When it comes to long-running modding communities, the Super Mario World ROM hacking one has to be the craziest. Starting in the 90s with folks on [site], and eventually moving over to its own site in SMW Central, the community hasn’t so much just modified the original game as they have turned into a Game Maker in of itself, with tools like Lunar Magic, GPS and PIXI letting hackers create almost anything in that engine.

It’s a truly incredible communnity, and one which has given us thousands of amazing hacks and games with all manner of interesting graphics, songs, enemies, objects and level designs.

And our interviewee of the day is one of the people pushing it to its very limits. Yep, as the title suggests, we’re interviewing Anonimzwx, creator of the DynamicX sprite system and hacks like Random Revolution.

Here are some examples of his work:

Yeah, he’s basically the Kaze Emanuar of SMW hacking, at least where assembly programming is concerned. He’s created enemies and bosses from Donkey Kong Country and Metal Slug. Recreated mechanics like the potion doors and subspace from Super Mario Bros 2, and even remade Dracula from Castlevania Symphony of the Night!

So today, we’re gonna find out how this is all possible. How one man can singlehandedly outdo the entirety of the 90s video game industry, implement mechanics not even thought possible on SNES hardware, and turn ROM hack programming into a full-time career. Let’s go!

The Interview

So first things first, who are you? Who is Anonimzwx?

I’m a SMW hacker and videogame developer, mostly known for creating ambitious and technically challenging projects. Some of my most notable works include Dynamic X – a powerful dynamic sprite system packed with features to handle complex entities efficiently; Dyxen – a tool for designing graphics, animations, and hitboxes for SMW enemies; Super Mario World Random Revolution X – a hack featuring high-quality Metal Slug bosses; and Super Mario World Random Revolution First Blood – a hack built entirely with pre-rendered graphics and rich Donkey Kong Country-style content.

Where did that username come from?

I started using that name a long time ago when I was playing Ragnarok Online. Honestly, I don’t remember exactly why I chose it — it just stuck ever since.

And how did you get into gaming? What was your first game?

I started playing video games when I was just 3 years old. The very first ones I played were Super Mario World and Super Mario Kart.

What ones are you playing at the moment?

I’m not playing anything actively right now since I don’t have much free time. I do have a few games on my backlog, like Metroid Prime Remastered and Donkey Kong Country Returns. Occasionally, when I get a bit of spare time, I play Age of Empires II.

What franchises are your favourites in general? Based on your works you seem to be really interested in Donkey Kong Country, Metal Slug, the Mario series in general, etc…

I’m a big fan of the Donkey Kong Country, Mega Man X, Metroid, Mario, and Metal Slug franchises. They’ve all inspired me in different ways, both as a player and as a developer.

Onto ROM Hacking in general now. How did you get into the hobby?

Back in 2012, when I was in university, I took a Computer Architecture course where I learned x86 Assembly. I wanted to put that knowledge to use, so I started watching SMW hacking tutorials on YouTube and gameplay videos of Brutal Mario. Seeing the amazing things that hack could do made me think, “I want to create stuff like that too.”

And why did you choose Super Mario World hacking instead of, say, making a Metal Slug, Mega Man or DKC hack?

Because Super Mario World is one of the easiest and most flexible games to hack. Other games don’t have as many tools or resources available, which makes hacking them far more challenging. For example, I’ve tried hacking titles like Fire Emblem: The Sacred Stones and Blazing Sword, as well as PSX games like Yu-Gi-Oh! Forbidden Memories, but the lack of tools, resources, and documentation creates a huge barrier.

What was your first project there?

My first project was an unfinished Kaizo hack called Super Mario World: Bowser’s Empire. I had some interesting ideas for it, but honestly, I’m not a Kaizo player, and testing each level was a nightmare. I quickly realized I prefer creating standard-difficulty hacks.

How did you get into assembly hacking and programming? Your SNES coding skills are absolutely insane…

I’ve been coding since I was 17 (I’m 33 now). Later, I studied Computer Civil Engineering at university, where I strengthened my knowledge of low-level programming and computer architecture. I love coding because it feels like giving life to something — and seeing the results is incredibly rewarding.

I have extensive programming experience and have been working in video game development since 2021. Currently, I focus on SNES ASM commissions. Some of my recent work includes:

1. The final boss of Castlemania II, Count Koopula.
2. The final boss for Arbitrium, inspired by the obscure SNES game Musya: The Classic Japanese Tale of Horror.
3. Commissions from the author of Invictus to create sprites and gimmicks for Invictus 2.
4. A custom player system for Super Mario World that supports fully reskinnable characters with custom animations, power-ups, and hitboxes.

