KPop Demon Hunters was largely animated by Sony Pictures Imageworks in their Vancouver and Montreal studios.
The film has tax credit money from Quebec, British Columbia and Canada — see the end credits.
Note there will be a KPop Demon Hunters Sing-Along limited theatrical event for one weekend only in the US and Canada on August 23 and 24. See SingKPopDemonHunters.com for more.
My take: kudos to Netflix for bankrolling this production.
“To resolve distribution issues in a sustainable manner, it is essential to tackle the structural problems throughout the entire value chain, both upstream and downstream of distribution.“
My take: this is a fascinating, if depressing, report on the state of film distribution in Canada. The best part, in my estimation, is the exhaustive list of (almost) every distributor in Canada, found in Appendix A on Page 25.
“You cannot compare street food to Michelin. Is Michelin food going to take over street food? No, not a chance. But there is an appetite for street food. There is an appetite for Michelin dishes.”
But she forecasts:
“People keep saying, you know, ‘TV is dying.’ It’s not dying. It’s just moving to the verticals, moving to the fragmented viewing. So if people move, you have to move with them, otherwise you’ll be eliminated.”
My take: You could say that TikTok forced Meta to introduce Reels to Facebook and Instagram, and Google to introduce Shorts to YouTube. Does that mean the Chinese love of microdramas is guaranteed to play out in North America too? I hope not. (I did like six-second Vines though.)
I’ve generated 30M+ views in 3 weeks using this exact workflow:
Write a rough script
Use Gemini to turn it into a shot list + prompts
Paste into Veo 3 (Google Flow)
Edit in Capcut/FCPX/Premiere, etc.
Concept
Kalshi is a prediction market where you can trade on anything. (US legal betting)
I pitched them on a GTA VI style concept because I think that unhinged street interviews are Veo 3’s bread and butter right now.
I guarantee you that everyone will copy this soon, so might as well make it easy and give you the entire process.
Script
Their team give me a bunch of bullet points of betting markets they wanted to cover (NBA, Eggs, Hurricanes, Aliens, etc)
I then rewatched the GTA VI trailer and got inspired by a couple locations, characters, etc.
Growing up in Florida…this wasn’t a hard script to write, lol.
Prompting:
I then ask Gemini/ChatGPT to take the script and convert every shot into a detailed Veo 3 prompt. I always tell it to return 5 prompts at a time—any more than that and the quality starts to slip.
Each prompt should fully describe the scene as if Veo 3 has no context of the shot before or after it. Re-describe the setting, the character, and the tone every time to maintain consistency.
Prompt example:
A handheld medium-wide shot, filmed like raw street footage on a crowded Miami strip at night. An old white man in his late 60s struts confidently down the sidewalk, surrounded by tourists and clubgoers. He’s grinning from ear to ear, his belly proudly sticking out from a cropped pink T-shirt. He wears extremely short neon green shorts, white tube socks, beat-up sneakers, and a massive foam cowboy hat with sequins on it. His leathery tan skin glows under the neon lights.
In one hand, he clutches a tiny, trembling chihuahua to his chest like a prized accessory.
As he walks, he turns slightly toward the camera, still mid-strut, and shouts with full confidence and joy:
“Indiana got that dog in ’em!”
Trailing just behind him are two elderly women in full 1980s gear—both wearing bedazzled workout leotards, chunky sneakers, and giant plastic sunglasses. Their hair is still in curlers under clear plastic shower caps. One sips from a giant novelty margarita glass, the other waves at passing cars.
Around them, the strip is buzzing—people filming with phones, scooters zipping by, music thumping from nearby balconies. Neon signs flicker above, casting electric color across the scene. The crowd parts around the trio, half amazed, half confused.
Process
Instead of giving it 10 shots and telling ChatGPT to turn them all prompts, I find it works best when it gives you back only 3 prompts at a time.
This keeps the accuracy high.
Open up three separate windows in Veo 3 and put each prompt in there.
Run all three at the same time.
3-4 min later, you’ll get back your results. You’ll likely need to change things.
