DETAILS


COLUMNS


CONTRIBUTIONS

a a a

ENTERTAINING THE FUTURE

Vol.32 No.3 August 1998
ACM SIGGRAPH



A Whiter Shade of Pale



Mike Milne
FrameStore


August 98 Columns
About the Cover Visfiles


Mike Milne
Previous article by Mike Milne Next article by Mike Milne

Money into light. What a powerful metaphor for filmmaking. It was, of course, the title for John Boorman’s 1985 book describing the making of his film, The Emerald Forest. As a description of the filmmaking process, it is startlingly accurate if a little on the reductionist side. There are many who would dispute that it tells the whole story — film music composers and sound effects gurus among them.

My fascination with the metaphor is rather more literal, and was revived recently while revisiting some of the stomping grounds of my youth. Actually, they were probably not so much stomping grounds as lounging-around-in-what-I-hoped-was-a-cool-and-languid-pose-grounds, if I were to be honest with myself.

As a teenager I used to live in Hampstead, a leafy suburb of London that in those days was a haven for dreamers, intellectuals and misfits of all descriptions — mainly because it was bordered on two sides by several square miles of wilderness, known as Hampstead Heath, and because it was relatively cheap to live in (today it has become a trendy neo-Chelsea for yuppies, where even the supermarkets have been replaced by haute couture boutiques and mobile phone stores, and you have to have a seven figure salary to live there). It also has the distinction of being the highest point in London, towering a full 300 feet (count ‘em!) over the rest of the sprawl, and consequently a perfect place from which to watch the sunset and muse upon the meaning of life. And it was while I was doing just that, recently, that I happened to remember Mr. Boorman’s metaphor, and to think that while filmmaking was indeed turning money into light, often there isn’t enough light — quite literally. I should like to develop this thought by referring to another writer or two, and a handful of painters.

Just near the junction of Hampstead High Street and Heath Street, there was a little coffee bar — imaginatively called the Coffee Cup — that, in the ‘60s, served as a gathering point for all the waifs and strays of Northwest London. During the day, we would sit around its little tables and strive to be intellectual — but in the evenings, the serious work would begin — swapping party addresses and trying to persuade members of the opposite sex to accompany us to wherever the wildest party was likely to be. The Coffee Cup still performs this function, 30 years later — and for all I know, might have done so since our ancestors were running around in blue woad (and very little else).

One of the cool things to discuss during the long days at the Coffee Cup was the works of the currently fashionable writers, especially if they were (a) not English-speaking and (b) dead. And the coolest of all the cool writers in the ‘60s was, undoubtedly, Hermann Hesse. He may not have been dead at the time, but he was certainly not English, and most of his writing was done in the ‘20s and ‘30s — which, we supposed, was as close to being dead as you could decently get. Hesse had written a book in 1927, Der Steppenwolf, which was enjoying a new popularity among the nascent hippies and would-be mystics of the flower-power decade. As a fully paid-up member of that clan, I naturally read the book from cover to cover and memorised key passages to recite to any passing nubile party-goer. (It was said in the ‘60s — I forget by whom — that if a bomb were dropped on Hampstead, the artistic and intellectual brains of England would be wiped out at one stroke. Whoever said it, they forgot to mention the sad poseurs).

In Der Steppenwolf, the principal character has a moment of epiphany in which, while listening to a scratchy recording of Beethoven on a wind-up gramophone, he realises that it doesn’t matter how poor the reproduction is, because there is a luminous thread — the soul of the music, if you like — that passes across to the listener. Faithful reproduction is not essential — the Idea would Get Through. (I suppose you could call this an example of early digital thinking — discrete ideas surviving transmission through narrow, and noisy, bandwidths). Hermann Hesse had never, of course, sat down in front of a pair of state-of-the-art JBL’s with a brand-new compact disc spinning in the CD player — not that it would have altered his argument — but it might have made him more reluctant to throw away the concept of faithful reproduction as a desirable goal.

And certainly, his argument can be made to apply to other uses of audio technology, rather than to music. We still put up with the terrible sound quality of telephones because they are performing the digital function of transmitting ideas (“I’ll be home late”) rather than evoking the true quality of the speaker’s voice. But we no longer put up with the sound of music from a wind-up gramophone, because — at the heart of it — Hesse was wrong. Beethoven was not trying to tell us he’d be home late, and what he had to say cannot really be heard in any other way than by listening to the orchestra. Sure, you can whistle along with a well-known melody when you hear it on a cheap radio, but the essence of music is precisely that sound quality that entranced the composer’s inner ear, and the nearer we get to it, the better the experience.

