Lewis Hamilton painting – design sketches
During my Valentino Rossi painting I thought I wanted to explore a bit more what would happen if I pushed the contrast between the dark and light areas of my next painting. Even though I had planned out the pattern of light and dark areas in my design, the final painting still includes a lot of in-between tones.
Here is the histogram from the Rossi painting.
The histogram shows how many pixels of each tone, from black on the left through the various shades of grey to white on the far right. As you can see there are three lumps corresponding with dark tones, mid tones and light tones as I had planned, but there are also a lot of in-between tones as well. I want to try pushing it a bit more to get rid of the wishy washy in between tones. What if I make sure my painting only includes dark, medium and light tones, and almost nothing in between, would that increase the impact of the painting? Here’s a great post from James Gurney about this subject.
I’m thinking about a histogram that looks more like this.
I might find it difficult with an impressionist style, or it might look weird, but I thought an experiment would be worthwhile.
Meanwhile, I have been asked to paint a picture of Lewis Hamilton in his 2008 McLaren, the year he won his first championship. I thought that car might be a great one to try pushing the tonal contrast – the reflective silver paintwork tends to show up either as very light or very dark, and the bright red Vodafone wings will be a strong midtone.
What race should I depict? One that he won, of course, which means Australia, Monaco, Great Britain, Germany and China. I like the white stripe in 2008 that identified the soft tyres, it is very characteristic of that period. You can see the stripe I mean around the centre of the tyres in the photo above. That rules out his wins in Monaco and Silverstone I think, both of which were very wet races – Lewis didn’t use the soft tyre in those races. So I’ve chosen to portray the race in Melbourne. I have wanted to paint a picture of Albert Park with the distinctive city skyline in the background anyway.
The skyline I want to portray is visible from the Lakeside Drive section between turns 10 and 11. Something like this, but with a lower viewpoint (this photo is Lewis in 2007 actually, the year before the painting I want to create).
In particular, I want to include the distinctive gold-tipped Eureka Tower, as seen here from the previous corner, turn 10.
Getting the skyscrapers in the picture along with a close-up of the car requires a low viewpoint. To understand why, let’s think about the horizon. The horizon intersects everything you can see that is the same height above the ground as your eyes. If your eyes are 2 metres above the ground, the horizon will cut through everything that is 2 metres above the ground, no matter how far away it is. If the horizon is higher or lower, it cuts through things at a higher or lower height accordingly. If you were standing on a 100m high cliff, you are assured that anything the horizon cuts through is 100m above the ground.
In the photo above we can find the horizon by extending the two parallel yellow lines into the distance and finding where they meet – parallel lines on a flat surface seem to converge at the horizon. So the horizon in the photo is somewhere in those trees just above the yellow bridge, and the photographer is standing that same height above the ground (he/she is in the grandstand at turn 10). Note that all the way along the straight the catch fencing reaches up about half the way from base of the fence to the horizon. Judging from the marshal in orange standing next to the fence, the fence is about 5m tall, so the horizon and the bridge and the photographer are about 10m above the road.
Here are some sketches exploring the possible viewpoints from each side of the car in the place I want to paint it from. These are quick little thumbnails but they still follow the rules of perspective in the relative positions of the tyres, the top and bottom of the walls and the lines on the road.
Practically all the Eureka Tower will be above the horizon line – my viewpoint will be only a metre or so above the road and the tower is 300m tall, so 299/300ths of it will be above the horizon! So to get a decent amount of tower in the picture I need to lower the horizon a lot. I’ve put the horizon about half way up the canvas, but more importantly, it cuts through the top of the tyres and some of the car is above the horizon, and we are looking slightly upwards.
Of the three thumbs above I like option number 2, so I developed it a bit further.
This little drawing is still quite small. It divides up the tones into three, what we might call white, mid-grey and black. I want to try to stick to this degree of separation in the painting, just to see what happens.
I’m starting to organise the tones into an interesting design with this drawing. But how do I know what tone to put where?
In order for the picture to “read” properly, I have to correctly show how the light falls on the forms in the scene. If everything in the picture was the same local colour – for example the car, tyres, road and walls were all a mid-grey colour – then all I’d have to do is paint the bits that were in full sunlight in mid-grey and the bits in shadow in a darker grey. (OK, it’s a bit more complicated than that, I also have to consider half-tones where the object is only partly turned away from the sun, or where reflections bounce light on to a shadowy object). Andrew Loomis calls the tonal difference between the mid-grey sunlit colour and the dark-grey shadow colour the tonal separation.
Of course, the objects in the my scene are not all the same local colour, or even the same local tone. The tyres are black, the road and the red wings are a mid-tone and some of the decals are white. Nevertheless, the same tonal separation must apply to each of these local colours – the difference in tone between sunlit-white and shadowed-white is the same as the difference between sunlit-black and shadowed-black. It’s just the strength of the direct light coming from the sun versus light coming from the blue sky, which is the light falling on the shadowed areas. Your brain is pre-programmed to expect this rigid structure, so if you want the viewer’s eye to interpret the picture you have to play by the rules.
I drew this tonal map for my painting. As you can see, the sunlit part of the black tyres will actually be painting in a mid-tone grey, as will any white decals that are in the shadow.
Hold on a moment though, my last thumbnail above breaks that structure. The local colour of the tarmac is going to be a mid-grey, so in the sunlight I must paint it in a midtone, the same tone as the red Vodafone decals. In my sketch earlier the road was a light tone. That will not work I think, your eye would interpret the road as being white if I did that. Let’s redraw that design.
Good, a mid-tone road means we can add some white lines to increase the sense of perspective anyway. The tones in this drawing follow the form principle described above, and they are still arranged in a nice abstract way. You have to think about organising the tones because a good abstract arrangement of tones will grab your attention from the other side of the room. Here’s the abstract arrangement of white vs grey/black
We have ended up with a design that includes a lot of mid-grey again, but my aim really is to keep this design clean and try not to include a lot of tones in between the mid-grey and the white. I don’t mind the mid-greys that I have designed into the composition, I just want to keep control of them and not end up including every shade of grey.
I seem to have trouble mixing very light colours, I always underestimate how much white paint I need and end up with something too dark. I find it only requires a tiny bit of colour to darken the tone of a light mix a lot, so for this painting I will probably start with a wash that matches the light tone I’m going for and then draw in the darker areas, and not start with my usual mid-tone wash and try to lighten the sky up.
I couldn’t find a reference photo that had just the right angle of the car, so I found one that was similar and then did a careful “technical” drawing adjusting the perspective to fit my requirements. I had to lower the perspective from the reference photo, so perspective skills came in handy.
It’s so important to get all the perspective points right – the height of the rear tyre in relation to the front tyre, the top edge of the rear wing end plate, the curve of the walls and catch fences etc. If you don’t pay attention to those details you brain will tell you there is something mysteriously wrong, even if you can’t put your finger on what it is.
That’s enough for now. I’ll be making a start on this painting in the next few days.
