General approach to lighting
If you haven’t already, please do check my journal entry about the lighting for Paragon’s Riktor. Where that focuses on the general lighting setup, this one delves into more of the creative considerations and problem solving. Therefore I’ll be referring to the original article throughout this one, rather than repeating myself.
Plus, the more page views I get, the more I can shove the meaningless of such endeavours to my mental periphery.
Life hack – successful self delusion is purely a question of consistency.
Themes and mood
The Countess was the third character I chose to work with and the first female (at least, I assumed she’s female, probably because I’m a privileged monster). This was a conscious decision since I wanted to mix things up.
The previous pair comprised large, bold shapes, relatively simply silhouettes and strong, aggressive personas:
These turned out well, but as previously mentioned – I don’t like repeating myself.
Which is actually quite problematic, as the ability to repeat actions is pretty crucial for drummers. It’s been suggested that its omittance from my own skillset might be why my own band-mates always seem so exhausted and without hope. But I digress.
The benefit of such impertinence is that it can encourage diversity in approach, because you’re always looking for something new. Since I wanted to explore the scope of this lighting setup with a distinctly different character, the Countess was ideal.
We’ve got numerous smooth and subtly textured surfaces, intersecting and conflicting shapes, ornate details and a real sense of culture. Altogether there’s a ridiculous range of creative options (and potential technical ball-busters), while never detracting from a distinguishable silhouette.
Literally the exact opposite of anything in Lawbreakers. Shots fired.
It was important to embody this change in the themes and goals of my approach, also. Otherwise, I wouldn’t be learning so much as… well, not-learning and I did enough of that in my teens.
Where the previous images were all about emphasising and showcasing the profound, dominating physical presence of the characters in question, this would be more about the mystery. Working with intricate details to imply the character’s purpose and backstory, rather than depict them in a very bold, deliberate manner.
Fundamentally, this was going to be less about what the light allows the audience to see and more about what it hides from them.
So deep, I’ll be writing for Black Mirror next.
Unless watching paint dry comes up as an alternative.
I honestly don’t know how anyone sits through a single episode of that show without blowing their brains out… which would be oddly ironic.
As with Riktor, I worked on getting the composition right first. Unlike with Riktor, I did this within my lighting structure rather than a blank environment. This was because the lighting served a different purpose on this occasion, so I needed to consider it at a far more immediate point in the process.
And it was frankly, bloody frustrating.
Not because it was difficult to get something that looked good, quite the opposite. Many of the poses and lighting compositions worked nicely, so choosing one was tricky:
This is no bad thing, in and of itself. Because design can be a largely iterative process, responding to and realigning both creative and technical concerns is only natural. So long as your end goal is clear, such digressions can be productive rather than a means of procrastination.
I reminded myself of and clarified my creative goals, now that I knew what I was working with:
- Create a sense of mystery and narrative
- Portray the character as determined but not specifically good/evil
- Highlight as many geometric details as possible without washing out the overall lighting
Refining the composition
Most of the poses I’d experimented with up to this point worked with one of the above goals. None of them however, ticked every box and I like my boxes to be ticked.
The closeups for example, provided an opportunity to suggest character motivation and backstory. On the downside, it meant losing all those really cool details across the rest of the character.
Compositions which introduced more of the body faired better, as the silhouette was concise and many of the finer details really stood out. Based on this, a blend of the two seemed logical.
Ah, logic – my old nemesis.
I eventually settled for a pose not too dissimilar to that used on Riktor.
This provided a little bit of everything. The upwards camera angle gave the character’s pose a sense of strength and determination. It was far away enough from the body that you could see plenty of the silhouette, while being close enough to allow scrutiny of all the finer details.
With that finally, mercifully sorted – I could move onto the fun stuff.
Lighting the Countess
For the most part, Riktor’s lighting worked in a clear, step by step process: I sorted the key lights, then the back lights, followed by the fill lights etc.
The Countess involved a lot more back and forth. I’d nail the key lights, but then do something with the fill light which ‘inspired’ me to go back and make alterations. Or, I’d struggle to achieve what I was looking for with a particular light, so would focus my attention elsewhere in the hopes that doing so would remove the mental block.
