolor#Three-strip_Technicolor

Especially interesting is the film stock used:

"Green light was recorded through a green filter on panchromatic film, while the other half of the light passed through a magenta filter and was recorded on bipack film stock with two strips running base to base. On this stock, the front film was sensitized to blue light only, backed by a red gelatin layer which acted as a light filter to the panchromatic film behind it. This process accurately reproduced the full color spectrum and optically printed using a dye-transfer process in cyan, magenta and yellow."

gbogatko

26-Jul-2010, 16:35

from last night -- this is fun!! (but tedious)

George

jnantz

26-Jul-2010, 18:47

i guess i will look for orange and cyan filters :)

thanks !
john

Henri Gaud

26-Jul-2010, 23:08

i guess i will look for orange and cyan filters :)

thanks !
john

http://trichromie.free.fr/trichromie/index.php?post/2007/04/26/235-la-bichromie-expliquee

I can send you cyan and orange filter if you tell me your adress.

gbogatko

27-Jul-2010, 19:08

Did anyone else watch that Curtis Tri-Color on ebay skyrocket to $2034?? Gulp.... Obviously a snipe battle.

Henri Gaud

27-Jul-2010, 23:06

Did anyone else watch that Curtis Tri-Color on ebay skyrocket to $2034?? Gulp.... Obviously a snipe battle.

I see,
And I stop at 1550 $

I have an other Curtis :
http://trichromie.free.fr/trichromie/index.php?post/2007/12/27/548-quatrime-achat
Not exactly the same, but similar Curtis Scout color 4x5

Henri Gaud

27-Jul-2010, 23:12

you can read 1942 new about one shot camera

http://trichromie.free.fr/trichromie/index.php?post/2010/07/23/OneShotCamera

Henri Gaud

29-Jul-2010, 02:45

you can read 1942 news about one shot camera

http://trichromie.free.fr/trichromie/index.php?post/2010/07/23/OneShotCamera

Eirik Berger

2-Aug-2010, 07:17

I would really like to take a look under the hood on this one:
http://www.collectiblecameras.com/product.php?productid=169958&cat=0&page=

Really sweet... :)

emo supremo

5-Aug-2010, 18:21

I haven't a clue whether or not this is a consideration but after 18 pages (which I haven't read totally) it might be so here goes: has anyone with savvy given thought xray films offered in green sensitive and blue sensitive emulsions. Scratches VERY easy when tray developed but cheap. Not that 'everyone' travels about with this in their pockets but is this relevent? Emilio

dh003i

4-Nov-2011, 04:42

Does anyone have suggestions on what to use with the Cokin X-Pro system (170mm x 130mm filters)?

I bought Hitech 85mm RGB filters HI25CP (http://www.bhphotovideo.com/c/product/158988-REG/Hitech_HT166485_85mm_Red_25_Resin.html) Red #25, HI47CP (http://www.bhphotovideo.com/c/product/158990-REG/Hitech_HT167085_85mm_Blue_47_Resin.html) Blue #47, and HI58CP (http://www.bhphotovideo.com/c/product/158991-REG/Hitech_HT167385_85mm_Green_58_Resin.html) Green #58 for the P-system, but I can't seem to find corresponding filters for the Cokin X-PRO.

For the X-PRO sized filters, I can only find...

Cyan, Magenta, and Yellow color-compensating filters (http://www.bhphotovideo.com/c/search?atclk=Rectangular+Sizes_Cokin+X-Pro&ci=110&N=4256189601+4254301308) -- these are the only ones that appear to have the complete set in the right size

A variety of resin filters, but no blue (http://www.bhphotovideo.com/c/search?atclk=Rectangular+Sizes_Cokin+X-Pro&ci=113&N=4256189599+4254301308)...

and some very expensive Tiffin 5x6 in Blue #47, Green #58, and Red #25 filters (http://www.bhphotovideo.com/c/search?atclk=Filter+No.%2fColor_%2358+Green&ci=113&N=4256189599+4294951821+4225042081+4225042075+4225042070)...but the size for these appears to be wrong. Cokin X-PRO takes filters that are 130mm wide. These filters are 5 x 25.4 = 127mm wide. I presume that won't quite work?

