ID:89585
 
...of how not to make a rogue-like.


I saw the banner rotating around on the BYOND developer forums. PMUR. It sounded familiar, but off the top of my head I didn't know what it was, so I thought I'd give it a click. "A roguelike developed in under 20kb code-size" says the hub page. I don't remember ever trying this game before - although I probably did and just forgot. But its listed as a favorite through Dungeon Crawlers, so I thought I'd give it a try.

Okay, I did actually read (most of) the instructions that popup when you first start the game. So in all fairness, I know this game isn't complete. (Its still in alpha, and at the rate its going it probably always will be.) You only get to choose the one class, so really you don't get to choose a class, and the random dungeon generating is less than ideal.

New game started. Its pretty much what I'd expect from a rogue-like. I have my character, and there are rooms full of monsters and loot scattered around each level. So I started off collecting loot and slaying monsters. Collecting loot consists of just stepping on it, and slaying monsters consists of bumping into them. If you want, you can be clever in maneuvering so that you can have the first strike again the monsters, and so that you'll only ever be within the reach of one monster at a time. These monsters also aren't very smart and they won't walk around walls or go around each other, so its very easy to get them all bunched up.

Soon I'm on my third dungeon level and my third character level. I've slain a good number of monsters which all seem to be very much alike (walk my general direction, bump into me, die shortly thereafter). At this point, though, some aspects of the game that seemed innocent on level 1 are starting to get obnoxious.

First of all, the dungeons consist of writhing, twisted tunnels. They simply are not fun to walk through, and since there are never any monsters or items in them (although there are often potions at the end), they only serve to make the game more tedious.

To make matters worse, your movement is restricted so that you can't walk very fast. Now keep in mind, this is a single player turn based game. Why in the world a turn based game would need to restrict the player's movement is beyond me. But because of this, making your way through the very long, twisted tunnels that often end up returning to their place of origin is even more tedious.


At this point in the game, I find myself not being so careful around the monsters anymore. Perhaps I'm hoping to die so this game will end.

Having finally reached the 4th dungeon level after roughly 10 minutes of tediously crawling through the winding passages of misery on the 3rd level, it occurs to me that he must have programmed this game to generate the exit stairway only after you've explored every room on the level. Then it must automatically appear in the very last room that you explore. Because that's how it happens every time.

It should be pointed out that there are only two reasons why I've even progressed this far. One is the diminishing hope that there may yet be something interesting to discover further in the dungeon. Two is because I'm valiantly striving to find a better screen shot for the start of this post.

Conclusion

I don't know how long this game continues, but I'm not eager to find out. I've had enough gaming agony for today. Even if I could save my progress, then hack my savefile and turn myself into a demigod before coming back to finish this game, I wouldn't do it. You couldn't make me.

Delgertome, if you're still around, I'm sorry. I suppose you had good intentions. But this game is just awful. I sincerely encourage everyone not to play it unless they're a fan of masochism. The scary thing is that this was considered a favorite in the Dungeon Crawlers guild. Now I'd be afraid to play any of their other games. Maybe that's why nobody has posted there for so long.
Maybe I should demote it from favorites. The list of favorites in the guild is pitifully small but it does make sense to pull it out. I agree completely on the tunnels being maddening.

Much of the reason behind nobody posting there though is I just haven't found adequate time to work on a Roguelike game of my own.
I liked it. [:
How much time did you spend playing it? :P
So...not a single unique thing about the game? No interesting approaches or UI or anything? Seems like something could be found. :) Generating the door in the last room is kinda neat.

ts
I dig this new CSS.
Foomer wrote:
How much time did you spend playing it? :P

PMUR was my first encounter with rogue-likes, I stopped playing after I realized everything was just being generated.

I thought I was simple, to the point, and obviously very unfinished.
Boring map generation is one of the reasons why I've always been fond of games where all the content is user created. Content created by actual people will always be more interesting than randomly generated, because people are unpredictable and they're capable of creating much more interesting things.

But this game was the epitome of boring map generation. Maps consist of long winding tunnels and square rooms. That's it. And nothing ever happens in the long winding tunnels.

