iiinova — where your digital self keeps evolving

My point in all this is that the arguments should be more honest on both sides (and often they aren't), and that once they are, it's easy to understand why it makes perfect logical sense for many people to relativize things. Not everyone has decided to draw the line here, and that's fine too.

000

We all draw a red line based on our own beliefs, and I understand that for many people, including you, that line is AI. With that in mind, of course it makes sense that you try to advocate for change.

100

But someone could equally argue that the environmental impact of everyone going vegan would be *several* orders of magnitude more significant than stopping people from using LLMs for hobby projects. And they would be right. So again, in the end it's just a matter of choosing your battles. 4/4

100

As long as the use itself isn't unethical, I think that's an individual choice about which battles you want to fight. For some people it's AI, for others it's veganism, fast fashion, fossil fuels, etc. And it's not that you're necessarily wrong for focusing more on one thing than another. 3/4

I'm also not specifically saying "phones are bad". I'm saying we live in an unethical society (i.e. capitalism), and being aware of the many things we use that were produced unethically doesn't necessarily mean you have to stop using them or that using them is inherently unethical. 2/4

100

Ironically, wasting resources to retrain models in an ethical way to get to the same point is arguably unethical when we already have them. Or in better words, I think it's no longer possible. I obviously would have preferred these models to be trained ethically, but that's a separate question. 1/4

100

Silly example, but can accept your phone (or clothing, chocolate, social media… you name it) was created unethically, and wish you lived in a world where it wasn’t, but still accept that using it is not inherently unethical. This is just a matter of choosing if it is or isn’t your battle.

100

Iván Delgado @toruzz.com · 25/09/2026

Mario now always centers himself when entering or exiting a pipe, so there are no more weird clippings. Also... transitions!

1184

I'm seeing this article a lot in my feed, and honestly, I think the arguments made here are very simplistic. So I've written a bit about it and also put some of my thoughts on AI in retro games into writing. toruzz.com/unlisted/202...

593

Iván Delgado @toruzz.com · 21/09/2026

Oh, and replying to your question: one nice thing about this approach is that I'm no longer limited to a particular console: not only DMG and GBC, but porting games from consoles like NES is possible now.

120

Iván Delgado @toruzz.com · 21/09/2026

I know some people feel very strongly about this, and I get it, but I still think this was a good move. At the end of the day, I didn't get as much interest from romhackers as I'd hoped for (not talking about fans, of course, they were super excited!), so removing all that complexity is a good thing

120

Iván Delgado @toruzz.com · 21/09/2026

Thanks! Everything in DX 2.0 will be included here (+ a bit more) Keep in mind that I've moved away from GB Bridge to native GBA ports (or "matching reimplementations" done by AI), which serves me as a base to later implement my code, pixelart, and enhancements. Read here: bsky.app/profile/toru...

111

Iván Delgado @toruzz.com · 15/09/2026

Thanks! Nothing planned at the moment, but I'm considering it

020

Iván Delgado @toruzz.com · 14/09/2026

Intro screen implemented, plus lots of tweaks here and there. What do you think?

117519

Iván Delgado @toruzz.com · 09/09/2026

Super Mario Land Advance is coming along pretty nicely, and I'm having a lot of fun designing and tweaking all the little details. I hope I can share more soon! In the meantime, here's the tentative intro screen

Start screen from Super Mario Land Advance, a port of the original Super Mario Land for the GBA

2398

Iván Delgado @toruzz.com · 09/08/2026

Super Mario Land Advance is coming along nicely! Here's an almost finished 1-1 (raw pixels in the next post)

2213

I understand, I have conflicting feelings about it, too. While I don't think it's realistic to expect anyone else to continue my work, I might still release some of it (the SPEC, at the very least).

