Traditional Matte Painting Technique Development

So the next paragraph of my dissertation is going to focus on the traditional techniques of Matte Painting, I think this might be challenging to sum up as it was such a lengthy process and there were three key developments of it. With it being really technical I don't want to paraphrase it incorrectly.

Barry Salt record Norman O. Dawn as the first to use a glass shot in still photography in 1905 (see 1983: 159). His employer at the Thorp Engraving Co. in Los Angeles, Max Hands leigh, had shown him how to enhance the appearance of a building by placing a sheet of glass in front of it and painting the adjustments directly onto glass. He also used this technique in Missions of California (1907), allowing partially built sets to be 'completed' by painting extensions onto glass. In 1911 he modified this process to create the glass matte shot, in which a sheet of glass is set up in front of the camera and a matte of black paint applied to it to cover unwanted parts of the frame. This is then filmed (along with several feet of test footage from which the glass painter will then work) and developed. A paused frame from the film can then be projected onto the artist's easel. The painter paints the sections to be placed in the matted area, and blacks out the areas of the original film. The painting can then be filmed as a second exposure with both elements combined in the same frame (see Salt 1983: 160). This process required meticulous planning to ensure that the 'real' and the painted elements were integrated with minimal evidence of the seams on display. It could be used to place actors in virtual locations or to place painted ceilings on roofless sets.

North, D (2008). Performing Illusions, Cinema, Special Effects and the Virtual Actor. London: Wallflower Press. p75-79.

So technique 1. Camera, glass panel with painting, live action footage all in camera effects. Required precision, and had to be done in a timely manner due to lighting changes.

The use of glass shots in film making is a laborious process. As with hanging foreground miniatures, the camera is positioned to film the scene in question, it’s tripod chained in place, and sandbags arranged around its legs to prevent any movement during the lengthy preparation process. A sheet of glass, held in a secured wooden frame, is positioned about 3m (10ft) in front of the camera. It is vital that the combination of camera lens, film stock and lighting produce sufficient depth of field to keep both nearby painting and background landscape in perfect focus. 

Rickett, R (2007). Special Effects: The History and Technique. 2nd ed. New York: Billboard Books. pg189

I think it's important to note that this technique also has it's flaws with fact that the glass could often get chipped on the edge causing it to shatter. I think it's also interesting that the distance the glass has to be from the camera has to be perfect focus. Richard Rickett continues:

To create the painting itself, the matte artist looks through the camera viewfinder and directs an assistant, who marks the glass to show the points at which real scenery and painted areas are to merge. The artist then sketches the scene on the glass with a wax pencil, frequently checking the marriage of drawing and background elements through the camera viewfinder. The required image is then painted directly onto the glass, usually in oil paint, using tones and colours that blend perfectly with the scenery in the background.

I'm going to have to invest in a wax pencil. Is this just a crayon? Apparently not, also known as grease pencils, they are overpriced. I bet the college shop has some. I ordered some glass Panels, finally, I went to Andrew's Glass but it's shut down (as the woodwork technician suggested), but I found another glass shop called Designs on Glass in Harehills, and I ordered a couple of sheets 900x900 for £31 a sheet.  I have lots of oil paints. But I like this paragraph because it tells you the tools of the past trade. I particularly like this one page, which I'm going to discuss in the techniques of painting chapter. But it's just mind blowing so amazing to read.

Back to this chapter though here is the evidence to back up my comment about the lighting:

While painting the image, the artist must judge how light and shadow will look at the time of day the scene is to be filmed, in order to ensure that the painting will blend with the live-action component of the shot ... Once the sun has moved to a position where shadows on both painting and scenery match, the director has perhaps an hour to capture the scene before the light quality becomes too dissimilar.  

It is vital that there is no movement of glass or camera during filming. If either moves fractionally, the background scenery and foreground painting will appear to shift, destroying the illusion that the two elements are one. When there is bad weather, filming can only proceed between gusts of wind, since this might wobble both camera and painting. Some camera movement is possible during filming if the camera itself is mounted on a nodal-head tripod, which allows basic pans and tilts to be achieved without painting or background becoming separated.