One of your most impressive projects is a dynamic sprite framework called Dynamic X, which lets you create sprites with dozens of frames of animation and effects you’d never expect were even possible on the SNES. What made you decide to create this?

It all started back in 2013, when I wanted to create sprites with many animations and frames. At the time, SMWCentral used a system called DSX for dynamic sprites. I researched and analyzed it, but found it had too many limitations and issues for what I wanted to achieve.

This led me to develop my own system, Dynamic Z. It worked for a while, but SNES limitations still made it difficult to create complex enemies like those from Donkey Kong Country or Metal Slug. Over time, Dynamic Z evolved from a very basic version 1.0 to a more capable version 4.0, which could handle DKC-style enemies — but it still wasn’t enough for me.

One day, I had an idea that completely changed my approach to dynamic sprites. That’s when I decided to start from scratch and build Dynamic X, a new framework designed to push the SNES far beyond what the older systems could do.

And in simple terms, how does it work on a technical level anyway?

The main improvements from Dynamic Z over DSX were:

• 30 FPS synchronized dynamic sprites – allowing twice as many dynamic sprites on screen.
• Dynamic VRAM allocation – space is assigned more intelligently for sprites.
• Flexible sprite sizes – unlike DSX’s fixed 32×32 or 64×64 sizes, Dynamic Z supported any size.
• Conditional VRAM transfers – data is only sent to VRAM when necessary, instead of every frame.
• No RAM buffer – DSX filled a slow RAM buffer each frame, causing heavy slowdown; removing it made Dynamic Z much more efficient.
• Customizable VRAM space usage – the user can define how much space dynamic sprites take.
• Configurable transfer limits – control how much data can be sent per frame to prevent flicker.

From Dynamic Z to Dynamic X, the main breakthroughs were:

• Animation freeze-on-wait – instead of syncing sprites at 30 FPS, animations pause until their graphics are ready in VRAM, similar to how Donkey Kong Country handles it. This change is rarely noticeable to the player.
  VRAM as a pose cache – individual animation frames (“poses”) are loaded into VRAM and reused across sprites. When VRAM is full, poses that haven’t been used recently are replaced (least-recently used policy). This drastically reduces how much data needs to be uploaded.

• Multiple instances reuse poses – different copies of the same sprite can share the same VRAM poses, avoiding redundant uploads.
• Tile overwrite control – users can choose which VRAM tiles can be replaced by dynamic sprites for even more flexibility.
• Per-pose allocation – instead of reserving the maximum frame size a sprite might use, each pose only takes the exact space it needs. This is especially efficient for sprites that change size.

Other features present in both Dynamic Z and Dynamic X:

• Dynamic player graphics and palette changes – modify the player’s appearance and colors in real time.
• Sprite palette management – sprites can load their own palettes, reserve shared palettes, or auto-assign free ones, allowing you to bypass the SNES’s limit of 8 palettes for all enemies.
• Palette effects – apply real-time RGB or HSL changes, such as lowering the red channel, shifting hue, or adjusting saturation. Some Examples of use:
  • Day/night transitions.
  • Status effects like poison or frozen states.
  • The rainbow effect during Starman.
  • Creating visual effects without relying on limited hardware color math.
• Drawing System: A centralized way to do graphic routines with minimum effort.

With all these improvements, Dynamic X allows for sprites with dozens of animation frames, complex effects, and minimal slowdown — pushing the SNES far beyond what most thought was possible.

Do you hope the community will eventually make it the standard instead of DSX? Because it’d certainly make dynamic sprites more accessible if they did…

When I finished Dynamic Z v4.0, it was already far superior to the DSX system. I proposed to SMWCentral’s staff that they upload it and run a dynamic sprite remoderation to convert existing sprites to Dynamic Z (at the time, only about 30 dynamic sprites were hosted on the site). My proposal was rejected — some staff members even said, “Nobody cares about dynamic sprites, so this is totally unnecessary.”

Since then, I haven’t been concerned about whether Dynamic Z or Dynamic X is hosted on SMWCentral. I have my own website where I share my resources, and people who want to use the system usually contact me directly on Discord. At this point, I’m more of an outsider, and I don’t mind if it ever becomes the standard on SMWCentral.

Why do you think dynamic sprites seem so rare in the SMW hacking community anyway? Even after a decade it seems like barely anyone uses them, let alone codes any new ones…

Mostly because many still rely on the old DSX system. However, some users who have adopted Dynamic X have created a large number of sprites — for example, Stivi in Mario Stellar Destiny or .Cookie in his Kaizo hacks.

