Hello you lovely people! How are you?
Dang, I really wanted to have a video done in April. And it's getting there! But it's taking a bit longer than planned, for a few reasons.
The obvious first reason is my son - things are still going well with the little man, but he's simply taking up a lot of my time at the moment!
He's growing all the time, developing more of a personality. He giggles every day and it's such a joy to find some silly noise, face, or game that will make him crack up. He's just started teething though, which is impacting his mood, and his sleep. (And so, in turn, my sleep!)
The other reason is that this video is turning out to be a lot more complex than I was expecting.
The video was supposed to be a quick look at how GTA 3 pioneered the use of streaming technology to fit an entire city into the PS2's paltry 32MB of memory.
But there's a lot more to the topic than I first expected. Which required a lot of research. Loads of motion graphics. And a bigger script.
And on top of that - I decided that I really wanted a great visual to show how GTA streams the world in and out of memory. A birds-eye view shot of Liberty City with buildings and trees appearing around the player, and then disappearing in their wake.
But as much as I played around with mods, debug options, and noclip cameras, I couldn't make it work.
That's because (as I would eventually learn), the streaming is pinned to the camera, not the player. So I could never move the camera away from the streaming. No matter where I looked, the world materialised in front of me!
So I did the only sensible thing - I downloaded the decompiled GTA source code, rewrote chunks of the streaming code, and compiled my own GTA3 executable. All to make a really cool visual that you'll see in the video when it's done.
And then there's one more reason. I've been thinking a lot lately about how AI might change the way I make videos. Only... not in the way you might think.
Consider this: just a few years back, practically every video on YouTube was made by a person. Obviously. How else would a video be made? It wasn't something you even had to think about.
But in 2026, after an explosion in generative AI tech, it's now possible to conjure a YouTube video out of thin air, using LLMs to generate the script, voice over, and even editing.
Which means that, now, some of the videos on YouTube are artificially generated. And it's getting harder and harder to tell if a video is made by a real person. It's definitely happened to me: numerous times now, I've started watching a video only to realise it's just AI.
And I don't want to watch an AI generated video! AI makes mistakes, it regurgitates plagiarised work, and it writes in the most boring manner possible. YouTube is great because it's ordinary people sharing their passions, opinions, hobbies, and thoughts - in a way that feels unique to every creator's voice.
Basically, I'm finding myself feeling suspicious when watching a YouTube video... unless it has a person on camera, that is. That's much harder to fake, and the uncanny valley kicks in a lot quicker.
So, I realised that AI has made me suspicious of YouTube videos that feature a disembodied voice speaking over footage. Which is... not good for a person who makes YouTube videos that feature a disembodied voice speaking over footage!
So I feel like I might need to do a bit more on-camera stuff in my videos. And I'm starting with this GTA episode as a test, with a few clips of me talking to camera where it makes sense. I'm still figuring out the best way to do it (I definitely don't want to go back to the way I did things for Developing, which required a studio space (that no longer exists) and lots of time spent learning the script).
Anyway! That video will be out in a couple weeks. Hope you dig it.
After that, I'll be announcing the date for this year's GMTK Game Jam. I'll try to come up with some way to make this one a bit more special, seeing as it's the 10th edition of the jam! How time flies!
Outside of videos, I'm still struggling to find the opportunity to do much gaming. I played a little Cairn and I'm enjoying that so far. But I'm spending more time watching movies and reading. Lots of reading! A 4AM bottle feed isn't so bad with an audiobook in one AirPod.
I really enjoyed the book Recursion, by Blake Crouch. I liked the Apple TV adaptation of his other book, Dark Matter, so thought I'd read this. It's about a woman who makes a chair that lets you vividly revisit your memories - only, she discovers that she can change the events of the memories, and then live a new life from that point, on a new timeline. But when she catches up to the exact time she went into the past... the timelines converge and everyone is suddenly given a second set of memories. It's twisty, goofy, fun stuff.
I also read Fight Club. The movie was a big deal for me. It was the first film I watched that showed that movies could mess with the audience, and do surprising mind-bending meta-textual nonsense - and led me to films like Eternal Sunshine, Adaptation, and Memento. Anyway, I realised I had never read the actual book, so it was interesting to finally see the source text. I'll need to revisit the movie, now - I think it just came out on a 4K Blu-Ray.
And right now I'm reading The City and The City, by China Miéville. It's a detective story set in a "split city" - two nation states that exist in the same physical location, as residents of one city tactically "unsee" anything to do with the other. It's weird and imaginative, and has real Disco Elysium vibes at times.
Other than that, life is mostly just baby and work. Work and baby. Baby and work. That's okay - but hopefully I'll one day play a video game for more than 30 minutes.
Mark