I imagine many people think of sight size mostly as a technique used in ateliers to train drawing accuracy. Bargue drawings, cast drawings, that kind of thing.
But I wonder if sight size sometimes gets a bit of a bad press, as if it was a kind of self-flagellation realist artists must go through before they earn their stripes.
It can be that, but it can also be outrageously useful if applied a little less stringently. It’s not just about measuring. After all, it just means drawing or painting something the same size as you see it, from a fixed position. And that can be an incredibly useful thing to do, especially if you’d like to be able to paint more realistically.
I’ll try to explain what I mean by showing you how you can use a sight size approach to paint a small still life – one I did last week, actually, in a couple of sessions. The first session was about 5 hours, and the second one two or three (somewhat lengthened because I was filming it, which is always a bit of a mess about). Here’s the painting:
So this isn’t some huge, month-long undertaking. You can try it too in the comfort and safety of your own home studio.
So grab a cuppa, pull up a chair and I’ll take you through the steps I took to make this little still life painting, and tell you where I think doing it sight size made a difference.
This little painting was intended as a demonstration, one I filmed for a still life painting course I’m putting together. So I wanted to set myself a couple of challenges.
The first is the inclusion of a white object – the little pot. Having that in the painting means that I can’t match the values like for like, because the range of values I have available in paint is more narrow than those I can see in my subject. So I have to do some magic with the values to get it to look convincing and realistic. That’s usually the case, and is one of the main reasons so many people struggle with values. So it’s good to get that in there and practice with it.
Secondly, I chose a high chroma object to paint. Modelling form is a little more difficult on high chroma objects. I use a nifty little method that gets round some of the problems of modelling high-chroma form convincingly – judging the change of colour across the form from dark to light. That’s more difficult to do on high chroma objects because the chroma itself tends to interfere with our perception of the value.
The set up
So let’s talk about sight size.
The set up for working sight size demands a bit of care and attention. I usually work next to a window, because I like natural light. For this demo, I used a viewfinder with a grid of red cotton thread that I can replicate with a pencil grid on my painting surface. These days, I tend not to use this myself, or to use a simpler version. But if you haven’t done much sight size work before, this is a really useful thing to do. It really helps to visualise the picture plane, since the viewfinder effectively is the picture plane.
Now please, don’t start on the whole “cheating” thing. I can’t even be bothered to argue the point. Using a viewfinder like this is a great way to train your observation and isn’t much different than holding out your pencil to measure with. A tool is just that, you still have to do the painting yourself!
One of the view-finder’s most useful features is helping me make sure that I keep a consistent viewpoint on my subject right the way through the painting. That’s pretty much a prerequisite for sight size. By sighting through the viewfinder and lining up the threads with particular parts of the subject, I can make sure I get back into exactly the same point each time I look at the subject. Handy.
If you’ve never tried this, it will surprise you how much you move about when you paint or draw, each movement showing you a slightly differing viewpoint on your subject.
For this demo, I’m just using a sheet from a Windsor and Newton canvas pad, fine for little studies like this. I’ve stuck the gridded up sheet to a drawing board on the easel. The viewfinder is stuck to the back of the board with blu-tack.
This pic shows a little more of the set up, although it’s not taken from my viewpoint as I painted it. The still life itself is sitting in a simple shadow box made of black foam board. I’ve stuck a sheet of tinted drawing paper to the back of the shadow box to put a bit of colour into the background.
The art of comparison
Notice here how the painting surface is right next to the subject. Here is perhaps the most useful feature of sight size: direct comparison. If I want to check the shape of a patch of colour, say, against the corresponding part of the subject, I simply have to flick my eyes from one to the other. You can do it repeatedly and quite quickly, giving an effect like a flick-book animation. It really helps to show you where your shapes might be out.
I also need to be a little distance back from the painting when I’m comparing, or distortions in the drawing will occur. Because this is a little painting (7 X 5 inches) I don’t need to be too far back. I find that if the painting surface is just out of reach of a long brush, I’m far enough back not to introduce too much distortion. I want to actually have to move forward to be able to make a mark.
Balancing the light
One more point about the light. This is really important.
Because my painting surface is right next to my subject, it has the same light falling on it. But there’s one little tip I find really useful that I want to share with you about values. I spend a bit of time with a three step scale on the painting surface, and one in the subject.
By comparing the white swatch on the easel with the one in the subject, and making sure they match, I give myself a much better chance of getting good values in my painting.