Frustrating? Yes, like explaining to students why backing up is a good idea, for the tenth time that semester.
Having said that, as much as I will implore thoughtful planning, I consider myself a bigger advocate of resourcefulness, flexibility and responsiveness. Sometimes the plan fails, so when you can’t rely on it – ensure you can rely on yourself.
With all that in mind, I’ve organised the following breakdown into a summary of challenges rather than specific lights. This better represents my intentions, problems that occurred and how I responded.
Step One – Back lights
As with Riktor, a pair of lights were positioned either side of the head, illuminating key areas of geometry with the intent of creating a sharp, varied silhouette. I tried innumerable variations of light intensity and positioning to address this:
Each had their own strengths and weaknesses. I’d already entered a scenario where I was struggling to be decisive over what worked and what didn’t. I liked some parts, didn’t like others but couldn’t discern a clear sense of direction through either.
It’s not uncommon to end up in this situation when you’re undertaking any type of design work. It certainly won’t be the last time I wander into it during this article.
At this point, it can be tempting to ask yourself or someone else – which do you think looks best?
Personally, I think there’s no worse a question you can ask. Except perhaps – would you mind please checking my heaped piles?
Goddam, I miss Bullfrog.
Back to the subject at hand – whether or not something looks good is largely subjective. Walk the streets of Stoke-on-Trent (armed and recently innoculated, I would advise) and you’ll see countless people who think that a combination of shirts that are three sizes too small, pants that are three sizes too big and facial tattoo’s courtesy of the local blind amputee looks good.
Some people think Metallica T-shirts look good.
Hell, some even feel the same about Miley Cyrus. My point is – there’s no accounting for taste.
More importantly, concerning yourself with whichever looks best only detracts from what you’re setting out to achieve. We’re not trying to create something that looks good (though we’d like it to), we’re pursuing a series of clearly defined objectives. So let’s not concern ourselves with which looks best. Let’s concern ourselves with which better achieves the goals?
Which is a question I promptly asked myself… to no avail. Great, logic has failed me. It’s like the whole “The best person to be around ladies – is yourself!” bullshit all over again.
They all created that sense of mystery and hidden strength I was wittering on about initially, with very little between them. At this point, I figured perhaps I was addressing these lights too soon, that I would get a better idea of how they needed to be handled, with a little more context.
So, I settled on a rough approximation of what I knew I wanted and moved ahead.
Step Two – Key lights
Key light positioning needed to be elevated for consistency with other images in the set and in preparation for the particles. I often struggle with the concept of thinking any further than five minutes into the future, but in this case – I nailed it.
It quickly became apparent that one light source would suffice and that the bold shadows created by a high lighting intensity really worked to create the sense of mystery I wanted.
The trick at this point, was in establishing exactly where those shadows needed to be cast. Audiences discern a lot about who a character is from their facial expressions. What I chose to show and hide here, would essentially determine who this character is.
The above pair of images are the ones I eventually found myself considering. Again – this wasn’t a question of which looked best, it was a question of which achieved the goal.
I was happy to consider this – for the most part – done… mainly because I was hungry and needed the toilet.
Not that the two things are related.
Step Three – Return to Back lights
I returned to the back lights, because the key had essentially given me more to work with.
While I was happy with the key light, I was conscious of it not being a particularly strong focal point. This wasn’t helped by the intensity of the back lights practically washing it out.
The light on their left was moved to produce fewer strong highlights and its intensity was also reduced.Because of this we got plenty of really cool areas of controlled bloom across her armor and lighting across the image benefited from a bit more diversity.
With this, I was quite pleased. So I moved on, lest I improve it into an abject failure.
Like Lawbreakers. Back-handed shots fired.
Step Four – Fill Lights
The main problem created by the back and key lights at this point, was the total absence of illuminated surfaces across the majority of the character’s front.
Which could work in a different scenario. Howitzer in particular worked because of the emphasis on back lighting with pretty much nothing else.
But that’s not what I was going for with this one, so this needed to be addressed.
