Whatever we digital artists create and do, we always end up flat.

That is quite a bold statement for the first tutorial on this site, but let me explain my statement with an image:

What do we see? A sphere, of course. But is it correct? Let’s change the view a little bit by rotating around the sphere:

Ok, that didn’t work as expected. Why? Because it is an image, of course. Moreover I cheated a little bit with this example in hope to communicate my point (plaese bare with me): What we see in the first picture is of course a sphere. But in fact, what we see in the first image is a circle with a color fill that gives us the illusion of a sphere.

This fact has many implications but there is only one I am interested in at the moment: everything we 3D-artists do in our 3D applications of choice will end up in two dimensions of a screen or monitor. If you grasp this fact you can save a lot of work in the right moment.

The Virtual Cone

Ok, before we get down to business I will write some introduction first which will hopefully help you to understand what I try to achieve with this tutorial: always try to skip 3D. Admitted, that is another bold statement. But let me explain.

A 3D scene is in a way much more complex than a 2D image because of the added dimension. Let’s take a 2D image that consists of 400×400 pixels. If you multiply both axes of the image you get a whole of 160.000 pixels. This might sound like a big number but this is nothing compared to the full three dimensional numbers of the image. If you take the corresponding 3D scene of this 400×400 pixel image you can say that there is the image ‘catched’ by the virtual camera during rendering. Or, in other words, the virtual celluloid. But there is more to it. Since the camera catches something it has somehow to record its surroundings. This is where the virtual cone comes into play which is the visible area of a scene. Depending on the settings this cone can reach a very long distance. Therefore, this cone can add a lot of possible space you have to fill and therefore complexity. Even if the cone is only 100 ‘pixels’ deep you already get a theoretical 16.000.000 (16 million) cubic ‘pixels’ you have to fill with something.

So, although my advice sound strange in the best case and outright crazy in the worst case one can’t deny that 3D is way more voluminous than its 2D counterpart. And with this in mind we now can get down to business.

Time saver one: Draw your vision!

If you want to create a good or great looking image in 3D, go back to the roots and draw it. In other words, draw the image you want to build later in 3D. Because in the end you want to create a regular 2D image, you only use different techniques and methods to achieve this goal.

Old masters like Rubin took a brush and painted their vision – often after many sketches. You take a computer and build your vision in a three dimensional representation that is afterwards rendered in a file you can watch on your monitor and print.

To draw your image beforehand will save you a huge amount of time because of several advantages:

  • If you make a mistake in the concept drawing, you can correct it way faster than in 3D
  • You can draw different versions of your vision until it is perfect in your eyes
  • You only have model those objects or parts of them present in your concept

The last point is especially important. Working in a 3D application means that you have to create everything for your image. You have to model an object, you have to light it, you have to texture it. Each and every additional object or even single polygon means work and therefore time. If you already have a concept image at your hands you can save time since you can model only those parts that are absolutely needed for the image in that particular perspective. Of course this contradicts somehow the essence of 3D modeling but why should you bother yourself with a perfect modeled back of a head when it will never be seen in the final image? Or even parts of an object outside of your image? Skip the unnecessary work and use your time to further enhance the visible parts.

You can use this trick even when you don’t have any concept drawing to use as a reference. Simply model a rough outline of your model or your scene and then place the camera accordingly. If you have done that – and if you are absolutely satisfied with the view and composition – start modelling those things you can see through the camera. Delete the rest.

Time saver two: Skip 3D as much as you can!

Working in your 3D application of choice is often fun and creativity, but in most times it only is work. And most of the time it is a really time consuming work. Each change you make can cost time because in the worst case scenario you have to test render your scene after each subtle change. I know this very well because I lived through that sort of hell a lot of times in the beginning. Whenever I was not absolutely satisfied I changed some color settings and hit the render button. What did follow? Some waiting, of course. While some settings were absolutely okay after one change I often had to tweak the setting so often I cannot count them anymore.

Therefore, if the changes affect mostly coloring and subtle lightning save time and make those changes in your image editor. Not only will you get a faster feedback but you will make more changes because of that.

Time saver three: Enhance in 2D!

This tip overlaps strongly with the second one since it also deals with the fact that you have to avoid doing some things in the 3d world. So to save again more time time for better things you not only can change colors or subtle lightning (as already stated) but you can also add objects in your image editor or increase the quality of the ones you already modeled.  In other words: paint.

A good example for this tip is one image I created some time ago, Ethereal Dreams (see below). In the 3D scene for this image I tried for a long time to create a good looking waterfall. The results I got were never as comforting as I wanted them to be. Therefore, after hours and hours of struggling, I changed my strategy and went into my image editor of choice and drew it. Not only could I save much time but also got a result I would never have achieved in 3D without relying on particles (and even then it would be not guaranteed). The same applies for the clouds and mist.

So, as you can see, try to avoid 3D at any costs. Add and enhance in 2D. It will save you a lot of time.

This Tutorial in Action

To show you some practical results of this tutorial, I’ll show you an example. In the picture below you can see two versions of the same image. The upper one is the result I got from the renderer, split up in several separate layers. The lower one is the result after some hours in Photoshop including added objects (the birds, the mist, the clouds), enhanced objects (the waterfall) and color corrections (the whole thing):

Imagine how much time it would have took to create all the 2D parts in 3D. If I think about the particles only that would have been necessary for the mist I am still glad I switched to my image editor of choice.

Final Notes

Everything we do in our 3D application of choice ends up two-dimensionally on a screen or monitor. Therefore you can save yourself a lot of time if you switch to a 3D application as soon as possible in the process. Moreover you can use the benefits of the flatness to save yourself even more time (and use it for real fun like football, meeting friend, or writing tutorials). For example by drawing your vision, or enhancing your result in Photoshop, Pain Shop Pro, Color Draw or a program that you like and work with.

And don’t forget: have fun with your work!