Take the first prompt back into ChatGPT and dictate what you want changed.
Then it will give you a new adjusted prompt.
Let that run while you then adjust prompt 2. Then prompt 3. Usually, by the time you’re done with prompt 3, prompt 1 has its second iteration generated.
Rinse and repeat for your whole shot list.
Tips:
I don’t know how to fix the random subtitles. I’ve tried it with and without quotes and saying (no subtitles) and it still happens. If anyone has a tip, let me know and I’ll add it to this post.
Don’t let ChatGPT describe music being played in the background or it’ll be mixed super loud.
If you want certain accents, repeat “British accent” or “country accent”, etc. a couple times, I’ve found that it will do a decent job matching the voice to the face/race/age but it helps to prompt for it.
Edit
Editing Veo 3 videos is easy.
Simply merge the clips in CapCut, FCPX, or Premiere, and add music (if necessary).
I’d love to know if anyone has found good upscale settings for Veo 3 in 720p. My tests in topaz made the faces more garbled, so I try and cover it with a bit of film grain.
I like to add a compression/bass to the Veo 3 audio because I find it to be “thin”.
Cost and Time:
This took around 300–400 generations to get 15 usable clips. One person, two days.
That’s a 95% cost reduction compared to traditional advertising.
The Future of Ads
But just because this was cheap doesn’t mean anyone can do it this quickly or effectively. You still need experience to make it look like a real commercial.
I’ve been a director 15+ years, and just because something can be done quickly, doesn’t mean it’ll come out great. But it can if you have the right team.
The future is small teams making viral, brand-adjacent content weekly, getting 80 to 90 percent of the results for way less.
What’s the Moat for Filmmakers?
It’s attention.
Right now the most valuable skill in entertainment and advertising is comedy writing.
If you can make people laugh, they’ll watch the full ad, engage with it, and some of them will become customers.”
The BTS:
My take: high energy, for sure! That’s one detailed prompt for a three second clip.
In honour of our upcoming Canada Day, I thought it would be patriotic to take a look at the biggest all-time world box office numbers for Canadian movies produced in the last ten years.
Confused by these titles? Paddington in Peru? Movies are international productions now and get certain work done in tax credit-friendly jurisdictions for the money – and the excellent work skilled artists and technicians provide, of course.
CAVCO certifies Canadian movies. The CRA then refunds money to the Canadian producers of these movies, to the tune of millions of dollars. (Wouldn’t it be nice to know which movies Canadian taxpayers are helping fund and for how much? Other than checking the credits, I don’t know of a comprehensive list. But here’s a cost report from BC.)
Is Canadian investment in Canadian movies worth it? Check Cinema of Canada to see how small the Canadian production and box office numbers actually are. And direct employment in Canadian movies is less than 8,500 people. That’s a government subsidy of something like $25,000 per job. (Someone please tell me my math is wrong.)
My take:The Hollywood Reporter recently published a list of The 51 Best Canadian Movies of All Time – now where have I heard that particular number recently? Curiously, I don’t think any of these movies appear on their list.
Google Veo is arguably the best (but most expensive) AI video generator today. And Google Flow is arguably the best AI filmmaking tool built with and for creatives. Want to peak under the hood and reveal the prompts creating the magic? See Flow TV.
Amazingly, 10 questions accounted for 52% of the total. They are, quoting Stephen:
“1. What camera and gear should I buy for filmmaking on my budget?
The search for the “right” camera and kit never ends, no matter how much technology shifts. People want to know what will give them industry-standard results without breaking the bank. The conversation includes price brackets, compatibility, and whether brand or model really matters to a film’s success.
2. How do I start a career in film or get my foot in the door?
This is the practical follow-up to the film school debate. Filmmakers want straight answers about first jobs, entry points, and which cities or skills lead to real work. Many people are looking for pathways that do not depend on family connections or luck.
3. Is film school worth it or do I need to go to film school to work in the industry?
Filmmakers want clarity on the value of a formal degree versus real-world experience. They are trying to weigh debt against opportunity and want to know if there are shortcuts, hidden costs, or alternative routes into the business.