In order to do that, I need to angle my easel slightly towards the light. If you’re working with artificial light, you can do the same thing more easily, to an extent, as long as you have a separate light for your easel and one for your subject. Either way, it will really help you if you take the time to balance the light.
If you don’t, you may well find that the lightest light you can get on your canvas is noticeably darker, perceptually, than the lightest light in the subject. That can easily throw all your values out.
This point may be arguable (what isn’t, in painting?) but I generally find that it really helps to make sure you can get visually near to the lightest lights in your subject. Otherwise, when I paint the little white pot I have in this set up, it’s going to come out looking very grey. Because I can hit the other, middle values I see with reasonable accuracy, since they’re well within the value range of paint, I’ll tend to end up with a painting that’s too light overall and compresses the light end of the value range. Most likely, the realism of the piece will suffer because the balance of the values will be out.
Spending a little while making sure you have sufficient headroom in your lights makes a huge difference to the balance of the values in the painting, I find.
But bear in mind that I still can’t match everything I have in my subject here with paint. I can’t match the range from the darkest shadow to the lightest light on the pot. I’m just making sure that I don’t make the challenge more difficult than it needs to be. I find it’s very rare that you find yourself in a position where you can actually match everything you see. In fact, what we’re really doing when we paint realistically is making something new, something a little different than what we see.
That’s why painting realistically is less simple than it seems at first. The illusion of reality in a painting, most of the time, means employing some strategies to create a more realistic effect, not simply replicating what you see. This is where the “you’re just painting what you see” crowd have it exactly backwards. In my experience, you’re painting something deliberately different than what you see, partly because you can’t actually match it, and partly in order to create a more convincing effect of reality.
So here’s the accuracy part. You knew it was coming. But I’m not sweating over it, like a cast drawing. Actually, this goes in pretty quickly, and I’m not being obsessive about the accuracy of each mark. But because I’m working sight size, I know that everything is in pretty much the right place, I won’t have to make any major corrections later.
Much of this I’ll actually be drawing with the brush as I go, so I don’t need to produce a laborious drawing. I do need to make sure that things are in the right place, though.
By the way, the above shot isn’t taken from my viewpoint either, which is why the drawing and the subject don’t line up – this is actually a still from the video, as most of these pics will be.
Background and ground
Next I paint the background. Here, I’m helped by being able to make a direct comparison between my subject and my painting – the most useful aspect of sight size.
It immediately shows me that I can’t quite reach the darkest darks. But as long as I’m pretty close with the mid values (the upper left of the background is closer to a mid value here) I should be fine and won’t get into any value difficulties. I’ve scrubbed the background in with a bright (a short, stubby, thick flat bristle brush). It takes some muscle power to do it, but it results in a thin. even coating of paint.
After the background, the ground goes in – in this case, it’s the wooden chopping board that the objects a sitting on.
The value of the chopping board is around the middle of the range, so I can match this fairly closely. And because I spent some time setting up the balance of the values at the start, to make sure I can get close to the lights, I know I won’t get into trouble with my values if I do.
Judging colours is pretty difficult. All kinds of things conspire to derail our perception. Simultaneous contrast, high chroma interfering with our perception of value, neighbouring values throwing out our perception. It’s really not easy, which is one of the reasons it takes a long time to learn to paint well. Translating your perception of your subject into paint is complicated by the very equipment we use to see with – our brains.
Here’s a handy little method you can use to help judge colours: get a little clear acetate strip, about half an inch wide by about an inch or two high. Then make a little “colour isolator”, a piece of card with a hole in the middle.
When you’re trying to match a colour, paint a little swatch of the colour you’re mixing onto the acetate strip and hold it with the colour isolator against the subject, so you can see a bit of both.
This works best in the mid value ranges (the extremes will be outside your range of paint) and is a great way to get in the general ballpark of the colour you’re trying to match. The viewfinder is also useful here, since you can make sure that you hold the isolator parallel to it, ensuring that you always have the same amount of light falling on the acetate strp. This is important because if you angle the strip slightly one way or the other and the light on it changes, so does the colour. That will completely screw up your comparison!
Now bear in mind that this is a rough and ready approach. Very often you won’t be able to match the colour you see exactly, if the value or the chroma are beyond what you can achieve. But it is a very practical and useful method, because it gets you close – closer than you might be if you relied on your judgement alone. Once you have something near down on your painting, you can manipulate it as you like in order to create a more realistic effect in the painting.