Henri Gaud

5-Nov-2011, 07:42

You can take Lee Filter filter, you have sheet 21"x48" and 25'x48", i think it-is sufficient.

http://www.leefilters.com/lighting/products/colours/ref:C4630710C51307/
http://www.leefilters.com/lighting/products/colours/ref:C4630710C5EDC4/
http://www.leefilters.com/lighting/products/colours/ref:C4630710C570C5/

A+

http://trichromie.free.fr/trichromie/index.php?category/Pour-commencer

dh003i

5-Nov-2011, 12:10

Good afternoon Henri,

These seem like they are disposable gel filters, if I am not mistaken. Would you then use them for one session, as they can easily become damaged? And how would you hold them in the proper place? Looks like you'd use the gels with these:

http://www.bhphotovideo.com/c/search?atclk=Brand_LEE+Filters&ci=3837&N=4247996974+4291378429

Henri Gaud

6-Nov-2011, 04:20

Hello,
You can use Do-it-your-self with cardboard and cutter.
And this filter Lee can by use during many days, and if you have un trouble, you can change, they are not expensive at all.

Juste do it ;-)

swilf

15-Nov-2011, 21:52

Does anyone have suggestions on what to use with the Cokin X-Pro system (170mm x 130mm filters)?

I've got some pleasing result with Cokin 003 (red) and Cokin 004 (green). Had no chance to compare them to the recommended #25 and #58, but the picture looks good:

http://farm7.static.flickr.com/6175/6191468837_23b53a936e_z.jpg

(sorry, it's MF)

As you already mentioned, there's no blue filter in the system, so I had to buy Hitech #47, but at least 2 of 3 filters are cheap. Anyway, spectral sensitivity of the film in use should be much more of a concern.

dh003i

16-Nov-2011, 20:23

Swilf,

This is a beautiful result, thank you for posting it. You say that you use a Hitech #47 with the Cokin X-PRO system. However, the X-PRO system takes filters 130mm wide. I only see a 4 x 4in (100 x 100 mm) filter on BH Photo (http://www.bhphotovideo.com/c/product/100629-REG/Hitech_HT100BW47_4x4_Blue_47_Resin.html).

I've got some pleasing result with Cokin 003 (red) and Cokin 004 (green). Had no chance to compare them to the recommended #25 and #58, but the picture looks good:

http://farm7.static.flickr.com/6175/6191468837_23b53a936e_z.jpg

(sorry, it's MF)

As you already mentioned, there's no blue filter in the system, so I had to buy Hitech #47, but at least 2 of 3 filters are cheap. Anyway, spectral sensitivity of the film in use should be much more of a concern.

D. Bryant

16-Nov-2011, 21:32

I've got some pleasing result with Cokin 003 (red) and Cokin 004 (green). Had no chance to compare them to the recommended #25 and #58, but the picture looks good:

http://farm7.static.flickr.com/6175/6191468837_23b53a936e_z.jpg

(sorry, it's MF)

As you already mentioned, there's no blue filter in the system, so I had to buy Hitech #47, but at least 2 of 3 filters are cheap. Anyway, spectral sensitivity of the film in use should be much more of a concern.

So what kind of film did you use?

swilf

17-Nov-2011, 01:37

You say that you use a Hitech #47 with the Cokin X-PRO system. However, the X-PRO system takes filters 130mm wide.

Glad you like it. No, I didn't use Cokin X-Pro, I used Cokin P, but there should be no difference. After all, they are just sheets of plastic cut to size. And we are too far anyway from meeting Luther-Ives condition to care about filters' spectral densities, exact brands or Wratten numbers. Almost any deep red, blue and green should be OK. The deeper the better: dense filters produce saturated colors.

As for film, I used TMX. I suggest generous bracketing for red and for blue.

Corran

17-Nov-2011, 11:43

So this is really intriguing me and I have some ideas that I would love to do this with.