Nodal Head tripod pictured above, i didn't have a clue what one was. I think it's interesting that movement be so important too because I guess this would be like a bad composite in after effects if it shifted animatedly it'd look like a bad green screen keying out job. The next paragraph I thought was interesting trivia.

Even after the development of more sophisticated methods of combining live-action images with paintings, the basic glass shot remained popular for many years. During World Ward II, the process was used during the production of newsreel films in armaments factories. To avoid having to remove all sensitive equipment , plans or, or signs from a factory floor, a sheet of glass was placed in front of the camera, and any visual information considered useful to the enemy painted out or disguised. 

The final paragraph of this section is the most important because it shows that even Norman Dawn saw this process as limiting. This is definitely quotable.

Simple Glass shots are only occasionally used today - the inconvenience of preparing and painting a shot on location while a whole film crew waited for the results led Norman O. Dawn himself to look for other techniques. 

I know digress back to performing illusions, I'm not going to include this in this chapter, because I haven't found many examples of it, but I think it's something to bare in mind if my practical goes wrong and I break all the glass panels, I could give wooden blocks a go. cut out to fit the image, i think this would be super challenging compositing it with the live action to not have a really shit line. However this is where the inspiration of digital could come in altering it to make it look beautiful. Note to self.

  'Hall's background process' was an extension of the glass-shot technique, but using a cut-out plywood background instead of a sheet of glass. The wooden shape would act as a mask for the scene involving actors performing many metres away from the camera to lend an impression of great scale to the composited elements (see Low 1971:246). Paramount held the exclusive American rights to the use of this process, and the minor variations in similar processes were forced by copyright restrictions requiring each studio to develop their own distinct method of compositing, rather than by an urge to standardise the 'most realistic' process. In the period 1907-13, some special effects that had previously been used only in trick films were first used regularly in 'substantial dramatic films' (Salt 1983: 130). These include composite photography using mattes and counter-mattes with part of the frame blocked off in each exposure, which can be seen in Italian epics such as Cabiria (Giovanni Pastrone 1914) Any slightly unsteady movement of the film in the camera gate would reveal the appearance of a black line between the two separately exposed parts of the frame. North, D (2008)  

Black lines are what you get in green screening when you feather the edge too much bad compositing, I think this is something that I've come to notice more and more and I've started reading this new book called, Klein, N M. (2004). The Vatican to Vegas: A History of Special Effects. New York: New Press. Which basically steals my thunder of my becoming revelation because he's got it written down first that essentially Matte Painting is Compositing. It must be really obvious. But what I really like about this book is how he discusses every method of matte painting really succinctly. In a list. Who doesn't love a good list?

The composite is fundamentally a matte. Standardly - before the computer algorithmically turned all of these into date - mattes came on glass, as mirrors, or through gauze (very much like magic lanterns):
Glass Shots (as in paintings and transparencies);
Mirror Shots (for sharper focus superimpositions, including miniatures);
In-the-camera matte shots: Where part of the frame was matted out in front of the camera, while shooting. This began simply as an opaque card, or as glass matte, a trick established by 1911;

I'm not going to include mirror shots in my chapter because I don't think it's particularly important, it was something George Méliès started in theater, as a way of showing ghosts on stage.  I think that although, this is succinctly summed up in this book it is a little brief. I think rather than doing two paragraphs one on traditional techniques and one on digital I think I'm going to have to do two on traditional techniques, the first being on glass shots and in-camera matte shots and then the second on these kind of shots:

So we move onto optical printing in matte painting: An optical printer is a device consisting of one or more film projectors mechanically linked to a movie camera. It allows filmmakers to re-photograph one or more strips of film. The optical printer is used for making special effects for motion pictures, or for copying and restoring old film material.

Optical Printer^

Bi-pack Optical Printing: One film image is projected on to part of the frame, while the camera shoots the rest of the frame. The the moving matte for wipes and various overlaps, as well as in action (traveling) mattes. Bi-pack gadgets evolved into optical printing machines, some as big as a room, long capsules inside a massive steel housing, long enough to hide a body, as if for an MRI scan. But big or just on tabletops, the same principle applied: a projector sends back images to the camera, which, in turn, keeps shooting the scene.
Until the nineties, optical printing was the workhorse of special effects. It could engage as many as three printing heads at once - to wipe, slide and composite movement. But by 2000 it was gone, relgated to obscure downstairs rooms at the few effects houses in L.A. The printers became a piece of nostalgia, a machine twenty years old, operated often by old-timers near retirement, for an ever decreasing cadre of older clients. .... blahblah goes on to discuss this masterpiece example of optical printing: 

"Tango" by Zbigniew Rybczyński, 1980 from Tito Molina Faceblog videos on Vimeo.