Once again for consistency, I turned to blue fill lights to compliment the keys. I did a shed load of tests and honestly, none worked. They were either too strong and detracted from the mood created by the back and key lights, or too weak and frankly achieved bugger all. In both cases, the entire area was just too washed out anyway. There was no depth and I wanted depth.
I remembered how with Riktor, it really came together when I identified the light on his armor.
Emphasising this immediately bought more life to the character, diversified and balanced the lighting overall.
I questioned whether such a thing existed on The Countess.
Spoilers: it did.
A simple, bright red stone surrounded by countless, beautifully intricate details. This could be the exact thing I was looking for.
Step Five – Detail Lights
I dropped a high intensity red light just above this piece of geometry and immediately got everything I had been missing.
It bought more emphasis to this area which balanced nicely with the key light. More importantly, it bled in perfectly from the character’s right hand side back lights while creating much, much stronger shadows on their left.
The whole image just feels much more alive than it did before. Every light has a purpose but also relevance within the scene.
With that sorted, I was eager to revisit the fill lights.
Step Six – Return to Fill Lights
I experimented with a few different approaches, mainly seeing how intense I could go with the lights before they detracted from the overall scene.
Rather than try to achieve what I was after with one light source, I created three of very, very low intensity. These were then positioned very carefully to highlight the pieces of geometry I felt would create the best sense of depth: shoulder, chin-strap/lower jaw, left eye socket.
Step Seven – Return to Key Lights Light
With those sorted, I went back to the positioning of my key light, as I felt a slight tweak of the shadows and highlights created would work better with the new fills.
You can see at this point how things started working together and the process became progressively easier.
Unlike Lawbr… nah, too easy. I’ve not even played the dam game, in fairness.
Step Eight – Return to Back Lights
Happy with that, I then ditched the (character’s) left hand side back light. Because it was bleeding into the blue areas more than I’d like and wasn’t really highlighting much of the armor, I moved it further back. This helped to bring out some of the finer details and further emphasise the shadows.
Now was also the time to introduce some volumetrics to give a better sense of atmosphere and context to the back lights.
I wanted the volumes to feel layered, rather than just a single spot light. To do so, I repeated a trick I used when creating Rampage: drop in a couple of spot lights, stick a cube in front to mix up the level of fog:
With that sorted, the primary lighting setup was complete.
I took the exact same approach here as I did with Riktor: hell fire, raining from above. This time however, I just went with two particle emitters rather than three.
Two’s company, three’s a crowd, after all.
Unless you’re… nevermind.
Depth of field
Given how close the camera is to the character, I felt the scene would benefit and feel more believable with the introduction of some subtle depth of field effects.
Originally, I bought the character’s face into focus.
Which was utterly moronic. I’d already said earlier on that I wanted to make the red gem thingy the main focus of the image. All the lights had been steadily built around this principle, so why was I now contradicting myself? Surely it made the most sense to focus on this object?
As it turns out – I struggle with the obvious.
This might seem like a very minor detail, but it does add further depth to the image.
Finally came the lens flares. As per usual, they looked utterly horrendous at their default values.
Once dialed down, they contributed just the right amount of atmosphere and even mixed up the darker areas.
With those sorted out, the Countess was finished:
The Countess was undeniably tricky to get right, but also more ambitious than the others I’d done up to this point. The goals I’d given myself this time involved much more nuance and complexity, so it was expected that more work would be involved.
I do suggest this to all designers from time to time – push yourself, try something different… unless it makes your eyes water, in which case – pray for forgiveness first.
But always be mindful of what you’re working towards. I’ve witnessed and experienced directly, many situations where people give themselves a technical/creative challenge, but get so wrapped up in learning all the new processes involved that the final product ends up missing the point.
You don’t want to become someone well versed in repeating button clicks alone. You want to become someone who is comfortable relying on their skills and resourcefulness to achieve their design goals, whatever they might be.
That’s a question of confidence and understanding, not just software knowledge.
Or perhaps that’s just me trying to justify the fact that I seem to forget everything. I mean, I honestly have to Google ‘image masks in Photoshop’ every time I need to use the dam things.
Now it just feels like I’m undermining a genuinely valuable point.
Perhaps just don’t listen to me. I’m irrelevant.