Game of Tobi Has Twilight Princess Running on a Nintendo 3DS

Homebrew developer Game of Tobi has The Legend of Zelda: Twilight Princess running on a Nintendo 3DS in an early, partially playable state, built on the decompilation that brought the game to PC. The build is unofficial and carries visible bugs. Even so, it works as a proof of concept that a 2005 GameCube-era adventure can run at all on Nintendo’s 2011 dual-screen handheld.

A writeup of the project surfaced on July 2, 2026, citing Game of Tobi’s full 3DS port demonstration on YouTube. The port sits in what the developer calls “an early state and there are a lot of visual bugs.” It can already be played in parts of the game world, with bug fixes and full optimisation still ahead. Game of Tobi told fans the build would be released “when it is ready and I feel like it,” and encouraged other homebrew developers to attempt their own 3DS ports in parallel. Right now, the build is a working proof of concept that the game can load and run on the handheld.

Game of Tobi Has a Working Build on the 3DS

The build is unofficial and built on the same Twilight Princess decompilation source code that brought the game to PC earlier in 2026. That PC release, named Dusk, arrived May 9, 2026 on Windows, macOS, Linux, Android, and iOS. The 3DS version targets Nintendo’s handheld directly without going through emulation. Game of Tobi has framed the current state as a working proof of concept, not a finished product. The visual bugs and instability match a port that hasn’t yet been optimised.

Game of Tobi, the same homebrew developer behind prior Super Mario 64 and Minecraft ports for the Game Boy Advance, told the writeup author that the build runs in parts of Hyrule. “It is already recognisable, it can already be played in some areas,” the developer said in comments cited by the FRVR Blog writeup. The bugs cluster around rendering and stability, the same first-hours-of-debugging pattern every early homebrew port runs through. Long-session play isn’t yet reliable enough for a public release.

What it comes down to right now is just the work of fixing the bugs and of actually optimising the game. Because, so far, I haven’t done a lot of optimisation work, and there are incredible things you can accomplish if you actually do the optimisation work.

The quote comes from Game of Tobi, the homebrew developer behind the project, in comments cited by the FRVR Blog writeup dated July 2, 2026. Prior YouTube work on the same channel covers a Super Mario 64 Game Boy Advance port and a 3D Minecraft for the same handheld, both leaning on hand-coded optimisation rather than official Nintendo releases. The 3DS Twilight Princess build joins those two on a list of decomp-based ports from the same creator.

How a Five-Year Decomp Made This Possible

The Twilight Princess 3DS port rides on a decompilation that took the community roughly five years and four months to reach 100% match against the original retail GameCube disc. The project, hosted at the zeldaret Twilight Princess decomp repository, finished byte-for-byte matching on every regional GameCube build by December 2025. GameCube versions covered in that milestone are GZ2E01 (North America), GZ2P01 (Europe and Australia), and GZ2J01 (Japan). Work on Wii revisions, an Nvidia Shield port, and a debug build was still listed as in progress at last check. By early July 2026, the GameCube portions of the codebase were already the foundation of two PC ports. The 3DS project is the first handheld target to inherit that work since the decomp’s December 2025 wrap.

The first consumer result from that completion was a PC port named Dusk by developer TwilitRealm, with a second one called Courage Reborn by Linifadomra public-demoed but not yet released. Both fork the December 2025 decomp source as their starting point. The 3DS port sits in the same generation of downstream work from the decomp as those two consumer builds.

1. August 2020: Project founder Pheenoh (Howard Luck) creates the zeldaret/tp repository on GitHub.
2. November 2025: Project hits the 99.02% matching milestone for GameCube versions.
3. December 23, 2025: All three GameCube regional builds (GZ2E01, GZ2P01, GZ2J01) reach 100% matching.
4. May 9, 2026: TwilitRealm releases Dusk v1.0.0 on Windows, macOS, Linux, Android, and iOS.
5. July 2, 2026: Game of Tobi demonstrates a Twilight Princess build running on a Nintendo 3DS.

This build joins a pattern that has already played out twice on prior Zelda decomps. Once a decomp reaches high completion, a downstream port team forks the source and ships a playable build on new hardware. Ocarina of Time followed that path: the oot decomp reached high completion, then Ship of Harkinian appeared. Majora’s Mask followed the same approach. Twilight Princess is the third major Zelda decomp and is now following the same release curve, with the 3DS port slotting in as a downstream use of the same codebase.

The decomp gives every downstream port team the same thing: source code written in C++ that compiles back to a binary byte-identical to the original retail disc when paired with the original Metrowerks CodeWarrior toolchain for PowerPC. To target a new platform, a port team replaces the platform abstraction layer, retargets the build for the new CPU, and recompiles. The decomp proper is, by design, not a port itself; zeldaret/tp publishes only the source, not the game assets or assembly. Anyone porting has to bring their own copy of the game data, which limits any commercial-style release even as it leaves the technical path wide open. Total code volume on the repo is around 11.5 MB of C++ source plus 2.5 MB of data, with the decompilation timeline describing Twilight Princess as roughly 10× larger than prior Zelda decomps. The 3DS port inherits all of it as a starting point.