Ts - If there was anything interesting about this game, I didn't encounter it. The 'door in the last room' thing is just a theory.
Rogue itself has a very simple map generator, but I think that game is better about surprising you with discoveries and forcing you to make the most of your strategic assets, which overshadows the simplicity of the maps. I also like Nethack's approach which blends the two. The real problem in PMUR is that the twisty passages make up a great deal of the map and they're too time-consuming, which turns most of the game into navigating the map instead of getting into the atmosphere.

I do find good user-created content enjoyable, which is one of the things I liked about Victol's/Ryuo's Dark Dungeon, although that was much more interesting for the visceral experience because it made liberal use of sound effects and effects like s_damage. Later in the game monsters were terribly unbalanced.

I've never been an ultra-firm believer in the supremacy of written vs. generated content though, because replayability and the element of surprise matter a lot. Where written content (in terms of maps) excels, it's often something that a good generator could have made. Written content though is much more difficult to recycle even in a random way. To be really excellent, such content has to be extremely well made and a great deal of time has to be spent crafting a rich environment that does not require linear play. Interplay's 1990 game Lord of the Rings not only had great artwork, but the best map and puzzle design I have ever seen in my life, making a truly huge game world to explore--one where you pretty much always had choices as to how you would proceed. But that level of quality is exceptionally rare, and for richness in a game world I think at least some level of random generation is usually a very good idea.
I've got a thing for Dungeon Generation as well. I made a pretty significant effort a few years back. I tried to do some pretty heavy stuff before I was ready. Anyway, this is my current "Dungeon Crawler" dream/vision or what have you.

- Every tile and experience in dungeon is unique. I want gross detail. GROSS! Every floor tile will be unique. I know this is a "no way" but it's not that crazy. It must appear that way for all intensive purposes.

- Automated generation is fine as long as we are randomizing a few halls or the direction of a table in a room. The world has to be submersive and screaming atmosphere. There are a few clever tricks we can do (have done in my project) using swapmaps.

- One life (period, forever, unless you buy a new BYOND key) Take this dungeon seriously or get out. See next "-" before huffing.

- Design one level at a time as a semi-public dev effort. Those of us who have the passion for such detail can contribute here and there. Jam-pack it full of rich content and story. If a player succeeds in fullfilling the levels requirements (kill boss or whatever) and exits the dungeon alive, they get a token for module two when it comes out. Module two may never come out but if it does, each one would be a unique theme for the adventurer.

- If you die in module one, your screwed until module two comes out. We would of course warn the crud out of people about this unique experience.

- When module two comes out, everyone who died will get a second shot at module one.

- A unique fighter battle system. Utilizing a turn based battle system which enables you to select regions of body to attack and from a large number of combinations which are put up against the monster combinations.

- Monsters would have many animations and many approaches.

- Each battle should be "epic" and an exciting experience.

- Monsters will be unique. One large mama spider on level and she has a name and she has a purpose.

- Many clever puzzles and many obstacle challenges.

- Module two might introduce a mage for which I have a very very clever casting system. Actually, this system should be utilized in some run of the mill RPG so it gets more use but whatever.

So that's about it. Thoughts?
Tsfreaks wrote:
So that's about it. Thoughts?

Sounds like a lot of work. O_O
C_Dawg_S wrote:
Tsfreaks wrote:
So that's about it. Thoughts?

Sounds like a lot of work. O_O

Well, yes and no. It is a lot of work where players may only get a relatively small amount of play value. However, your goal is to "awe" the player and provide them with an unforgettable experience.

To reach an awe enspiring level, it would require a fair number of people contributing. I think a fun goal would be to impress one another with our ridicouly detailed contributions. As an example, a dev went and added a dynamic torch system so every torch sadded in the map editor has a totally unique flame, angle, tinge, dirt, shadow and etc.

It's really an eye candy experiment in BYOND. It could be used as a flagship for potential.