There is 19 pages of this thread though, so can someone sum up the pertinent info? Like, what the specific 3 filters you would need for this are and what filter factor to use for the film when using them? There was lots of discussion and it was a bit confusing as to what was right.

Ben Syverson

17-Nov-2011, 12:26

You really only need two filters. For example, you might have one Red image, one Green, and one RGB (panchromatic). You can derive the blue record with:
B = RGB - (R+G);

The blue filter is quite dark, which means you're either exposing for longer or living with a thin negative. This gets around that limitation.

Another thing to try would be orthochromatic (blue-sensitive) film for the B record.

Drew Wiley

17-Nov-2011, 13:59

A simple trick is to use TMX100 for R & G, but TMY400 for B. You will need to dev B
neg longer, but it's not too diffcult to match the curves. Also, a 47 filter is not too
bad and about a stop faster than the ideal 47B.

Corran

18-Nov-2011, 06:40

I will try just taking an R and G and subtracting it from an RGB exposure (which is convenient to have a "straight" b&w). Kind of reminds me of some of the audio tricks we use for stereo images in recording.

Do you think an X1 green filter will work instead of a stronger green filter? I don't have a #58 and don't feel like buying a $50 filter for simply testing this idea.

Thanks!

Tocal

18-Nov-2011, 10:21

Hi everybody.

I finally managed to read the whole thread. Separation is indeed a very interesting technique.

Now, a Sinar 4-lens board with lenses crossed my way accidentially and I wasn´t able to resist. I´ll try to start one-shot seperation with it on 4x5. I don´t expect the Schneider 125/8 Radionars to be astounding, but it is a start. Hopefully I can manage to get rid of the parrallax-error using photoshop.

I will report here once my project is going to start.

Best regards,

Tobias

D. Bryant

20-Nov-2011, 08:49

I've got some pleasing result with Cokin 003 (red) and Cokin 004 (green). Had no chance to compare them to the recommended #25 and #58, but the picture looks good:

(sorry, it's MF)

As you already mentioned, there's no blue filter in the system, so I had to buy Hitech #47, but at least 2 of 3 filters are cheap. Anyway, spectral sensitivity of the film in use should be much more of a concern.

Just noticed that your background isn't a uniform color. Interesting.

swilf

20-Nov-2011, 19:44

Just noticed that your background isn't a uniform color. Interesting.

Yes, this is a mistake in matching all three curves at highlights. Shadows are slightly greenish too, I should have been more accurate. Hint: include Q-13 when possible.

Drew Wiley

21-Nov-2011, 11:11

Corran - an X1 would be useless for this purpose. It lets way too much of the other
primary colors through.

Corran

1-Dec-2011, 15:48

Just now saw this. Thanks, I will look for a cheap #58.

jonreid

15-Oct-2012, 01:24

My first attempt. Stupidly chose a subject with subtle colours. It's not quite right. The timber background was a real chocolate brown, and my daughters trashed dancing shoes a dirty soft pink with what I remember were lime green reflections off some off the little hex sparkly bits....

Jon

82090

David Schaller

15-Oct-2012, 16:50

But still nice! Keep posting. This is great.

Zaitz

16-Oct-2012, 05:19

I really think that has an excellent feel to it Jon!

I haven't seen it mentioned and have been through the thread twice. What values are people using for their color layers? The french translation link posted a few pages back gave me his channel mixer values for RGB. But I saw Sandy mention magenta, yellow, and cyan too. Is there a preferred way to overlay the colors? I tried this on a digital color image of mine and the result was fairly encouraging. I may have to give it a try with some black and white negatives. Makes me a bit less concerned about the future of color film. Obviously though movement might be a problem.

dwross

9-Sep-2013, 08:36

Hi All,

I don't know if this topic still holds interest. Everything about photography morphs so fast. But, even if tricolor separation with film isn't currently of wide interest, I can hope that it resonates with an individual or two.

As some of you may know (gawd know I talk about it enough :) ) I work with handmade silver gelatin emulsions -- first colorblind, then ortho, and now panchromatic. Adding pan to the first two, in a suite of three emulsion types, allows for separations as done back in the day for color carbon, etc. Anyway, here's the info. I'd love to see a chemical color photography artist take a swing at emulsion making.

