Plextona SNES

Hi, guys I just wanted to show you what I've been working on recently. This is a game for the SNES which I have named Plextona. In the game you control a ninja who explores dungeons and is on an adventure. The will progress similar to how Doom on the SNES works. Once you complete a level you move onto another area not being able to return to the previous area.



Now, believe it or not this is actually a Super Mario World rom hack. Keep in mind that while I would like to create homebrew SNES games, I lack the knowledge to create an SNES game from scratch. I consider myself a game designer rather than a programmer, however I will be working towards being more proficient in programming as I continue to work on the project.

I like the idea & challenge of turning a 2D platforming game into an overhead style Zelda like game. Also all of the artwork seen in the gif is original and not lifted from other games. Since it cost money to hire artists for new artwork I am considering taking donations to help fund the project.

So if you would like to see more progress on this game and support this project donations can be made by my palpay e-mail erockbrox(at)yahoo(dot)com. The game will be released as a patch and will be free to play by anyone.

I was excited until I read "Super Mario World rom hack"... Like, hey someone new programming SNES games...oh never mind. Cool though.

I've played around with Lunar Magic, but I agree... looking at the SMW disassembly is like looking at Chinese...or maybe Russian, since I know what the basic ASM symbols mean.

I mean, it's probably way less work than you think, if it's this radically different from SMW already. How much SMW code are you actually using?

I've always found that reverse engineering other people's code and coming up with ways to modify it and have it do things it wasn't originally designed to do was much harder than just designing what I actually wanted from the ground up.

Why not using Zelda? (Genuinely curious, don't know the difference in tools/docs availability for SMW and Zelda)

I'm starting to sound like a broken record at this point, but this is what I mean when I say we need more engines and tools for homebrew dev. In any case, don't sweat it too much. If you keep your code separate from the underlying SMW base you should be able to more easily port your game over to an original base, if you decide to do so. Either way, we could always use more homebrew.

Agreed.


I'm still (sort of) trying to work on it.

Once you get it playable, one of us might help you port it to an original codebase.

This isn't intended as an insult, but the graphics you have look NES. It's fine to prototype your game on a more powerful platform even if you'll be underusing its graphics. But don't be surprised if someone offers to demake it for the actual release losing little or nothing. That's how STREEMERZ from the first Action 53 was made: Mr. Podunkian prototyped it in some environment that compiles to Flash and then thefox ported it to NES.

I quite like the graphics myself. I don't suspect it would take a lot of effort to add one or two more shades. I think they're at a pretty good starting point.

What's hilarious about smwcentral is how much they are afraid of slowdown when none of the stuff they do is CPU intense in the first place.

Can't say I agree with you on that, I think modifying the internals of an existing platformer game engine to make a top-down game is very impressive and certainly counts as programming a SNES game. It's arguably more work than starting from scratch, depending on the resources you have.

It really depends on how much work you're already doing each frame. I was told once that the score counter can lag the game because it's updated every frame even when the vast majority of frames the counter isn't going to change as a result of it.

That might be misleading - it's not the score counter itself that is the sole cause of lag, but the cumulative processing of many small or large things every frame that reduces the amount of available CPU time enough that a few cheap operations are enough to break the camel's back.

In any case I imagine a lot of this stuff can be optimized or just removed entirely.

SMW's specifically converting the score, lives, and coins from binary to decimal, I infer (as ones digits don't matter…and the score lacks a ones digit); this is why zeros in it tend to lag the game less.

I don't know how, specifically, the game calculates the iris-to-black, but I expect it's the major expense that makes the comparatively minor alterable one of B2D go over the "lag-frame" time threshold.

The source seems to indicate that this is mostly a result of the score counter calculation and fadeout transition occurring simultaneously (at least in the vanilla game?). Shaped fades seem to lag a lot of games simply on the basis of having to calculate hundreds of values per frame. This is actually one of the areas I'd like to investigate with homebrew: implementing an HDMA window circle fade effect and trying to see how efficient I can make it.

I've seen several SMW hacks and this has me scratching my head. Even hacks with new mechanics like Brutal Mario still keep the basic gameplay intact.

Does it calculate irises on the fly? I don't know about most games, but EarthBound has them precomputed, so that each frame it can just point HDMA to a different table in ROM.

Okay guys, thanks for your input.

The reason why this is a SMW rom hack is because I simply don't know how to code an SNES game from scratch, nor do I know how to code a level editor program and other tools that I would need. And yet at the same time I'm passionate about making new SNES games because the SNES was my go to console for many years of my life.

As a game designer when I go to the SMWcentral website and see literally hundreds of custom blocks and sprites that I can use at my disposal, I just feel like wow, I don't have to code anything. Instead I can just work on building a game with lots of premade resources.

I've been thinking of making game totally from scratch with gamemakerstudio 2, but I'm not proficient enough to use it to make a full game yet.. I would have to hire a programmer to code things that I don't know how to.

The colors for this game are simple, kind of like NES restrictions. This was done on purpose and I do not want to make it look like a 16 bit game. Why? Because it would cost way too much for a 16 bit game and its harder on me as the developer. By going with just a few colors I can make things a lot faster and it takes a load off my back.

Plus did you know that often times I end up having to be a pixel artist and correct things that the original artist missed or didn't get right. This actually happened a lot with this one tile-set that you see in the gif. With simple graphics I can do this myself and fill in the gaps, but if the graphics were more complex I would have to send the pixel artist tons of corrections and such that they would need to fix because it didn't fit in the game or wasn't the right size and so on. In other words it becomes a pain.

I also like the idea of taking the smw engine and doing something totally different with it.

Wow, I never thought about using lunar magic for a non-Mario game.

I looked up information about the dec to hex converter, and it works by decreasing the 32-bit score by 100,000 and counting how many times it takes before being negative, then it does the same for each digit. Yeah this is a slow way of doing it.

Not sure how to do 32-bit scores, but 16-bit scores can be quickly done by dividing by 100 with the division registers several times, storing the remainders, and using a LUT.

Or you can just calculate the score in decimal mode.

That's quite funny – the same algorithm is used in Fire Emblem: Thracia 776. So Nintendo did this conversion in the same inefficient way in both the first and last game released for the platform.

Does smwcentral have any patch to fix this?

I am all for the idea of taking an existing game and hacking it beyond recognition to become your own game. So many useful low level routines are already there, it can save a ton of time. I'm not doing the same myself because I set my goal to build a NES game from scratch, but doing what you're doing has been in my mind for quite some time. If your goal is just to have a new game, then by all means use whatever you have at disposal to do that.