With Dynamic X, creating dynamic sprites is much faster and easier thanks to Dyxen, which handles most of the work for you. Personally, I’ve already made around 200 dynamic sprites, including enemies, projectiles, decorations, particles, and more.

Could DynamicX turn the trend around and make them mainstream?

I’m not aiming for Dynamic X to become mainstream on SMWCentral, as I prefer it to be used in projects that can truly take advantage of its capabilities. This way, hacks using it can stand out with sprites and effects that wouldn’t be possible otherwise. Over the last year, its use has grown, especially through my commissions — for example, bosses like Count Koopula in Castlemania II would be nearly impossible without a robust system like this.

Recently, you released a short Metal Slug themed hack called Random Revolution X, which has Mario fight all manner of bosses from that series. What made you decide to go that route?

It was an idea I’d had since 2013 — I wanted to create a hack that pushed the SNES hardware and Dynamic X to its absolute limits, and Metal Slug felt like the perfect franchise for that. The current release is just a demo, but the final version will also include content from Mega Man X, Metroid, Castlevania, and Contra, along with many more bosses.

How did you choose what opponents to include? O’Neil as the final boss seems logical enough, but we’re curious how the others fit in here.

Since SMWRR X was made for a SMWCentral contest, I had limited time. I avoided bosses that would require too much development time, so the Twins Tanks from Metal Slug Advance were a good choice — they’re simpler to implement than other franchise bosses. Allen was also a natural pick because he’s very iconic and has a wide range of poses to work with.

If I’d had more time, I probably would have chosen different bosses. Some of my favorites I’d like to include are Assy Nero and Big Shie from Metal Slug X, Jupiter King from Metal Slug 3, or Kessi II (Metal Slug 1 and 2) and Kessi III (Metal Slug Advance).

And how did you get the graphics and music so spot on here? Despite the limitations of the SNES, this feels like a Neo Geo or arcade game…

For the graphics, I received a lot of help from MetalYoshi94, who ripped most of them, and from a friend known as Klonoz, who worked on Mario’s portrait.

For the music, I was helped by Masterlink and Maxodex.

As for the sound effects, they were done using a sample streaming system developed by an anonymous friend — it’s a private tool that allowed me to achieve higher-quality sound effects on the SNES and bypass ARAM limits.

One aspect of the Metal Slug series it captures quite well is the franchise’s brutal level of difficulty, with Mario dying extremely quickly against a barrage of difficult to dodge attacks. Do you prefer your games to be on the more challenging side?

I enjoy designing challenging bosses, but always aim to keep them fair. I like including mechanics that reward skillful players — for example, in SMWRR X, using the spin jump not only deals more damage than a regular jump, but also makes the fight much easier and faster. Some players didn’t realize this at first, but discovering it changes the whole strategy and creates a satisfying “aha” moment.

There’s also Super Mario World Random Revolution First Blood, which seems to be a Donkey Kong Country style hack with Mario elements. What made you decide to create that game?

It started with the idea of creating a series of hacks called Random Revolution, based on the premise that everything in the video game universe has been absorbed and corrupted by an unknown and mysterious dimension called the Random Dimension.

Originally, my plan was to change the player, enemies, graphics, and mechanics in every single level so players could never predict what was coming next. But I soon realized that making a custom player and set of mechanics for each level would take an eternity — and that content made for a single level could instead be expanded into a full hack.

So, I decided to split the concept into multiple hacks, each with its own theme. In the case of Random Revolution First Blood, the theme is pre-rendered graphics, heavily inspired by the Donkey Kong Country franchise, but I also plan to include content from many other games that fit that visual style.

Why make it DKC/NSMB style anyway?

I’ve always loved pre-rendered graphics. As a kid, games like Donkey Kong Country and Killer Instinct blew my mind because their visual quality was far beyond most other titles on the same platform. Using that style felt like the perfect opportunity to showcase what Dynamic X is capable of.

How do you rip and assemble the graphics for the project? The faux 3D look must be a real challenge to work with…

It depends on the sprite. If it’s from Donkey Kong Country or another game with existing rips, I usually take a sprite sheet from The Spriters Resource and adapt it for Dyxen. Then, I check in the original game to make sure all the poses are correct, since sprite sheets often miss frames or have them arranged incorrectly.

For pre-rendered sprites, I start by finding a 3D model with all its animations. I open it in Blender, adjust the camera and lighting, and render the animations. After that, I rescale the images, reduce the number of colors to fit SNES restrictions and assemble a sprite sheet.

Once the sheet is ready, I load it into Dyxen, set up the animations and hitboxes, and then begin coding the sprite.

