Fixed: Frame Grabs of Technicolor Test

[Brian Rose]

Thanks to the help of Mike Collier, I am able to show the frame grabs from my first experiment in creating natural color from black and white film.

 

 

 

 

 

I shot it with a bolex, using Plus X BW Neg. I shot each scene three times, one through a red filter, green, and finally blue. The camera was secured on a tripod, to prevent any movement between takes. I then tinted each color record in premiere pro, and combined them with a screen key. Of course, at this time, I am limited to static scenes, since I took the exposures successively. My next challenge is to figure out how to make simulataneous exposures to capture real action. But, I hope to demonstrate that the three color method is practical, it can be done, and the results are amazing!

 

Brian Rose

[nathan snyder]

Brian, This is awesome! Really very cool. Thanks for posting. I cant wait to show these stills to my friend who is actually restoring an old technicolor camera.

 

I have an idea for you for doing live action... simply gang the drive shaft of three bolexes. Kind of like this:

http://www.sci.fi/~animato/3d/3d2.html

You have probably already thought of that but I propose that you set all three of them up in a row side by side, and set the two on the out side on some sort of pivot so they turn inward for close focus and outward (parrallel to the center camera) for infinite focus. This wold be kind of like the parallax adjustment on old parrallel view finders on bolexes, auricons, and mitchells. What ever was focused on would have sharp color focus but there would be misalignments of things out side of the depth of field. I actually think this would only lend to the charm of this format, kind of like using anamorphic lenses were the stuff outside of the depth of field is either extra skinny(behind the subject in focus) or extra fat (infront of the subject in focus).

 

Just an idea. Again, really good stuff.

 

nathan snyder

Edited November 1, 2006 by nathan snyder

[Brian Rose]

Thanks for the link! I was thinking of something along those lines, but I hadn't figured out how to make it workable. Your link gives me an excellent start! So you know someone actually restoring a tech camera? Like a real, three strip camera? Awesome. What I would give to sit down and see the guts of one of those babies, especially that beamsplitter prism. Any info on that, or a website with updates? I'd love to learn more. Why is he doing the restoration...for a museum, or actual working condition...

Again, thanks for the help, and the encouragement. I'm going to do some more tests soon, and then, I'll post some screengrabs.

Best,

Brian Rose

[Hal Smith]

Thanks to the help of Mike Collier, I am able to show the frame grabs from my first experiment in creating natural color from black and white film.

Awesome pictures!

 

You might ask around if anyone's salvaged the beam splitting optics from an old, pre-CCD TV color camera - the image size on Plumbicons is pretty close to 16mm - for 35mm you'd look for the guts out of an old Image Orthicon camera like an RCA TK43 or 44. A lot of originally very expensive but elderly cameras end up on junk piles - someone's got something you can use for experiments.

 

What exactly did you use for separation filters? Your colorspace looks pretty darn good - you obviously hit the nail on the head in your choice of filters.

[Brian Rose]

There are several color filters available through Kodak's Wratten line. I used Red 24, Blue 47B and Green 61. With the filters, I overexposed the film by between 3.5 and 4 stops. In hindsight, I think I went a litte too far, as some of the images were a little blown out, especially the plant photo. Next time, I might try 3 stops or so. Thanks for the suggestion on the beamsplitter-i'll definitely have to look into it. I'm interested in experimenting with prisms, and with the three camera method suggested by nathan.

 

Best,

Brian Rose

[Dominic Case]

It's a great thrill to recreate colours this way!

 

You are searching for a way to eliminate the time slip between R, G, & B exposures. But what is even more interesting - creatively if not technically - is to exploit it. Note the fringing of the clouds that have moved between exposures. You would get some terrific shots of - say - waves breaking on a beach - or a fountain - or traffic moving along a freeway - or cornfields blowing in the wind - if you used your technique like this.

