El sol del membrillo (The Quince Tree Sun, also released as Dream of Light) is a film about process. The artist Antonio López García is shown carefully and meticulously setting up his workspace, making a canvas, mixing paints. He builds a wooden frame around his subject, the small quince tree in the shared backyard of his apartment block, to hold plumblines and markers and, later, a small tent. He hammers pegs into the ground to mark where he stands. The film's most striking visual, reproduced on various posters and DVD covers, are the cruciform white marks he paints on the quince fruits themselves, tracking their slow movement as they ripen and sink under the force of gravity.

The camera mirrors his obsessive focus, lingering on wide shots of his lengthy setup and zooming close on individual brushstrokes or pencil marks. At times however its focus broadens, encompassing the other people in García's life, the other residents of the apartment and the city around them. Occasionally we hear snippets of news broadcasts from García's radio, reminding us of the presence of a whole world surrounding this tiny quince tree.

In taking such care, García aims to capture the sunlight as it falls on the bright quince fruits. We see him explain his goal to friends and onlookers in a way which makes clear that he doesn't know exactly how to express the beauty he finds in this image, other than reproducing it in paint as carefully and exactly as he can. Ultimately his meticulousness is his undoing, however, as first the weather turns foul and then, despite his struggles, the fruits ripen and fall before he can finish his work. But the ending is bittersweet. He has still created some beautiful, if incomplete, drawings. The film itself stands as a powerful work of art in itself, and the film ends with friends, family and acquaintances in his small community sharing in the sweet fruits. And, of course, there's always next year.

García's scrupulousness stands somewhat at odds with my own artistic instincts. I suppose you could say I lean more towards an Impressionist style in my own comparatively amateurish artworks, embracing the inherent flaws of my medium and allowing the viewer's mind to fill in details which paper or film or a digital camera sensor could never adequately capture. However, I cant deny that I also feel an odd kind of attraction to the careful, precise and slow way in which García works. Many films, fictional and documentary, cut away from moments in which El sol del membrillo luxuriates, seemingly afraid of boring their viewers with details. But I am always curious, and wish they would let me watch just a little longer. I experience a strange craving to find some similar goal which I can work towards, methodically and with purpose, repeating steps over and over until it's finished.

García's attempts to reach a deep understanding of his quince tree through careful measurement and replication stirred a surprising memory in my mind. I remembered being a child, watching some promotional material for the Pixar film Toy Story, or perhaps its sequel. I have not been able to find the original clip but I remember one of the animators saying that the hardest scenes to animate were those that took place outdoors, because the background trees had to have such a vast number of leaves, in the hundreds of thousands or perhaps millions. I think as a child I assumed these leaves were all individually sculpted and painted, when in reality I'm sure the animators took various shortcuts to ease that impossibly tedious task. But I began to wonder, could you count all the leaves on a tree? And if you did, what would you learn?

A couple of days later, I packed a notebook, colourful thread, scissors and some snacks and headed to my local patch of woodland. I knew there would be some serious limitations on what I was trying to do. Firstly, I was only one person, limited energy and time. Secondly, any climbable tree would be far too big, so I would be limited to trees short enough for me to reach their tops from the ground. Thirdly, it was late August at this point, and many leaves were already turning brown. I would probably not have another chance to do this this year, before the leaves would start falling faster than I could count them.

I decided to start as small as possible. I found a spindly baby sycamore, so small and bare that if it was leaning over a little further I would have assumed that it was a young branch fallen from one of the larger trees nearby, not a tree in its own right. But it had roots and branches, and I thought it would be a good way to test whether the method I had come up with would work. I started working my way up the branches, tying a small piece of thread at each fork to mark where I had been. I was shocked by how immediately tactile the process became, as I instinctively started trusting my fingers over my eyes, feeling for the smooth pliant stems which distinguished the base of a leaf from the rough woody bark of the twigs. Despite the miniscule size of this tree, by the time I had finished with the main stem I had already counted 99 leaves, far more than I would have estimated just from looking at it. Including the even tinier side stems, the leaves totaled 148.

Having found the sycamore so easy to manage, I started looking for something larger and bushier. At first I ignored the hawthorns which were everywhere in the part of the woods I was searching, afraid of hurting myself when feeling up and down their spiny branches. But I couldn't find any other trees which felt right, and so I returned to one especially nice multi-stemmed copse. The thorns turned out not to be a big issue as my slow and careful work allowed me to avoid them easily. However the counting was not as simple as the sycamore had led me to expect. Hawthorn leaves are much smaller and more densely packed, with side-branches remaining at an angle far closer to the stem they split from, making them harder to tell apart even when marked. Hawthorn thorns, for those who don't know, are actually modified leaf buds, and in spring they unfurl into multiple leaves, stacked almost vertically and very close together so that the terminal leaves of each twig were layered like a sandwich, very difficult to keep track of. Often, one or two of these leaves was already yellow or had even died completely, being held in place only by the grip of its siblings. I couldn't decide whether to count these leaves or not, and I'm fairly certain I changed my mind multiple times when finding these ambiguous cases. The wood of hawthorn is also famously hard and stiff, and even the thin stems of this young copse were far less flexible and harder to manipulate than the thicker stem of the sycamore.

After counting about 200 leaves, I realised I had bitten off more than I could chew. I headed home as the quiet of the woods was broken at intervals by the soft thud of the year's first conkers landing on the already-fallen leaves of their parent trees, and I knew I would not be able to try again this year. Perhaps one summer day I will find a more cooperative tree and collect some friends to count all its leaves. My absolute dream would be to climb some ancient oak and spend a whole weekend taking stock of what could easily be millions of leaves. I still want to do this because, despite my extremely mixed success, I found the experience very rewarding. I feel that I understand the shape, structure, and personality of the sycamore and the hawthorn on a much more intimate level now, perhaps approaching the extent to which Antonio López García understood his beloved quince.