The info about pan starts here, but there is more info leading up to it on the previous two pages. http://www.thelightfarm.com/cgi-bin/htmltutgen.py?content=02Sep2013

d

martinyanus

29-Oct-2016, 10:46

Hello guys!

Your thread here inspired me hugely, even if it happened only with some delay.
Some weeks ago I finally bought the right filters, exposed, developed and scanned the negatives, and... don´t know what to do next.
It´s quite embarrassing but I have no clue how to merge the three negs in Photoshop.
I´ve tried hard in the last many days but find no way to do it.
Could one of you tell me step by step (please!) how I can create a color image on a computer screen?
I´m sorry, I´m probably too analogue to solve this problem on my own...
I will really appreciate your help on that.

Kind regards,
Marcin

jnantz

29-Oct-2016, 11:18

hi marcin

http://photo-utopia.blogspot.com/2007/11/colour-images-from-b-film.html

it explains how to do it...

i have been taking black and white prints and making color separations in ps
and adding my own cmy levels and merging them in an open file's color channels
and getting great results.

have fun!
john

martinyanus

1-Nov-2016, 01:47

Hi John,

thank you for your quick answer.

I´ve been trying to follow the instructions you found on PhotoUtopia.
Unfortunately, even if I adjust the size of all three pictures the field "Merge Channels" remains inactive.

What I could do (and did) was "Split Channels", after which "Merge Channels" got activated.
But even then, that was it and I couldn´t procede to make a color image...

156846

*****

A few attempts later, with a little help from my girlfriend...

156847

Now, it looks like color!
I´m very happy with this result!
:-)

The problem (hopefully the last one) is that my scanner doesn´t crop half-frame negatives as it should.
The scans therefore have to be adjusted because at the moment the three layers simply don´t fit exactly, as you can see.
What is the best way to do this?
:-)

dwross

1-Nov-2016, 04:13

martinyanus

1-Nov-2016, 07:49

You're on your way! Here's the site with the merging technique I use: http://www.vintagephoto.tv/photoshop_assembly.shtml It includes all steps for moving and merging the layers seamlessly. It's actually a lot simpler than the number of steps makes it look like in the tutorial. It's just that he's has spelled out every little detail (always a good thing :)).

I'm close but... Ups...
Your link doesn't work.
:-o

+++

Some two hours and eighteen minutes later...
I thought, I´d try to adjust the pictures by cropping just a bit here and there at a bigger magnification...

Finally, after giving up color photography years ago as inherently grainless (and hence disappointing to me), I feel now, I´m on the right track towards an interesting rendition of grain in color:

Ladies & Gentlemen, this is Ilford Delta 3200 in Rodinal (APH09) 1+25 on an 18x24 mm negative.

:-)

156850

Corran

1-Nov-2016, 08:43

I know nothing about Photoshop CE, I have CS6 and CC. But if it's in CE as well, you should be able to go to FILE -> SCRIPTS -> Load Files into Stack, and then select your 3 scans and check the box that says "Attempt to Automatically Align Sources Images."

Once those three images are in the file and aligned, I personally just select the whole canvas and copy each layer into a new file, dropping them directly into the R G and B channels.

I've done this a few times, looking for very specific times when water is moving just the right amount with debris in it to make certain images, but it's difficult. Here's one I've done:

http://www.garrisaudiovisual.com/photosharing/colormotionfinished01ss.jpg

martinyanus

1-Nov-2016, 09:07

I'm afraid I have only this under FILE:

156856

The plants in the water look lovely.

Corran

1-Nov-2016, 09:18

Darn - but, did they potentially put it under the "Automate" tab? For a couple of years I taught a Photoshop Elements course and they did some weird things with the menus and UI compared to the full version, like moving or combining things in different places, which was really annoying!

martinyanus

1-Nov-2016, 09:52

If I recall correctly, this is the way Sandy mentioned he did this a couple of pages and a few years ago.
I tried to check it but as you see, my Photoshop skills are quite unsufficient, so I failed.