What’s the process for choosing levels and mechanics here? We’ve seen Stop & Go Station and a Bramble Blast style level, but will all 3 DKC games see their level themes replicated here?

I usually start with a brainstorming session with my collaborators, then decide which enemies and mechanics fit best. Not every mechanic makes it in — for example, I’m not a fan of the strong wind gimmick from DKC2, so I doubt I’ll include it.

We also add original ideas, like a pyramid level inspired by Super Mario 64 where all enemies can crush you (Whomps, Thwomps, Spindles, Grindels), or an Age of Empires II-inspired stage where monks convert you to their color and enemies of the opposite color attack you.

Are there any other games you’re basing levels and elements on? We recall seeing a Kirby Super Star style area in one older video…

Yes — I want to add elements from Super Mario 64, New Super Mario Bros., Mario Maker, Killer Instinct, Age of Empires II, Doom, Diablo, Mortal Kombat, or really any game that works well with the style.

For example, I’ve already coded Killer Instinct resources like Glacius performing a punch or Cinder doing his signature flip kick. I’ve also coded Doom enemies and even a Doom weapon that the player can control.

I’d also like to add content from modern 3D games if I can get pre-renders of the models. For instance, I want to create a boss based on Smithy from Super Mario RPG, but using the remaster version to match the game’s perspective. Another idea is adding a Ridley boss from the modern Metroid titles, also i would like to implement Super Mario 64 sliders using FMV technique like Winter Gold.

Do you ever worry about running out of ROM space here? Or hitting other resource limits? Every stage seems to have a unique graphics set, tons of unique enemies and objects and its own music.

Right now, I still have plenty of free space since it’s an 8 MB ROM. I’m not entirely sure if everything will fit in the end, but if it doesn’t, I’ll likely split the hack into two parts to include all the content I want — without relying on the MSU-1 chip.

Heck, have there been any instances where an idea just didn’t work out? Like say, where an enemy, boss, item or mechanic wouldn’t work due to SNES limitations?

So far, everything has been doable, but in some cases I’ve had to make sacrifices. For example, I wanted to create a Cyberdemon boss, but the original was simply too large, so I had to scale it down significantly. Another case was the Cacodemon from Doom — after reducing the colors to meet SNES limits, it lost some of its original detail and didn’t look as good as I wanted.

I just need to implement projectiles and one attack i finish this boss, but it is just the first phase of the boss. pic.twitter.com/pqO6Lwi88N

— anonimzwx (Commisions Open) (@anonimzwx) December 5, 2022

When do you plan to release the game?

It’s hard to set a release date for the full hack because it has so many resources and mechanics, and I’m the only coder on the project. My current goal is to release a demo either later this year or in the first half of next year (2026).

Are there any other hacks or projects you’re personally working on?

In general, I avoid participating in too many projects at once, but right now I’m involved in a few:

• Rediscover Super Mario World – I was commissioned to create around eight bosses for this hack.
• Hetaumas – A homebrew game featuring multiple minigames.
• Homebrew Base with Dynamic X – A development base for SNES games that includes the Dynamic X system and various tools.
• Tools for SNES and modern development:
  • Dyxen Ultimate – For creating graphics, animations, and hitboxes for enemies in SMW.
  • HDMA Effects Tool – For creating advanced HDMA visual effects.
  • A level editor for my homebrew base (similar to Lunar Magic) with a generic format usable in modern engines.
  • A modern version of Dyxen for creating animations and generating ready-to-use code and assets.
  • Unity 2D Asset Packs – Tools and resources designed to make 2D game development in Unity easier and more efficient.

These days, I prefer to focus on one project at a time — finishing one before moving on to the next.

Either way, another thing you’ve done a bit recently are commissions, where you’ve coded mechanics and sprites for other people’s projects. What are some of your favourite works so far?

I’ve received a lot of commissions over time — most of them for custom bosses, but also for sprites, gimmicks, and even full systems. Here are some of my favorites:

Complex Systems

• Player Revolution – A system that allows multiple custom players, each with their own power-ups, reskinnable graphics and animations, and any possible size.

Bosses

• Super Fry Guy – For Knetog’s hack, featuring many unique and creative attacks.
• Super Bowser – For Rediscover Super Mario World, similar to vanilla Bowser but with many extra attacks, custom animations, and no Mode 7 — it uses dynamic sprites so all three layers (1, 2, 3) remain available.
• Count Koopula – For Castlemania II, based on Dracula from Castlevania: Symphony of the Night.
• The Evil One – For Arbitrium, inspired by Musya: The Classic Japanese Tale of Horror, with many new attacks.
• Reaper Sigma – From Mega Man X4.
• Gunner Unit* – From Metal Slug 5.
• Super Mario Maker Bowser – With power-ups for Super Mario World Mix.
• Kaizo Super Bowser – For Alvinwawa’s hack, a harder version with Kaizo design.