4. Which editing software should I use?
Software choice raises both budget and workflow issues. Filmmakers want to know which tools are worth learning for professional growth. Questions focus on cost, features, compatibility, and what is expected in professional settings.
5. How do I find cast, crew, or collaborators for my film?
Building a team is a constant sticking point. Most low-budget filmmakers do not have a professional network and are looking for reputable ways to meet actors, crew, or creative partners. Trust and reliability are major concerns, as is the need for effective group communication.
6. What is the legal, rights, permits, and music aspect of filmmaking?
Legal uncertainty is widespread. Filmmakers are confused about permissions, copyright, insurance, and protecting their work and collaborators. They want step-by-step advice that demystifies the paperwork.
7. How do I improve as a filmmaker, cinematographer, editor, writer, director, etc?
Self-development is a constant thread. Filmmakers search for the best courses, books, tutorials, and case studies. Clear recommendations are valued and people want to know what separates average work from great films.
8. Is my gear, equipment, location, or crew good enough for filmmaking?
Questions about minimum standards reflect deeper anxieties about competing in a crowded field. People want reassurance that their toolkit will not hold them back and want to know how far they can push limited resources.
9. How do I submit my film to festivals, distribute it, or what happens after my film is done?
People want clear instructions on taking their finished work to the next level. Festival strategies, navigating submissions, and understanding distribution channels are a minefield. Filmmakers want to know how to maximise exposure and what steps make the biggest difference.
10. How do I get feedback or critique on my work?
Constructive criticism is in high demand. Filmmakers want practical advice on scripts, edits, and showreels. They look for honest reactions to their work and advice on how to keep improving.”
My take: my answers:
The camera on your smartphone is totally adequate to film your first short movie.
Make your own on ramp by creating a brand somewhere online with a minimum viable product – you need to specialize and dominate that niche. Or move to a large production centre.
Maybe, if you can afford it and you’re a people person. Otherwise, spend the money on your own films because every short film is an education unto itself.
Davinci Resolve. Free or Studio.
Your local film cooperative. Don’t have one? Start your own.
Google is your friend. Don’t sweat it too much (and create your own music) for your first short festival films. As soon as your product becomes commercial, you need an entertainment lawyer on your team.
Watch movies, watch tutorials, make weekend movies to practice techniques, challenge yourself. Just do it.
See Answers One and Seven. Note: this is an audiovisual medium; audiences will forgive visuals that fall short but WILL NOT forgive bad sound. Luckily, great sound is easily achievable today.
He completes the package by taking us behind the scenes to reveal his workflow:
The software or services he used and their cost per month (or for this project)? See below:
Midjourney – $30 (images)
Gemini – free (prompts)
ElevenLabs – $22 (voice)
Hume – free (voice)
Udio – $10 (music)
Hedra – $10 (lip sync)
Premiere – $60 (NLE)
RunwayML – $30 (stylize)
Magnific – $40 (creative upscale)
Veo 2 – $1,500 (video at 50 cents/second)
Topaz – $300 (upscale) TOTAL – $2,002 (plus 40 hours of Tim’s time)
In addition to the great AI news and advice, Tim is actually funny:
“At some point in the process Gemini and I definitely got into a bit of a groove and I just ended up ditching the reference images entirely. I have often said that working this way kind of feels a bit like being a writer/producer/director working remotely with a film crew in like let’s say Belgium and then your point of contact speaks English but none of the other department heads do. But like with all creative endeavours you know somehow it gets done.”
My take: Tim’s “shooting” ratio worked out to about 10:1 and there are many, many steps in this work flow. Basically, it’s a new form of animation — kinda takes me back to the early days of Machinima, that, in hindsight, was actually more linear than this process.
1/ If you’re not using a LLM (Gemini, ChatGPT, whatever), you’re doing it wrong.