I actually really don't like this short, Controversial, I know. I dislike how fake it is, and how it's quite fine arty, in the sense there's no real narrative, it's super repetitive and not very entertaining. That's personal opinion. However I can understand why it's important to the evolution of special effects particularly that of the optical printer because before this nothing really had been seen like this, where live action people are in the same sequence repeating an action not interacting with each other, it was innovative. Now it's not matte painting but it shows the use of an optical printer. This later allowed for things like skies of moving clouds being composited with a live action different scene. see proper explanation of how it works with a matte painting below:

The first uses of travelling mattes date back to the early 1920's, following Frank Williams' patenting of the first successful technique in 1918, (see Field 2002). The Williams process involved shooting the foreground action against a highly illuminated white background, and from the negative of this a high contrast film was taken to produce a moving black silhouette on clear film. This was then combined in an optical printer with the negative of the background plate, followed by a second pass to place the original foreground action in the scene (Salt 1983:206). North, D (2008)

I think I can't use the above quote because of how much I've read on the Norman Dawn Vs Frank Williams Court Case. So I have these two contradictory sources, and I could triangulate them with a third source and go for that, but actually I don't think it's important, and my history of Matte Painting chapter doesn't actually give a shit with who came up with this traditional technique it's all about the fact that it exists, and it was an absolute nightmare which forced them to come up with new methods asap because it was using like a ridiculous amount film stock to composite one shot.

The Dunning Process was patented in 1928, a travelling matte technique that eased some of the troubles of the Williams Process (see Salt 1983: 234). Otherwise known as the Dunning - Pomeroy self-matting process in acknowledgement of the Roy J. Pomeroy, this was a complex matting technique which proved prohibitively awkward and ultimately incompatible with colour cinematography. North, D (2008)

Travelling mattes explained:

Richard Rickitt's explanation is quite long winded and doesn't really make sense without the diagram's but here it is:

Since the earliest days of motion picture history, film makers have sought ways to convincingly combine, on a single strip of film, images that have been filmed at different times or locations. Melies and Porter, among other early film makers, used a simple technique of blocking, or ‘Matting” out, parts of the frame during initial photography in order to preserve an area of unexposed film for the later addition of other elements. Such simple ‘split-screen’ processes, performed on the original piece of film, could be effective if the additional action was to remain in a fixed area of the frame without interacting directly with the rest of the picture. More attractive to film makers, however, was a method that would allow characters filmed in a studio to appear to inhabit places that were located on the other side of the world, or that only existed as small models constructed by the special effects department. Producing the two elements to be combined is a simple process. The desired background can be photographed on location on one piece of film, and the foreground element - usually an actor performing against a blank studio backdrop - on a another piece of film. Combining the two pieces of film is where the difficulty begins to arise. If the separately shot film of an actor, who is to appear in the foreground, is simply placed or printed on top of the film of the desired background a composite will be produced in which the actor may indeed appear to be at that location. But due to the transparent nature of celluloid film, the background element will be visible through the actor, making the performer appear ghostlike (fig. 9.).The only way to combine the two elements convincingly is to produce a background image that contains and unexposed hole exactly the same size and shape and the moving actor. The photograph of the actor can be slotted into this hole to produce the desired result. To produce such a combination of images, several additional elements must be created. First, a ‘male’ matte, which is a black silhouette of the foreground performer on a clear background (fig.10a). This silhouette is placed over the desired background image (b), and the two are printed together to produce a piece of undeveloped film that has been exposed to the background except in the areas covered by the matte (c). This piece of film must then have the image of the foreground performer (d) copied into the area left unexposed by the male matte. This involves the use of a ‘female’ matte, an exact opposite of the male matte in entirely black, except for a clear area the size and shape of the foreground actor (e). The female matte is placed over the undeveloped film that has already been exposed to the background, the black area covering the already exposed background areas to prevent them from receiving more light during exposure and leaving uncovered the as-yet-unexposed area of the film. The image of the foreground actor is then copied into the area of the film not concealed by the female matte. When this piece of film is developed, the result is a combination of the desired foreground and background elements known as a ‘composite’. Since the male and female mattes must change in shape, position and size for every subsequent frame of film in order to accommodate the actor travelling about the frame, techniques in which they are used are known as 'travelling mattes’.