Painting the objects
If everything has gone ok up to this point, this stage isn’t too hard. I generally start with the shadow planes, because they’re better painted more thinly and because the light planes can then be painted more thickly into them, creating a more solid effect. this is one of those things you can do that you don’t actually see in your subject, but can help create a more realistic effect in the painting by creating a feeling of solidity.
Once the shadow planes have gone in, the light and half-tone areas can be painted in a bit more thickly. I usually do this pretty roughly, because it’s easy to refine something that’s already working well with everything else, than to spend ages refining some area only to find you have to paint it again because the values or the colour are out.
Sight size comes into it’s own here again. By closing one eye and throwing the other out of focus, I can blur my vision, making it easy to make general comparisons between my painting and my subject. Anything that’s really out will jump out at me, and I can correct it at this stage.
I can also begin to work more exclusively just on the painting itself. I know I can’t match the full range of the values in my subject. I know my little white pot is a little darker than the actual one. More, the white pot has a reflective surface, so if I want to be able include the little highlights of reflected light (which will greatly help the feeling of realism) I need to drop the value of the light planes of the pot a little in order to help those little highlights stand out.
Because those little accents of reflected reflected light are actually little light sources in themselves, I can’t get anywhere near how bright they really are. So I have to suggest them. William Nicholson, one of my favourite still life painters, is really good to look at for this. Here’s a couple of examples of him cleverly manipulating value in order to make those little reflected highlights stand out (technically it’s called specular reflection, if you’re interested).
A word here also on the colour of the apricot at the front of my still life: The local colour of this apricot was quite easy to match with paint, I could get higher chroma if I needed to, in the same hue, by mixing together cadmium yellow and cadmium orange. But the local is far from the whole story.
Perceptually, when I tried to match the colour of the light planes of the apricot on my painting, I had to push the chroma higher than the local. I think this may be because the light planes of the apricot are angled more towards the light than my painting surface, so are reflecting more light resulting in higher chroma at the value I need to match (lighter than the local).
Whatever the reason, I had to push the chroma of the apricot to get something near my perception of it. In the event, I actually pushed it a little higher than I perceived it, which gave the apricot a little more presence in the painting and seemed to bring it forward somewhat, increasing the feeling of depth and space in the painting.
The light planes of the apricot also have the thickest paint surface. It’s more obvious in the flesh than in a photo, unfortunately. You’ll have to take my word for it 🙂 That tends to give it a certain solidity and physicality too, which gives the impression of it coming forward in the painting.
The shadow planes of the apricot were interesting too. Here, I couldn’t get the chroma I needed at the value of the shadows, and keep the hue constant across the surface of the apricot. So to get more chroma, a moved the hue a little towards red and away from yellow. With paint, you can get more chroma in lower values in reds than you can in yellows.
Generally speaking, I’ve found that if an object has a single hue, that hue tends to remain constant across it’s surface from light to shadow (assuming no influence of coloured light either in the light planes or coloured reflected light in the shadows). So in this painting, I deliberately sacrificed some of the hue in order to get higher chroma shadows.
Now, I believe that painters who’ve been painting for a long time tend to make these sorts of adjustments from experience. Perhaps often, not consciously knowing that they’re making them. Working sight size, with my little colour-checking method, I can see exactly where I’m having to compromise and make decisions about what I’m putting into the painting that differ from what I’m seeing – always, in order to try to create a more life-like, more realistic impression.
By the way, you might also notice that my apricot is far from perfectly drawn, and doesn’t exactly match the apricot in the subject. That concerns me not at all, since it does look solid and real in the painting. That’s what I’m most interested in here. But when you’re n the early stages of learning to paint, I think it pays to push yourself for as much accuracy of shape and proportion as you can.
It’s worth noting that I haven’t been nearly so slovenly in the drawing of the pot which, whilst not perfect, needs to be more accurate since it’s geometrically symmetrical and will look obviously wrong if I don’t watch it!
At this stage, it’s really just refining, and you can take it as far as you like. Because sight size has helped me get this painting closer to my visual impression (within the very real limitations of oil paint) than I might have managed otherwise, and because I’ve spent some time trying to use that as a foundation from which to create a more realistic effect in the painting, I’m at liberty to choose the kind of finish I’d like.
I could keep it very brushy and chunky, or I could refine it more and more, smoothing and adding finer and finer detail. I see level of finish as an entirely personal thing. On the one hand, you have the high-finishers, who produce a smooth surface with lots of fine detail. The Dutch still life masters of the 17th century are a good example if this, and lots of contemporary Dutch still life painters still follow that tradition. I think they produce(d) incredibly beautiful work.