[K Borowski]

There was a famous photographer (Ulanov?) (forget the name at the moment), that went around pre-Soviet Russia photographing people using this same principle. I agree with Dominic that the fringing can add to the "feel". One of the shots this guy took involved water, and you got a very interesting effect in the water due to the water being in different places with the different color exposures. Some sort of beam-splitter would be best for this project. This method (with beamsplitter) actually has an advantage over technicolor in that you don't get that annoying color pulsation due to limitations in the accuracy of the dye imbibition process (it's almost like a slight color "flicker"; I notice it the most on faces). You also won't get the magenta halation effect that is the trademark of technicolor as well, but I find this too to be distracting.

 

Anyways, really good for you for actually moving beyond internet banter and actually shooting it.

 

Regards,

 

~Karl Borowski

[Phil Rhodes]

Hi,

 

What you could do is to shoot at some high frame rate (ideally three times normal, but anything would help) with a rotating colour filter synced to the camera, then separate out every third frame and do optical flow interpolation in post (twixtor, retimer, motionperfect) to make up the missing frames. I quite like combining old and new techniques.

 

Or even, if you like the fringing, don't do optical flow interpolation. Just stick them back together.

 

Phil

[Brian Rose]

You all raise some great points about using the "time slip" effect, and I'm trying to think of a project that will use this technique. The big challenge though, is trying to figure out how to capture action. These first tests were done in a crude manner. Essentially, what I did was wind up the Bolex as far as it would go, so I would get about 30 seconds at 24 FPS. I then shot ten seconds through a red filter, paused, inserted a green, shot another ten, and so on with the blue. So, there is a considerable time lapse between the three strips. At this point, I am down to three methods for achieving better coverage: 1) Use a color wheel, and shoot at triple the framerate. The problem is how to sinc up the color wheel.... 2) Use mirrors and/or prisms to capture the exact image 3)Finally, the one I lean toward, is using three cameras aligned to capture the same image. It won't be geometrically precise, but pretty darn close.

Brian

 

 

I forgot one other thing. Karl, the photographer you were talking about is named Sergei Prokudin-Gorskii. He made some really stunning images using a tricolor method. Looking at them, it's hard to believe they were shot nearly a century ago!

 

http://en.wikipedia.org/wiki/Sergei_Mikhai...rokudin-Gorskii

 

And thanks for the encouragement. I hope to have some new tests soon...

 

Best,

Brian Rose

[Leo Anthony Vale]

There was a famous photographer (Ulanov?) (forget the name at the moment), that went around pre-Soviet Russia photographing people using this same principle.

 

Would that be V.I.Ulanov AKA Lenin?

[K Borowski]

Would that be V.I.Ulanov AKA Lenin?

 

I didn't know that Lenin had a pseudonym. I wasn't referring to him, rather Prokudin-Gorskii, also a famous writer, no? As for Lenin, I think the only way he "advanced" photography was having his "friends" airbrushed out of a picture with him after he had them killed one by one. . .

[Nate Downes]

I've always wanted to shoot a technicolor-style picture, due to the technical challenges. Good luck to the guy rebuilding a camera, and let me know when it's ready for use. 8)

 

the best option is a beamsplitting prism with color-filtration. Aka, it splits away the green to go one way, blue to go the other. Panasonic and canon use such a prism for their 3CCD cameras, for instance. (I've opened one up and seen this)

[Leo Anthony Vale]

I didn't know that Lenin had a pseudonym. I wasn't referring to him, rather Prokudin-Gorskii, also a famous writer, no? As for Lenin, I think the only way he "advanced" photography was having his "friends" airbrushed out of a picture with him after he had them killed one by one. . .

 

Lenin was the pseudonym. Nom-de-guerre would be more accurate, though not my french spelling.

 

Stalin had the airbrushing done. Lenin was already dead by then, possibly helped along by Stalin.

 

Stalin is another nom-de guerre, Steel. No one said he was stainless.

[Will Montgomery]

I applaud your effort. Something that most people would only think about, you actually did. Good job.

[grantsmith]

I applaud your effort. Something that most people would only think about, you actually did. Good job.