This is what I find under FILE > AUTOMATE:

156858

jnantz

1-Nov-2016, 11:44

i have cs2 and i find it easy just to put them carefully 1 ontop of the other using opacity to make sure they are aligned,
using the cropping tool to trim a little off so they align right afterwards, opening a new file and just dropping them in manually
into color channels. i do this with rgb filtered images as well as cmy files, i guess it takes a little extra effort but it works OK ..

Corran

1-Nov-2016, 12:04

At some point you could get Photoshop CS2 free, I found this tutorial here but I can't vouch for it:
http://www.ibtimes.co.uk/download-install-adobe-photoshop-cs2-free-legally-windows-mac-1445595

Might be good to try. I am pretty sure the Stack command is in CS2 - way back when I started using PS that is the version I used.

You can definitely manually do it (and jnanian's suggestion on using the opacity on the layer to help is spot on) but I figure it might be worthwhile to try installing CS2. They can probably run concurrently anyway in case you want to return to CE, whatever that is.

koraks

2-Nov-2016, 01:47

Last time I checked (which, granted, is quite some time ago) CS2 could still be downloaded. Note that you need to pull a little trick to get it to work on 64bit Windows:

1. When installing, change your path to C:\progra~2\..... and install. It should work.
2. Copy the installer to your local drive. Right click Setup.exe, Properties > Compatibility tab > Run this in compatibility mode. Choose XP SP2. Click OK. Run Setup.exe
(from here) (https://forums.adobe.com/thread/1069386)
Also note that CS2's window management doesn't mesh well with more recent versions of Windows (8.x, 10); I often get the various Photoshop Windows not being placed on top, or the opposite: remaining on top even if focus is shifted to another program. Minimizing the main Photoshop Window and then maximizing it again works, but it can be a bit annoying at first. However, I have found the program to be entirely usable.

DravenJohnson

16-Jul-2020, 13:03

I understand that you are trying to make a color photograph more black and white in the old stylistics, right? I would advise you to start by trying to look for various phone applications such as Photo lab, Light room or Capture One Pro they allow you to edit and enhance photos. If you plan to edit through a computer, then there are two choices: an online editor or a special program such as Photoshop. Personally i don't really know how to use Photoshop and that is why I prefer online services. In general, there are a huge number of services that allow you to customize and edit images online, but not every service can colorize black and white photos (https://imagecolorizer.com/), especially if they're old. I have my personal favorite service that allows me to colorize black and white photos in just a few seconds. They use a special AI technology that is constantly training to make the result as quick as possible.

Rod Klukas

16-Jul-2020, 14:55

That was really cool!
Are the 25A, 58 and 47B-filters the optimal combination?

Yes that is the optimal set for 3 color separations.

This was used for Dye transfer especially, as well. See the color work of Paul Outerbridge in the 1930's.

Rod

Bolshoi

29-Mar-2022, 02:19

Hello!
I've recently finished my year-long project. This is Curtis Color Scout 4x5 restored with dichroic beamsplitters. Works fine. https://live.staticflickr.com/65535/51967393077_229c6cff47_o_d.jpg
226065
226066
226067
(cant attach full size photo, use link https://www.flickr.com/photos/132081012@N03/51967393077/in/pool-92087504@N00/ )

martiansea

19-Apr-2022, 21:30

First try at TriColor! Been wanting to for awhile and finally got around to it
226648
Shot on 4x5 with my battered B&J using Foma100 developed in Legacy L110 dilution B. Lens is a CZJ Tessar.
For filters, I used an antique set of thick glass gel Kodak Wrattan filters: 25 Red, 58 Green and 47 Blue. I also included a straight unfiltered shot that I added a little bit of on top to get some extra shadow depth.
I scanned by photographing with my Nikon D800 with 60mm Micro-Nikkor. Raw files processed with DxO PhotoLab and final image composited with Affinity Photo.
If you look close, you can see a few little artifacts here and there from dust on individual frames. Was anxious to see how it looked, so didn't uptight about dusting or retouching.
Fun! :D