Traditional Rogue Likes?

I’ve been a huge fans of classic rogues ever since middle school and I’ve spent many hours playing them. But, in my three years on this forum, I have never seen a traditional rogue.

Has anyone ever thought about making a traditional tile based rogue like? Other than E-Roguelike by @captainstupids and a brief post about Cataclysm Dark Days Ahead (my favorite). I haven’t seen any other games in this genre.

I can think of some problems that could seriously halt development like

  • How to have a entity occupy multiple tiles (big problem in these games) to properly display size.

  • Most games being custom engine. Some engines like unity and godot are receptive to rogue’s but can be limiting without custom programming.

  • How does weight gain actually affect the game itself and connect to the mechanics?

  • Most rogues are a labor of love and we all lack time and energy in this day and age. (sad but true.)

  • Is custom art going to be used or is ASCII tiles going to be used?

I would like other people’s opinions about the problems and benefits you could see with traditional tile based rogue likes and potential ways to offset this.


the closest I have thought about making would be something like Legend of Grimlock. though I haven’t fleshed it out more than “wouldn’t Legend of Grimlock with fats be nice?”

my personal problem with classic style rogue likes is that I feel you have to follow a “script” (like how there was a bot for rogue that would beat the game) when playing them to win. unless the game is like Nethack where (from my understanding) you can basically do anything. this might just be me not “getting it” when it comes to classic rogue likes

I think that a way to solve the multiple tiles problem would be to ditch tile based combat and go with “radial combat” instead (like in Arc the Lad: Twilight of the Spirits). it might go to far from classic rogue for some but I think you could make it feel more or less the same (it would be harder to gauge distances though)

I feel like if you are gonna use ASCII tiles then you would have to lean into the description way more and I am not sure people want to read the same-ish lines about how fat you are over and over again. though I am a really slow reader so I do have a bias against text in general

also I don’t think it being a labor of love would hinder most people that make games here considering the market for fat fetish games is probably quite small :​P

I understand the parts about ASCII, Labor, and Tiles. But I don’t understand what you mean by “script”. I think you’re saying that many rogues have only one way to go through the game or you fail.

Honestly NetHack has that problem where there is only one way to go through the game and I would like you to explain this concept more.

I think you understand what I mean. another (more accurate I suppose) way to put it would be you can follow a flowchart and win everytime. so after you have figured the game out you don’t need to really think anymore. hopefully I explained it better this time.

but again I just might not get it and misinterpret how the games actually works.
also I haven’t played Nethack so I probably have a wrong picture of how it is as well

It’s all good, rogues can be terribly obtuse with some games being only playable if you have played for 20 years!

I don’t have that time and i’m too young to have lived that long. But I really like your feedback! :slight_smile:

This actually wouldn’t be too hard to do, just gotta have the time and a general understanding of coding(if you don’t have this you could use a VBS system like playmaker to offset).
Granted, I’ve never made a roguelike so take what I say with a grain of salt, but I’ve built out a handful of systems that could easily be adapted to one. I work in unity so my answers will be tailored to that engine.

To tackle this you could have the player object be the image of the character and its collider. Then just have a child object that handles the locations and keep it centered on the image parent. No matter how many squares the character “takes up” the location will always be centered on one square. Then for sword attacks or whatever you could have them originate from another childed object that moves with the edge as the parent object scales. This would also create an interesting effect where the attacks come from the front and leaves your sides vulnerable especially as you grow.

Unity is actually very capable of a lot of things. It could easily handle a roguelike. So much so that they even have an official tutorial on how to make one! 2D Roguelike - Unity Learn
Game dev is literally custom programming so that’s not really a holdup imo.

More weight = more health but a bigger hitbox. This one is really up to the game designer. You could do stuff like slower movement speed, bigger inventory(more room for pockets!), all sorts of stuff. The limit is your imagination.

Not really something that can be worked around. If you want to make a game you’ll make time. Personally I dedicate about two hours every evening to passion projects, but I know not everyone has that luxury.

Unity has an excellent tile system you can use. There’s lots of free art assets out there, you can buy some from the unity store, commission assets from places like fiverr, or reach out to others on this forum that have the skills to help out. You could even go as simple as a grey floor, a colored box for the player then colored circles for enemies. Just depends on your budget/know how.

At the end of the day it’s all very doable, just gotta jump in and learn as you progress.
If game dev is something that interests you then it’s never to late to start learning something new!

1 Like

Thanks, I admit that I don’t know about Unity’s capabilities and coding but I’m glad to hear your points. I want to start to learn coding and I admit that I can be easily discouraged. Thanks for telling me about this!

I’ve played rogue, moria and nethack, and even modified nethack (it was a huge time sink where I worked, so our version got a “landlord” that would chuck everyone out at the end of scheduled lunch break) and they are comparitively simple games in terms of anything that could be called an engine - it’s all basic wait for input, do action, monster turn(s), with the screen updates handled by curses/ncurses.

The complexity comes from all the edge cases of various effects/spells/monsters - this is where the work is, and why they can be so random and fun to play. The result of many many developers working together over many years. Take a look at the source of one: the “core” of the game is tiny (though somewhat complicated by being written to be complied on multiple architecutes and OS’s), the coding for the various effects vast and rambling.

Of course, some of the stuff is only relevant for a multi-user environment, running into another user’s bones/ghost isn’t going to happen on a single user machine, and you don’t need to have the mailman delivering email in-game.

But the problem will always be making the game complex enough to be interesting and challenging. That and balancing it to be just the right level of frustration.

You could always opt to add WG elements into an existing game of course.

1 Like

That is some interesting talk about rogues. I can see your point that the actual mechanics, graphics and gameplay itself is where the complicated part comes in.