VEO 2 currently has a sweet spot when it comes to prompt length: too short is poor, too long drops information, action, description etc. I did a lot of back and forth to find my sweet spot, but once I got in a place I thought felt right, I used a LLM to help me keep my structure, length, and help me draft actions. I would then spent an extensive amount of time tweaking, iterating, removing words, changing order, adding others, but the draft would come from a LLM and a conversation I built and trained to understand what my structure looked like, what was a success, or a failure. I would also share the prompts working well for further reference, and sharing the failures also for further reference. This would ensure my LLM conversation became a true companion.
2/ Structure, structure, structure
Structure is important. Each recipe is different but same as any GenAI text-to something, it looks like the “higher on the prompt has more weight” rule applies. So, in my case I would start by describing the aesthetics I am looking for, time of day, colors, mood, then move to camera, subject, action, and all the rest. Once again, you might have a different experience but what is important is to stick to whatever structure you have as you move forward. Keeping it organized also makes it easier to edit later.
3/ Only describe what you see in the frame
If you have a character you want to keep consistent, but you want a close-up on the face for example, your reflex will be to describe the character from head to toe and then mention you want a close-up…It’s not that simple. If I tell VEO I want a face close-up but then proceed to describe the character’s feet, the close-up mention will be dropped by VEO… Once again, the LLM can help you in this by giving it the instruction to only describe what is in the frame.
4/ Patience
Well, it can get costly to be patient, but even if you repeat the same structure, sometimes changing one word can still throw the entire thing out and totally change the aesthetics of your scene. It is by nature extremely consistent if you conserve most words, but sometimes it happens. In those situations, trace your steps back and try to figure out which words are triggering a larger change.
5/ Documenting
When I started “Kitsune” (and did the same for all others), the first thing I did was start a Figjam file so I could save the successful prompts and come back to them for future reference. Why Figjam? So I could also upload 1 to 4 generations from this prompt, and browse through them in the future.
6/ VEO is the Midjourney of video
Currently, no text-to-video tool (Minimax being the closest behind) gave me a feeling I could provide strong art directions and actually get them. I have been a designer for nearly 20 years, and art direction to me has been one of the strongest foundations of most of my work. Dark, light, happy, sad, colorful or not, it doesn’t matter as long as you have a point of view and please…have a point of view. Recently watched a great video about the slow death of art direction in film (link in comments) and oh boy, did VEO 2 deliver on giving me the feeling I was listened. Try starting your prompts with different kinds of medium (watercolor for example), the mood you are trying to achieve, the kind of lighting you want, the dust in the rays of light, etc… which gets me to the next one
7/ You can direct your colors in VEO
It’s as simple as mentioning the hues you want to have in the final result, in which quantity, and where. When I direct shots, I am constantly describing colors for two reasons: 1. Well, having a point of view and 2. reaching better consistency through text-to-video. If I have a strong and consistent mood but my character is slightly different because of text-to-video, the impact won’t be dramatic because a strong art direction helps a lot with consistency.
8/ Describe your life away
Some people asked me how I achieved a good consistency between shots knowing it’s only text-to-video and the answer is simple: I describe my characters, their unique traits, their clothing, their haircut, etc..anything which could help someone visually impaired have a very precise mental representation of the subject.
9/ But don’t describe too much either…
It would be magical if you could stuff 3000 words in the window and have exactly what you asked for, right? Well, it turns out VEO is amazing with its prompt adherence, but there is always a moment where it starts dropping animations or visual elements when your prompt stretches for a tad too long. This actually happens way before the character limit allowed by VEO is reached, so don’t overdo it, it’s no use and will play against the results. For info, 200-250 words seems like a sweet spot!
10/ Natural movements but…
VEO is great with natural movements and this is also one of the reasons why I used it so extensively: people walking don’t walk in slow-motion. That being said, don’t try to be too ambitious on some of the expected movements: multiple camera movements won’t work, full 360 revolutions around a subject won’t work, anime-style crazy camera movements won’t work, etc… what it can do is already great, but there are still some limitations…
“The filmmakers who get noticed are the ones who don’t wait for permission. They write, shoot, and edit their own work. Even a no-budget short is better than waiting for the perfect offer that may never come.”