Or you could leave the painting brushy, perhaps even making a feature of the brush strokes. Lot’s of painters work like this today. A good example of loose, brushy painting is Velasquez, in his later years:
Sargent is a great example, of course.
Whichever kind of finish you’re most drawn to, if the piece is underpinned by sound drawing, values, colour and edge handling, and if an effort has been made to make it something different from a replication of the visual impression (and so more “real”) then it will work.
I think this is especially important for brushy work. If the values and the colour aren’t good, brushy painting simply doesn’t work, it looks messy. To an extent too, though, the same holds true for highly finished, detailed painting.
More than just measuring
If you thought sight size was just for ateliers and classical realism, I hope this post has at least given you a slightly different perspective on it. The benefits I got from doing this little demo sight size were:
- Direct comparison of shape, leading to more accurate drawing out.
- Concentration on the overall effect. Being further back from the painting and the subject helps to keep in touch with how the whole picture is developing as you go, rather than getting bogged down in detail too early.
- Direct comparison of the value range available in the painting against that in the subject – crucial in showing you where you need to do some thinking about how you’ll handle the values to create a realistic effect.
- Seeing the differences between the subject and the painting, especially in value terms, helps you see where you can and can’t match what you see. I believe it also helps with showing you that need to do more with painting to create a more realistic impression of depth and space – to make the painting look more real. I can also see where I’m struggling with being able to match the chroma in places, and can make adjustments accordingly. This really goes back to the point about direct comparison.
So here’s another quick list of the things I did in the painting that differed from my visual impression of the subject, in order to get round the limitations of paint and hopefully create something that looks more real:
- Thought carefully about how I would treat the values, since I couldn’t match what I saw – particularly on the highlights on the pot
- Punched up the chroma of the apricot to make it appear to come forward more
- Painted some areas thin and others more thickly, to increase the feeling of solidity of the objects
- Painted some areas smoothly (the background) and others with more texture (the apricot), to create a feeling of solidity
If you thought sight size was just for Bargue and cast drawings, I hope this little demo has given you a slightly different impression. I hope it’s shown you how it can really help you get a more realistic result in your still life painting, and that it doesn’t necessarily mean highly finished work.
Ultimately, I think sight size is about the overall visual impression, and can be a really effective way of getting more realism into your work, if approached in the right way. Sight size in particular is often criticised by people who don’t understand it as merely painting what you see.
In many ways, I think the real value of sight size is that it shows you why – and where – you can’t paint what you see. That helps show you where you need to think carefully about how you handle particular passages of a painting, so that you can, somewhat paradoxically, create something that looks more real.
This Thursday the 21st of January, at 8PM GMT (pacific and eastern) I’m giving a webinar in which I’ll go through the process of building up a still life like this live on camera. There’ll be a chat, with a live Q&A at the end, so if you have any questions about what I’m doing you can ask them and get an answer straight away.
Click here to register for the webinar. (it’s free 🙂 )
Best wishes and thanks for reading,
Update: January 21st, 2016
I’ve just had an email from Daniel Graves, founder of the Florence Academy of Art. He was concerned that I was suggesting that the methods I’m describing above (and the still life painting online course I’m putting together) are a substitute for three to five years of full-time study at an atelier.
They’re not, of course, and never could be. It was never my intention to suggest that they were, apologies if I gave that impression. As Daniel rightly pointed out to me in his email, at an atelier you will follow courses in anatomy, perspective, composition, colour theory and more. You also have the opportunity to study alongside accomplished artists, a significant drawback of working on your own at home.
No online course could possibly be a like-for-like substitute for that. To be clear (which I probably haven’t sufficiently been, yet, since I’m still working on it) the course I’m developing is for people who paint at home, like me, and is designed to help with painting more realistic still life. I believe that’s perfectly doable in an online course. The methods I’ll be demonstrating are easily replicated in a home studio, and the course will include practice exercises to help with the development of specific skills, and feedback to check that those skills are being developed.
The concepts involved are applicable to other areas of art too, inevitably, but it’s certainly not intended to be a substitute for a full atelier program. The goals of the course are much more limited and specific.
Free Value Tutorials
Subscribe: Join over 10,000 other artists and get free updates. I'll also email you THREE FREE value tutorials that will help you bring your pictures to life.