 

I second that Brian, Well done.

 

You have inspired me to actually go out and do this too. I made enquiries in a post about trying something similar a few months ago but as will mentioned, like a lot of people it was just theoretical talk so thank you for actually doing it (and doing it very well).

 

As Hal mentioned I think it would be great to see some movement in the frame.

 

I have a few questions however.

 

1) Why over expose?

 

2) are all three exposures equal?

 

3) When you tinted each 'strip' did you have to have different intensities of color? i.e. I wouldn't have thought each color would be 33.3%

 

Any other tips would be appreciated. I also thought about trying it with B+W reversal.

 

It would be great if you did manage to get 3 actual cameras filming the exact same scene and also being able to hand dye the film yourself.

 

Once more, well done

[Matthew Buick]

Do you think that this would be possible/feasible in Super 8, as Monochrome film is about 50% the price of colour.

[David Sweetman]

Brian, that's awesome! I like Phil's idea about a syncronized filter-rotator at 72fps, but that doesn't sound altogether feasible with your range of expertise, and I have no idea how it could be done. The side-by-side method may give it a cool look anyway.

 

Matt - this process takes three times the amount of film for the same amount of footage (meterage?) so if the cost of monochrome was 50 percent, this method would make it cost 150 percent.

[Brian Rose]

Synchronization, in theory, is simple enough. You have a color wheel in front, and link that through gears and such to the main drive shaft/winder of the Bolex camera. In practice, it will be a challenge to build such a device, but I hope to do it. My goal, at the moment, is to try and perfect a two color process. It's simpler than three colors, and has a very unique (if not accurate) look to it.

 

With Super 8, it should be just as feasible. All you need is black and white film, and three color filters. You then record each scene three times with each filter. It has some drawbacks, like you cannot capture action without color fringing, and you must keep the camera absolutely steady, but that's the fun of it!

 

In regards to my exposure, it was pretty simple. I just took a light reading based on my film's normal ASA (80), and then opened up several stops to compensate for the light loss due to the filters. As a rule of thumb, I opened up by 3 to 3.5 stops, and it came out fairly well. The blue record was a little overexposed, so in the future, I plant to lean toward 3 stops.

 

Tinting was surprisingly each, at least in Adobe Premiere Pro:

Step 1) Apply "Color Balance" to each color layer

Step 2) Zero out every color except the one you want remaining (So, if you're working with the red record, zero out blue and green, while leaving the Red record at 100.

Step 3) Stack each clip (I use blue as the base, with Green on Track 2, and Red on Track 3)

Step 4) Apply the "Screen Key" to the RED and GREEN layers ONLY.

Assuming you exposed the film right, you should see glorious, saturated, three strip color!

Brian

PS: I just completed some two-strip tests with two cameras--will report back soon....

[Christian Appelt]

Brian, did you see Marty Hart's Widescreenmuseum.com pages about early color processes?

 

AWSM Early Color Processes

[Leo Anthony Vale]

Do you think that this would be possible/feasible in Super 8, as Monochrome film is about 50% the price of colour.

 

Registration would be a problem.

[grantsmith]

 

wow! it really works. i only had a lee swatch to use as filters (and video). will try it on super 8 on wednesday.

 

I found that it worked better exposing red and green properly (exposing for the filters) and over exposing the blue part.

 

how are your two strip tests?

[Matthew Buick]

Matt - this process takes three times the amount of film for the same amount of footage (meterage?) so if the cost of monochrome was 50 percent, this method would make it cost 150 percent.

 

AYAH !!!

 

 

Registration would be a problem.

 

Oh well, there goes that then.

[Brian Rose]

It would be possible to do this with super 8. But yes, you would be shooting the film three times, and you would need to have the camera mounted firmly to capture the image. On the other hand, owing to its smaller area, it might be more difficult to recombine the image. It could wind up being very soft.

 

Best,

Brian Rose

 

PS: I'll have some two color tests soon, once I send my next batch of film to the lab.