A Track Record of Putting Big Games on Small Hardware

Game of Tobi’s earlier projects include a working Super Mario 64 port on the Game Boy Advance and a 3D Minecraft for the same handheld. Both shipped as compiled binaries intended to run on real GBA hardware via a flashcart, not as emulated images. Each project leaned on the same playbook of hand-coded optimisation rather than emulation. Prior videos on the Game of Tobi channel documented the work from early prototype through to a playable build. The 3DS Twilight Princess port is now the third such decomp-based project from the same channel.

The 3DS sits in a specific slot in homebrew history. Nintendo abandoned the platform in September 2020 and closed the eShop in March 2023, leaving the hardware behind in the aftermarket but still capable and well known to the homebrew scene. The 3DS is the only mass-market Nintendo system ever sold with a built-in stereoscopic 3D display. In recent years, the 3DS has become a regular target for ports of games Nintendo never officially released on it. The Twilight Princess build is the latest in that line, and the first to bring a GameCube-era 3D adventure onto the system via decompilation.

All three of Game of Tobi’s projects share a single approach: read a finished decompilation as the basis, then retarget and recompile the source for hardware Nintendo never aimed it at. The 3DS port is the third major decomp-based project from the same developer. The build is also Game of Tobi’s first project that could end up using the handheld’s stereoscopic 3D feature, which the developer said would let the game be played in a way not possible on many other platforms.

What 3D Would Mean for a Twilight Princess Build

The headline possibility for any finished Twilight Princess 3DS port is the stereoscopic 3D feature, the Nintendo-built capability the original GameCube and Wii releases could not use. Game of Tobi said the project would eventually aim to render the game in true 3D once optimisation is complete. The same 3D effect has been disabled in some official 3DS titles when Nintendo judged the workload too heavy on the system. That is a sign of how demanding the feature is on the handheld’s GPU. If a stable Twilight Princess build reaches that point, the player would see Link and the Hyrule environment in layered depth rather than on a flat screen. The hardware that would deliver it is the 3DS’s autostereoscopic top screen, an 800 by 240 pixel display split per eye into two 400 by 240 views.

The 3DS hardware that would have to run Twilight Princess was not designed for a GameCube-era game’s workload. The system sits well below GameCube-era compute baselines on both memory and CPU, with GPU capabilities built around a small screen rather than a TV. GameCube code would have to fit inside that tighter envelope after optimisation work the developer says hasn’t really started. The ceiling for any unofficial port is set by the Nintendo 3DS hardware specifications. The official intensive 3DS ports that did ship show what those specs can support.

During the 3DS’s retail life, Nintendo did ship a number of intensive ports that pushed the same hardware harder than typical 3DS games did. Luigi’s Mansion was one of them, a GameCube-era title reworked for the handheld. Xenoblade Chronicles was another, and Nintendo restricted that release to the New Nintendo 3DS, which had the extra RAM and CPU headroom to handle the workload. The comparison is not exact because those ports came from Nintendo’s own engineers with full source access, but the precedent is what matters. An unofficial Twilight Princess port would be running on the same set of constraints.

Game of Tobi’s stated plan is to fix the visible bugs first, then run the optimisation pass that, by the developer’s own account, has barely started. The first public build went live on July 2, 2026, but no release date has been announced for anything beyond the current proof-of-concept demonstration. In comments cited by the FRVR Blog writeup, Game of Tobi asked other homebrew developers to attempt their own 3DS ports in parallel if they want a faster release. The developer also indicated the build would be released “when it is ready and I feel like it.” Anyone hoping for a finished Twilight Princess 3DS binary in 2026 has to accept that the timeframe is open-ended.

3DS at a glance:

• 128 MB FCRAM main memory
• 6 MB VRAM
• 268 MHz dual-core ARM11 MPCore CPU
• 268 MHz PICA200 GPU
• 75.94 million units sold across the 3DS family (as of December 31, 2025)

The 3DS Has Quietly Become a Decomp Playground

The 3DS has become a regular destination for fan ports of N64-era titles using decomps as their starting point, including at least one complete Super Mario 64 port in recent years. The Twilight Princess 3DS build extends that pattern to a major GameCube-era release. Each project has, in turn, helped validate the decomp source as a portable base rather than a research curiosity. Game of Tobi’s YouTube demonstration is currently the most visible record of Twilight Princess running on the 3DS. It is unlikely to be the last 3DS decomp port to draw a wide audience. The same community has its sights on other big GameCube-era titles still locked behind commercial release cycles on Nintendo’s other hardware.

Right now, the 3DS Twilight Princess port remains a proof of concept rather than a shipping build. Game of Tobi has not announced a release timeline beyond promising updates on the same channel. The wider pattern matters more than this one build: each decomp that crosses the matching bar opens a new path for handheld ports Nintendo never authorised. The 3DS specifically has emerged as a frequent target because the homebrew tooling for it is mature and the hardware is widely available secondhand. The 3DS Twilight Princess port joins Game of Tobi’s two prior completed projects, a Super Mario 64 Game Boy Advance port and a 3D Minecraft for the same handheld. The Twilight Princess decompilation timeline lists Wind Waker, Skyward Sword, and Breath of the Wild as the three next-likely candidates for the same downstream-port treatment.