Do We Want Our Engineering Visuals to Look Like Real Life?

Progress has been the status quo for technology over the past few centuries and for computers and graphics in particular, it feels as if the progress has been exponential. We watched as, in only a couple of decades, video game graphics progressed from pixelated 2D figures (my 4 y/o son calls these games ‘crumbly’ which I think is pretty accurate) with unrealistic movements to cinematic adventures that blur the lines of reality.

Visualization in engineering has seen similar progress from wire frame graphics to photorealistic animations in the past 20 years or so. It will only continue to get better and the highest end renders that are already amazing will soon look like actual footage from a GoPro.

But this begs the questions: do we want our engineering visuals to look like real life? Will it be beneficial or just straight up confusing if our graphics look like actual infrastructure?

When you look at architectural renderings- especially interior work- the quality is usually so good these days that the renders could be mistaken for the real world. But this is okay for these types of renders because the viewer knows that it is a render. They know that all of what they are looking at doesn’t yet ‘exist’.

The problem we run into with infrastructure type visualizations is that we are blending the real world with a conceptual world. If the viewer of our render doesn’t know where the real world ends and the conceptual world begins, we start to have problems. They may not even recognize that we are showing improvements and may just assume that they are pictures  or videos of an actual location.

This isn’t good.

Hollywood often pulls this off to the point where New York can look like it is post-apocalyptic by blending real footage with special effects but this is okay because there are usually zombies, robots or aliens as part of the scene to tell our minds that it isn’t real. Even if there are no ‘non-real’ objects to help with this problem, the viewer is watching the film with the expectation that it is not real. No problem here.

This is kind of like the ‘uncanny valley’ problem experienced by films and games but instead of our work feeling ‘unnatural’ like hyper-realistic humans can feel, our work runs the risk of just blending in with what is natural and our work may largely go unnoticed by the target audience.

So what can we do? Do we intentionally make our work look cartoonish or avoid working towards real life? Do we add artistic filters or keep our textures unnaturally clean and homogeneous so that the viewer will know that what we are showing isn’t real?

I say no. I say we continue to work towards real life as fast as possible. However, there is a solution. With the same power we are using to make a project look like it actually exists, we can add elements that will inform the viewer that it doesn’t exist. These can be elements like highlighted lanes, 3D text in-scene, deliberate cross sections, construction phasing animation… and so on. We want to be in control of when and where our projects look as realistic as possible while still informing the viewer which parts currently exist and which parts are proposed.

Because at the end of the day we are telling a story. We are communicating design intent to stakeholders or the public. And if our work is so good that it blends in with the real world and nobody notices?

Then we aren’t doing our job right.