Movies have cottoned on to this as well — audio technology is at its peak in movies, and even television (or home cinema as it is referred to now by TV salesmen) have sound reproduction to rival even the most expensive stereo in the coolest ‘60s pad. We want that richness of detail — we want the deep bass notes (whether of the Titanic breaking up on that iceberg, or the double basses in Beethoven’s seventh) to shake our rib cages, and the high notes, (whether breaking champagne glasses as the first-class dining room sinks into the briny, or plinks on a glockenspiel) to be crystal clear and sharp.

However, even if the battle for faithful sound reproduction is almost won, the Hesse argument is still alive and thriving in the guise of a technical justification — the “normalised vision” argument — for an equally unsatisfactory state of affairs in the visual field. Let’s start with painting, since it’s probably the oldest of the visual arts.

Painting has always had the built-in requirement of “normalisation” — that is, you can’t paint a colour darker than black, and you can’t paint a colour lighter than white, so everything you paint has to fall between the two. For those of you fortunate enough not to have been presented with this problem professionally, you might well ask, “Where’s the problem? What’s lighter than white, and what’s darker than black?” To which I might well reply that while nothing much is darker than black, there are quite a few things that are brighter than white, even though they might not actually be white themselves, and these things are all emitters of light — either directly, or by fairly efficient reflection.

In CGI we now class these things as visible light sources, but from the earliest days the great painters wrestled with the problems of showing light sources with a limited colour palette (one of those rare cases where that word is used in its original meaning). These problems were imposed as much by the physics of light reflected from an opaque surface as by the limitations of coloured earth chemistry. Caravaggio (and fellow chiaroscurists) had already pinned it down back in the 16th century, when they discovered that in order to paint a visible light source it wasn’t enough to paint it in the lightest colour available (white, or more usually a pale yellow); the rest of the picture had to be darkened — so that anything previously white, (such as pages in a book or a white neckerchief) had to become pale grey, and anything previously pale grey (the torturer’s horse, or the old crone’s hair) had to become dark grey, and anything previously dark grey had to become ... but you catch my drift.

The novelist and wit Gore Vidal once said “To succeed is not enough. Others must fail.” Though originally describing the problems of human stars attempting to outshine their rivals, this is analogous to the problems of visible light sources in painting. Only by shifting the other parts of the picture down the luminance scale can the appearance of a light source be simulated, in conjunction with other techniques such as painting the effect of the simulated light falling on nearby surfaces. This, in turn, meant painting those surfaces a little lighter — but of course, that couldn’t be done, since the lighter colours were already spoken for, so the alternative was to darken the other parts, and so on. Eventually, as the painters neared the shadowed part of the painting, all colours were compressed to brownish-black, and details were lost — but that’s another problem (in today’s parlance, that’s just a bit depth problem — not enough subdivisions in the paint scale).

This is not to say that there aren’t magnificent examples of the genre. The Dutch tenebrists were no slouches when it came to lighting effects, and Joseph Wright of Derby’s superb pieces, The Orrery and Experiment with an Air Pump show what can be done by a master of his art.

Just a few minutes away (by ox-cart) from the Coffee Cup is the place where the painter John Constable painted Hampstead Heath With a Rainbow. When Constable hung around Hampstead in the early nineteenth century (perhaps the Coffee Cup didn’t exist at the time, and he wasn’t able to swap many party addresses), he was trying one of the most difficult of “paint into light” tricks — painting a rainbow. And although he was the acknowledged master of atmospheric lighting effects, I have to say that this one was not a great success — through no fault of his own, I’m sure — he was just up against the old normalising problem. He attempted to cover it by surrounding the rainbow with a dark cloud, but you can’t help noticing that the white clouds on either side are much brighter, as is the windmill below, and almost everything else in the scene.