000

...these projects actually happening or remaining ideas, I'd rather disappoint some people. Most importantly, I can now focus on the parts I enjoy most: manually crafting the pixelart, adding enhancements, and ultimately realizing my vision. (5/5)

230

I can't even put into words how much this simplifies a port like this, but to me working on the GB Bridge version seems almost stupid now. So, considering I have very limited time, and reducing months of engineering effort to a matter of weeks is the difference between... (4/5)

140

A few weeks ago, as an experiment, I used AI to automate the disassembly+rewrite of the game, ensuring frame-perfect accuracy and completeness is tested. To my surprise, it turns out you really can end up with a clean, native codebase in a relatively short amount of time. (3/5)

And of course I thought working on a SML port would be perfect to showcase GB Bridge. Truth is, with the advancement of AI coding agents, and considering how simple GB games are, that's no longer the best (and certainly not the only) way to make a project like this happen. (2/5)

Iván Delgado @toruzz.com · 02/08/2026

I want to talk about a big change in this project that may disappoint some people. As some of my friends know, I started planning GB Bridge many years ago because, at the time, it seemed like the only feasible way to "port" GB games to the GBA. (1/5)

241

Still a lot left to do, but I'm happy with how it's coming along!

64314

Iván Delgado @toruzz.com · 13/05/2026

Not a great candidate, unfortunately, so probably not

120

Iván Delgado @toruzz.com · 05/05/2026

It hasn’t, I’m always working on more than one project at a time. I’m currently reworking the palette printing system so it’s dynamic instead of a full set per-screen.

220

Iván Delgado @toruzz.com · 23/04/2026

I don't usually ask for donations or anything like that, but if you'd really like to support my work, a coffee or something is always appreciated. Thanks for asking! github.com/sponsors/tor... paypal.me/toruzz

010

Iván Delgado @toruzz.com · 23/04/2026

I did an interview about Super Mario Land DX that didn’t end up being published as planned, so I’m sharing it here. Hope you find it interesting! toruzz.com/interview/20...

2175

I mean, I'm naming a mid-step to port games from GB to GBA. I'd even say I'm naming a bridge between the systems. So yes.

- What I've shared so far are mockups, since they're the best way to communicate the end goal. The current POCs are still rough and a bit disjointed, so here's one to illustrate my point

1161

- The game can then use special registers to trigger native GBA events - Those events are implemented per game (graphics, palettes, extra effects, etc.) - This is not even close to automated. It's a framework for romhackers/artists (me) to give GB games the "Advance" treatment

1110

- The emulator doesn't target the GB or GBC, but a modified/fantasy GB-like system. That system + its spec is what I call GB Bridge - Porting a game requires a romhack that adapts it to GB Bridge (screen, inputs, layers, etc.), similarly to how DX hacks adapt GB games to the GBC

160

I'm seeing a lot of confusion, so I'll try my best to clarify what this is about: - I'm building a way to "port" GB/GBC games to the GBA. The end result is a native GBA game - This uses a custom GB to GBA emulator (think 'Classic NES Series' but for GB) + native GBA code

32812

Thanks! I'm surprised I didn't come across your poc sooner, looks super cool. I would absolutely have loved to see a fully fleshed-out version of it.

040

Yes? This is not meant for regular people, it's just a project that enables romhackers and artists to port games. And even then I explicitly stated how difficult this is to pull off...

150

Yes, something like that, but the emulator has code built specifically for each game so they can't be separated. At the end of the day, you get a real GBA game that you can run anywhere, including real hardware. It's the same as the Classic NES Series.

1100

It's tentative, but for now I'm using $FF58 and $FF59 for GBA events and $FF5A for JOYP2.

010

I think I didn't explain this very well, but the end result is a GBA game (as the emulator is *for* the GBA). Keep in mind this is just a concept for now, but if I (or anyone else) ported it, the game would be playable on a real GBA.

1390

What I've posted are mockups of the end goal, but I've tested key parts and they are feasible. I'm currently working on SML. Like most of my stuff this is long-term, so please don't expect releases anytime soon. That said, I hope I gave you something cool to look forward to.

1580

GB Bridge is basically a GBC with a different memory map, a bigger screen, and extra inputs. The real trick is that I use special registers to trigger native GBA routines from the emulator - meaning that, while very hard to pull off, I can push things as much as I want.

2791