Where as it's the opposite with Norman Klein in Vatican to Vegas.

Traveling Mattes: A part of the frame is masked off during shooting, then action added there later on. For generation, the mask was blue screen. On television, the blue screen was colour-keyed (chroma key).  

Created by Larry Butler the 'blue screen' travelling matte process.

In designing his blue screen technique, Butler was trying to create "traveling matte" elements that could be tailored to technicolor's primary color process. The classic approach was to allow a specific actor or object to be photographed against a neutral background and combine that film element with separately photographed backgrounds. The old approaches had technical drawbacks, but the beauty of bluescreen was that the primary-colour backing made it easier to photograph and "extract" the subject element. Since there had to be a color separation from the backing and subject, blue was deemed the easiest colour to avoid. "You might say, 'My God, how can that be true?;" Butler said. "'The sky is blue, everything is blue.' Well, that isn't true, there's very little blue on people. And blue eyes are not big enough to bother anybody."      ......
The bluescreen compositing was ultimately accomplished on an optical printer built at Denham by Butler (with the help of Jack Thomas and Tom Howard). The process would become vital for creating composite elements, even into the digital age.  

Cotta Vaz, M and Barron, C. (2004)

From about 1930 both travelling matte processes were abandoned in favour of back projection, an in-camera effect achieved live on the set with the concomitant reduction in costs and time schedules required. The earliest back projections had to be on small, ground glass screens (hence the frequent use of the process to show the back windows of Moving cars), until the introduction of a new cellulose screen material and improved optics in the projectors enabling use of much larger screens in 1932 (see Salt 1983: 269-71). North, D (2008)

Rear Projection: Action projected behind a screen, then filmed behind the action, for example, the rearview mirror inside automobiles during a car chase (circa 1935). Or a fuzzy ocean behind Fred Astair while he leans on the rail of an ocean liner, and pitches woo. Rear projected imagery sharpened during the television era. Chroma-key systems quickly became a standard in feature films as well. (Klein, N: Vatican to vegas)

Right so this next chapter needs to talk about bi-pack optical printing then travelling mattes and blue screening, and back projection/rear projection techniques because these were the three next big developments in matte painting, before the digital age. Also due to the dawn of Technicolour:

Technicolor, (trademark), motion-picture process using dye-transfer techniques to produce a colour print. The Technicolor process, perfected in 1932, originally used a beam-splitting optical cube, in combination with the camera lens, to expose three black-and-white films. The light beam was split into three parts as it entered the camera, one beam favouring the red portion of the spectrum, one favouring the green, and one the blue. Each image was captured simultaneously on a separate band of black-and-white film. The three strips were developed separately and printed, after which the prints were passed through their appropriate coloured dyes; when laminated together, they produced a reasonably faithful approximation of natural colour. In a later version of the process, only one integral tri-pack colour negative film was exposed during filming, and three colour-separated negatives could then be made from this. These three colour-separated strips were appropriately dyed and then superimposed on a final emulsion to produce a full-colour image.  http://www.britannica.com/topic/Technicolor


Another Matte shot explained in Performing Illusions is this:

Eugen Schüfftan, is best remembered as special effects manager on Metropolis (1927),

utilising to the full the 'Schüfftan process' of shot composition which he had pioneered for use on Lang's Die Nibelengen (1924), released in two parts, as Siegfried and Kriemhilds Rache, two months apart. Using a dual-lensed camera capable of exposing two focused images onto a single filmstrip, the process could combine miniature models and live actors in one composite shot without the need for multiple exposures or extra laboratory work. This matte-shot technique involved scraping off an area of the tain on the back of a mirror and thus composited in camera and on the set.  North, D (2008)

I think that's all the traditional techniques covered, now for the digital age!