Smoke, Sparks, and Twenty Years of Showing Up
VFX Artists React to Bad & Great CGi 91 (ft. Tim Miller)
The cheapest trick in the book is a little smoke comped in with some sparks flying around. That's not a confession of failure... that's the entire philosophy of building something that lasts.
Tim Miller built Blur Studio the way you build anything worth keeping. One shot at a time. One terrifying payroll week at a time. One cheesy star filter at a time.
He sat down with the Corridor Crew and walked through twenty years of CG animation... from early 2000s game cinematics rendered in 3ds Max with no global illumination, no ambient occlusion, no ZBrush... to directing motion capture sessions via iPad across continents during a pandemic. And the throughline wasn't technology. It was showing up.
The Art of the Workaround
Here's what strikes me about those early Hellgate London shots. The faces looked like chalk. The skin had no specular. Miller watched his own work and winced... "Ew, she looks bad there." But at the time? They thought she was amazing.
That gap between what you made and what you wish you'd made... that's not failure. That's growth you can measure. The artists who scare me are the ones who look back at old work and feel nothing. If your past work doesn't make you a little uncomfortable, you probably stopped pushing.
The team at Blur didn't wait for perfect tools. They composited smoke layers and spark effects over flat renders to create depth that the technology couldn't produce on its own. Miller's Quake name was literally "Cheesefilter" because he loved slapping a five-pointed star highlight on everything. His colleagues roasted him for it. He kept doing it because he loved the craft more than he feared looking silly.
That's the recipe. Love the work more than you fear the judgment.
Story Is the Hard Part
Miller said something that stuck with me like a splinter. His wife asked him why he didn't just do a straight action piece for the next season of Love Death + Robots. He paused. Then he realized... he'd done so much action in his career that it wasn't a challenge anymore. "Story is fucking hard."
Think about that for a second.
A man who directed Deadpool. Who built one of the most respected animation studios on the planet. Who has rendered war, apocalypse, and every variety of cinematic destruction... and the thing that still humbles him is narrative. Getting the why right.
His adaptation process for Love Death + Robots is a masterclass in the discipline of subtraction. He takes the original prose, drops it into Final Draft, and starts cutting. All the inner monologue. The philosophical conversations. The beautiful literary passages that work on the page but die on screen. He described adapting Bruce Sterling's "Swarm" and cutting, cutting, cutting... because "you just don't want to sit there and have people monologue."
Removing what you love in service of what works... that takes a different kind of courage than adding more.
The Empire He Didn't Build
This is the part that got me.
Miller used to dream of building an animation empire. He let that dream go. Blur Studio sits at about 105 people. He could scale to 400 or 500. He chose not to. Because at that size, "it's just cogs in a machine."
Instead, he gave episodes of his most personal project... his flagship show... to competitor studios around the world. Digic. Studios he admires. Studios that bring artistic perspectives he couldn't manufacture internally even if he tried. The variety in Love Death + Robots isn't an accident or a budget constraint. It's a deliberate act of creative generosity.
He made the show better by making it less his.
There's a leadership principle buried in there that most people miss. The instinct is to keep everything close. Control the output. Protect the brand. But Miller understood something deeper... the show's identity is its variety. Hoarding it would have killed the very thing that makes it special.
The Sphincter-Tightening Weeks
I love that Miller was honest about the terror. He described years where Blur was one lost bid away from layoffs. The Dante's Inferno cinematic... the job that kept the lights on... only came to them because another studio, The Orphanage, filed for bankruptcy. Miller didn't even know that at the time.
"If that piece of luck hadn't happened, I don't know, maybe some other one would have, or maybe I would have had to finally face that crowd of people and say, 'Sorry.'"
That's honest. No rewriting history. No "I always believed it would work out." Just... we almost didn't make it, and the reason we did involved factors completely outside my control.
And yet he kept showing up. Every single day. He said he's never once woken up and thought, "I don't want to go in today."
Not because the work was easy. Because the work was his.
What CG Still Can't Do
Miller asked Wren what the hardest remaining challenge in CG is. Wren said human faces. Miller said that answer was too easy. Wren thought about it and admitted... he wasn't sure anything was truly beyond modern CG artists anymore.
Miller's response was quiet. "That's kind of sad."
Because when nothing is impossible, what do you strive toward?
The answer, I think, is the same answer it's always been. The tools will keep improving. The rendering will approach photorealism and then surpass it. But story... story stays hard. Connection stays hard. Making someone feel something through a screen... that never gets automated.
Light doesn't fight darkness. It just shows up. And twenty years of smoke, sparks, and cheesy filters later... that's still the whole game.
A journey of a thousand miles begins with a single step, Padawan. Miller said that as a joke. But it's the truest thing in the whole conversation. You don't need perfect tools. You don't need an empire. You need the willingness to comp some smoke over a flat render, call it magic, and come back tomorrow to make it better. Whatever you're building... show up. The striving is the point. 💙
--- Source: https://www.youtube.com/watch?v=GIKqT9S2jPQ
From TIG's Notebook
Thoughts that surfaced while watching this.
google_doc_last_sync: '2026-04-03T21:00:50.682456-07:00'
It's a problem you think you need to explain yourself. Don't. To anyone.— TIG's Notebook — On Self & Identity
But how many new things have you let become old things without meaningful extraction?— TIG's Notebook — New Captures
Echoes
Wisdom from across the constellation that resonates with this article.
I could hire somebody else who would charge me the same rate but do it quicker.
USD in Bifrost - Lesson 3: Instancing - With Bifrost USD, you can quickly generate a large number of instances and customize their composition and display. In this tutorial, Yingying Z teaches you how to use a USD point instancer, together
Understand the role of share buybacks in inflating per-share valuations beyond organic growth