Fifty years later, another painter was trying his luck on the Hampstead party scene — Ford Madox Brown, one of the pre-Raphaelite brotherhood who were impressing their girlfriends by trying to get back to the vivid, natural-colour style of three centuries earlier — and who painted a picture of the very street where the Coffee Cup stands — but unfortunately, he was facing the wrong way, and so we are unable to see his friends as they haggled over the choicest venues for the evening’s revelry. No doubt he was anxious to join them, for he painted the scene by daylight so as to be ready for action by nightfall — but this was not true of one of his admirers, the lesser-known Atkinson Grimshaw, who also painted the same scene — but at twilight, with the gas lights a-twinkle.

In this picture, View of Heath Street by Night, we see the problem laid bare — for as enchanting as the painting is, there is no doubt that the visible light sources are a problem. The scene is at twilight, which means that the sky cannot be completely dark — and neither can the wet streets, for they reflect the sky. The moon is the brightest area — in white, and the gaslights are pale yellow — but the painter has had to leave out all the secondary lighting effects to avoid having to darken the pavements.

Now it must be said (and, in fact it was — by the American painter Whistler) that Grimshaw was probably the greatest exponent of visible light source painting of his day. After seeing his Nightfall down the Thames, Whistler said: “I considered myself the inventor of ‘nocturnes’ until I saw Grimmy’s moonlight.” (Whistler, you may recall, was the poor unfortunate who, upon showing his painting The Falling Rocket, was accused by Ruskin of “throwing a pot of paint in the public’s face” — and was nearly bankrupted by the ensuing court case).

It’s worth noting, in the Heath Street painting, Grimshaw’s use of a star filter effect — probably the first one, as it was painted 110 years ago — around some of the gas lights. This gives a clue to the fact that he was one of the first painters to use photography extensively as reference material — and that photography has no solution to the visible light source problem either. The white photographic paper is just as much an upper barrier as the Chinese White paint of the artist. Today’s digital artists, using rendering programs to simulate light and shade, are up against the same thing, although the barrier is electronic. However cleverly the algorithm reproduces the gradations of light and shade, they are still subject to the tyranny of the normalised spectrum — the awful one-volt barrier, as my video-engineering friends say.

Mike Milne is Director of Computer Animation at FrameStore, which together with its sister company CFC, forms one of Europe's largest digital effects teams. Mike started out as an artist and beachcomber in the '60s, moved into graphic design in the '70s and finally to computer graphics in 1982. Sometimes he regards his career as one long, downhill slide.




Mike Milne
FrameStore
9 Nole Street
London W1V 4AL
United Kingdom

Tel: +44-171-208-2600
Fax: +44-171-208-2626


The copyright of articles and images printed remains with the author unless otherwise indicated.

Now obviously it is true that in the visual arts, it is precisely the limitations of the medium that create the artistic value — for it is the medium of reproduction itself that embodies the work of art, which is not so in the case of music, where the recording is a secondary thing — something that transfers the work of art from the privileged sanctum of the concert hall to the wider audience out there in stereoland.

But my concern is perhaps less with the artistic merit of the medium and more with its actual physical limitations, especially as they apply to entertainment — which, although undoubtedly an art form, is far more reliant on faithfulness of reproduction than its progenitors. Or, to put it in simpler terms, the more we’re fooled, the better we like it, and the more we’re prepared to pay for the experience.

So the problem that faces all entertainers and artists working in paint, print, television or film is the same one — whatever we do, we can’t escape the major difference between the real world and our attempts to recreate it (apart, of course, from its three- dimensionality — a problem that I don’t intend to touch on) — and that is that the real world has no limit to the brightness of the things you can see, or if you like, to the amplitude of the visual signals that your eye receives (I suppose an artificial upper limit can be placed at the point where your eyeballs fry — as would-be observers of the 1999 solar eclipse might well find out) — while the visual media have a hard limit, whether governed by the whiteness of the pigment or the Broadcast Standards Authority’s rules on legal transmission levels.

The argument I often hear put forward has a familiar ring. “Of course the upper limit doesn’t matter,” they say. “Your brain just normalises the range, and you perceive the light areas just as if they were light sources.” Oh, I see. The Idea Gets Through, right? Well, the idea might, but the full visual richness doesn’t. When I stand at the top of Heath Street and watch that fiery red balloon of a sun sinking below the trees on the West Heath, my brain isn’t normalising anything, I’m glad to say. Instead, it’s glorying in the physical effect of a light source that’s almost (but not quite) too bright to look at, and thinking “What a beautiful sunset. Now I wonder where the best party is tonight?”