By Brian Curtis Brian Curtis Drawing From Observation

Drawing From Observation Is Just Looking Better Than You Probably Are

I used to skip the whole observational drawing process because I thought it was just about copying what you see. That changed when I actually sat down and tried to draw something without referencing an image or using a grid. It took me three hours to draw a coffee mug that still looked wrong. The problem wasn't my hand. It was my eyes. Observational drawing means looking at a real object or scene and translating what you see onto paper. That's the definition. The reality is it's a mental exercise that forces you to slow down enough to notice proportions, values, and edges that your brain normally filters out. Your brain likes shortcuts. It fills in gaps. It simplifies complex shapes into symbols. Drawing from observation is the process of overriding those shortcuts through deliberate, sustained attention.

By Brian Curtis Brian Curtis Drawing From Observation

Brian Curtis approaches this method with a focus on gesture and structural underpinning before value work. His system isn't complicated. You start with a light gesture line that captures the overall movement or direction of the subject. Then you map in the basic shapes and angles. After that comes proportion checking, which is where most people fail because they don't actually measure anything. The measuring part is straightforward. Hold your pencil at arm's length, close one eye, and use your thumb to mark off units along the shaft. Compare the height of your subject to its width. Record that ratio. Now you're not guessing. A lot of people just start drawing and fill the page without ever checking if their proportions are actually correct. That's why your drawings look off even when they feel right to you. Your brain lies to you constantly about spatial relationships. This is a documented cognitive issue, not a skill issue. One edge case I ran into that most tutorials don't address: overlapping transparent objects. I was drawing a glass vase with a pen inside it, and every perspective system I tried collapsed because the refraction shifted the apparent position of the pen. The workaround was to draw the vase first as an opaque form, then treat the visible portion of the pen as a separate element and align it to the vase's perspective grid rather than trying to account for the glass itself. The refraction distortion was minimal enough to ignore at that scale, but it would have wrecked the drawing if I'd tried to be perfectly accurate about it.

The Actual Process

Here's how I run through a session now, usually about 45 to 90 minutes depending on the subject: First, I set up the subject with directional lighting. Flat light kills form. A single window or a lamp positioned to one side is enough. Harsh shadows are better than dim ambient light for learning to see edges and values. Then I do a gesture. Ten to thirty seconds. Just capture the energy and tilt of the object. Don't render. Don't outline. Let the pencil move freely across the paper.

Next, I block in the major shapes. Circle, square, triangle, cylinder. Everything reduces to these. If you can name the basic geometry, you can draw it. This step takes maybe five minutes on a moderate study. Proportion check comes next. Measure height-to-width ratios. Mark key angles. Compare distances between reference points on the subject. This is the step people skip, and it's the step that determines whether the drawing works or falls apart later. After that, I refine the contours. Not every edge is the same. Some are hard, some are soft, some disappear into shadow. Varying edge quality makes a drawing read as three-dimensional rather than like a flat sticker on the page. This differentiation comes from actually studying the subject, not from applying a technique blindly.

Value work follows. Start with the darkest darks and work outward. Build up midtones before getting caught up in the lights. A common mistake is starting too light and never anchoring the drawing with sufficient contrast. Once the values are in place, you can adjust relationships. Pull a shadow darker. Lift a highlight. These adjustments are always comparative. You're not setting absolute values. You're setting relative ones.

Things No One Tells You About This Method

Your first twenty or so drawings from observation will look terrible. Not frustratingly close to good. Actually terrible. This is normal. The skill gap between what you see and what your hand produces shrinks over time, but it doesn't disappear. Even professional illustrators who do this daily make mistakes. The difference is they catch them earlier and they know how to fix them without starting over. Another counter-intuitive point: sometimes the best thing you can do is step away from the drawing for a few minutes and come back with fresh eyes. What looks correct while you're staring at it fixes itself the moment you break focus. This happens because your brain stops accepting the drawing as a representation and starts seeing it as a collection of marks. Distance gives you objectivity. There's a bottleneck with this approach that most beginners hit around the six-month mark. You're drawing simple objects well enough, but when you try a complex scene or a figure in a dynamic pose, everything falls apart. The issue is usually that you've been practicing passive observation rather than active analysis. Drawing a still life from a photo is different from drawing from life. Photos compress depth and flatten values. Life has parallax, atmospheric perspective, and color variation that a two-dimensional reference won't show you. If your observational skills plateau, it might be because you're not challenging yourself with subjects that require deeper structural understanding.

I also want to mention one scenario where drawing from observation hits a hard wall. Complex mechanical subjects with repetitive precision parts, like engines or watches. Pure freehand observation struggles here because the margins for error are tiny. In those cases, I switch to a combination method. I'll block in the gesture and major forms from observation, then use measurement tools or light grids for the fine details. Nothing says you have to commit fully to one approach. Practicality matters more than purity.

What You Actually Need

You don't need expensive supplies. A standard HB or 2B pencil works fine for initial work. A kneaded eraser is useful for lifting values without destroying the paper surface. Paper should be medium tooth, not glossy. Something like a sketchbook from Strathmore or Canson will serve you. The only thing you absolutely need is a subject. Anything will work. A fruit, a shoe, a piece of furniture, a plant. The complexity doesn't matter at the beginning. Simple subjects teach proportion and edge control. Complex subjects teach patience and organizational strategy. If you want to follow Brian Curtis's approach specifically, his materials are available through his website and various art instruction platforms. His methods emphasize gesture-first thinking and structural clarity, which separates his work from more representational copy-the-reference styles. The actual downloadable resources include reference sheets and demonstration videos that walk through the process step by step. You can find those through the standard artist supply channels without any special access.

The core takeaway is that drawing from observation trains your perception. That's the real product. The drawing itself is just evidence that the training happened. If you can see better, you can draw better. The correlation is direct and well-established across every visual discipline, from illustration to industrial design to scientific illustration. The method works because it forces your brain to engage with reality instead of relying on stored symbols and assumptions.