Gimmicks & Mechanics

• Metroidvania Skill System – Works like Metroid, where acquiring a skill permanently unlocks new abilities.
• Alternative Dimension (SMB2)* – Allows switching to an alternate version of a level, with its own enemies.
• Celeste Sportglass – Lets the player visualize the level layout.
• HP Bars – For enemies and bosses.
• Banzai Bill Rockets – That the player can ride, like in DKC Returns.

Custom Enemies

• SMB2 Logs – Fully customizable, with reverse gravity and different sizes.
• SMB2 Albatross – Can drop any enemy.

Derived Sprites & Improvements

• Bullet Bills, Banzai Bills, and mini Banzai Bills — with variants like homing or controllable platforms.
  Improved Mecha Koopas, bowling balls, explosions, Chuck rocks, and Bowser flames.

Editor’s Note: The ones marked with a * are for a project the writer of this article is developing

Do you think you could become a full time modder this way?

Right now, commissions are my main source of income. While it’s not yet at the point where I can live fully off them, the demand is growing, and I expect it to become more sustainable in the near future. Commissions have given me valuable experience and industry connections, but my long-term goal is to run my own video game studio — or join an existing one — where I can work on larger projects full time.

And uh, do you ever worry Nintendo will try and shut it down? Feels like relying on Nintendo IP for money isn’t always the safest move…

I’m not worried about that. Many other SMW hacks have been released over the years without receiving a cease-and-desist from Nintendo. These projects are non-commercial, made by fans, and shared for free, which historically has made them much less likely to be targeted.

What do you think the future holds for SMW hacking as a whole?

I think SMW hacking has slowed down in terms of innovation. Many popular hacks today are Kaizo hacks made for streamers, and a lot of them follow similar formulas, like a clone of others kaizos or level structure from jump team. However, there are still creators pushing boundaries — projects like Steven Atauchi’s Super Mario World Mix and Stivi’s Mario Stellar Destiny are great examples. Even in my commissions, I’ve seen people request unique bosses and mechanics, which shows there’s still room for creativity. I believe the community could move forward by organizing collaborative projects to develop new tools and resources that make it easier for creators to experiment and innovate.

Finally, what advice would you give someone looking to get into assembly programming, or game development as a whole?

Assembly programming and game development can be incredibly rewarding, but building a strong foundation is essential. One of the biggest mistakes newcomers make is relying too much on copying and pasting code without truly understanding it. My advice is to take the time to learn: read tutorials, experiment on your own, and avoid depending entirely on automated tools or AI. Learn about design patterns and how to structure clean, maintainable code — a solid code base makes future changes and bug fixes much easier.

For game developers, another common mistake is starting with a massive, ambitious project that’s far beyond their resources. Begin with a smaller, manageable game that matches your skills and budget. This approach will give you valuable experience, teach you the complete development process, and prepare you for bigger challenges later on.

Start small, keep learning, and focus on writing code you can be proud of — because every great game, system, or tool begins with a single well-written line of code.

---

Thanks Anon! Yeah, starting small is definitely the way to go where game development is concerned. We’ve all seen stories of people trying to make a super ambitious project as their first ever game, and well, the results are usually a very late release if lucky, or total burnout if not. Like with this guy:

And yeah, definitely don’t rely too much on AI or other automated tooling. They can certainly help if you’re dealing with small code changes or working on boilerplate code you’ve already handled a million times, but they can also breed complacency too. We’ve all seen people who stopped learning because there was an ‘easy’ solution, then got stumped when they faced a challenge their tools couldn’t solve. Or the morons that vibe coded an entire web app without knowing how programming works, then watched their entire business get hacked and destroyed overnight.

Regardless, thanks for the interview! It was amazing to talk to you about your ROM hacks and projects, and we’re excited to see what other stuff you come up with in future too. DynamicX is one of the most exciting innovations in the Super Mario World ROM hacking scene in years, and projects using it have been able to do things no one would ever think possible on the aging hardware.

Still, what do you think? Did you enjoy the interview? What questions do you have about Random Revolution, DynamicX or other projects Anonimzwx has worked with over the years?

Leave your thoughts in the comments below, on social media, or on our Discord server today!

Anonimzwx on Social Media

• Anonimzwx on Twitter
• Anonimzwx’s YouTube Channel
• Anonimzwx on Patreon
• Anonimzwx on Twitch
• SNES Office Discord Server
• SNES Storage (Anonimzwx’s resources)