I dunno... it's just a day dream mostly. I'm sure we all dream about the ultimate dungeon crawler. ;)

ts
I remember when he made this, it was a fun project of his. :)
A game that you only get to play once really has zero potential. I mean sure, most of Lexy's games were that way, but not on purpose. Rogue is characterized by a very high degree of difficulty, which is what makes it interesting because it's such a challenge. If you made it a play-once-or-never-again experience per module, anyone who got killed early on would be completely shafted. There's no learning curve involved. Who would go on to play module two after getting their butt handed to them ten minutes into module one and never getting another crack at it?

FAIL.

That said, some kind of semi-public dev effort put into a Roguelike could be interesting, but perhaps more as content with which to seed a procedural generator, or interesting characters to throw into players' paths. I think you have some good ideas tossed around there, but this play-once concept is a dog.
Lummox JR wrote:
A game that you only get to play once really has zero potential. I mean sure, most of Lexy's games were that way, but not on purpose. Rogue is characterized by a very high degree of difficulty, which is what makes it interesting because it's such a challenge. If you made it a play-once-or-never-again experience per module, anyone who got killed early on would be completely shafted. There's no learning curve involved. Who would go on to play module two after getting their butt handed to them ten minutes into module one and never getting another crack at it?

FAIL.

That said, some kind of semi-public dev effort put into a Roguelike could be interesting, but perhaps more as content with which to seed a procedural generator, or interesting characters to throw into players' paths. I think you have some good ideas tossed around there, but this play-once concept is a dog.

Sure, at face value, it can be a "dog". Think about it like this though. To die, meant you were really stupid. I certainly wouldn't design a game where a random monster walks up and kills you because you wandered a little wrong. I would "try" my best to get the player through the module.

In my vision of a rogue-like, its not about crazy hard luck. It's about visual experiences and wow factor. I want to wow a person with detail.

Another way to look at it... people will talk about a one chance rogue like game. Especially one that was impressive looking.

It can be done right... and it can also be done wrong and I recognize that.

As for going on to play module two, they wouldn't. :) They would have to make it through module one first. Perhaps they get to play Module 1 as much as they want once Module 2 comes out but they only get one shot at the latest. :)

and as for all my good ideas... upon review, I didn't really have any aside from...

1. Community Effort
2. Rich and detailed in the extreme
3. and perma death

So, a community effort is "neat" in theory so now its about how. Do we need to gauge interest or do we just start.

If we don't do perma death (which I agree with unless we could do it just right), then I'd like to do something where death earns you negative points on the leader board. I want death to mean something damnit. :)

ts


Just to be realistic about it, people will respond to your "one chance" game one of two ways. Either they will play it, not taking it seriously, die, be unable to play again, and decide that your game sucked, OR they will go through the tedious process of creating new BYOND keys or whatever else is necessary in order to start a new character, circumventing the whole one chance thing. For those people, you're only creating an extra hassle involved in playing your game.

I'm all in favor of huge amounts of detail, but remember that in the end people are there to have fun, and the more hours of fun you provide them, the better. Otherwise your game will be, at best, a brief novelty.
Seriously Ts, the play-once idea is indefensible. Recognize that not every idea coming out of brainstorming is gonna be a winner and be willing to ditch the losers. There's no way to do that right. Tweaking the difficulty as you described would only make encounters too easy to be meaningful, and you can still end up with someone getting locked out of the game just by a freak run of bad luck from the PRNG. I realize the point behind it is to try to keep the element of surprise at a maximum using only written content, but given that it's an open dev effort that's an impossible goal anyway. Besides, if the player wants to try again in spite of being aware of what lies ahead, why stop them? Knowing what lies ahead in a game like Super Mario Bros. doesn't make it less fun. We all have stupid ideas sometimes; it's time to bury this one.

There are ways to give death more meaning however, like in Rogue how a character only gets one life. Forcing people to invest more effort into keeping their characters alive will have numerous rewards. That's why play-through games like Super Mario Bros., Sonic the Hedgehog, Double Dragon, Golden Axe, etc. all were successful in their time--and games of that mold that allowed continues sucked down a lot more quarters than their counterparts.
Lummox JR wrote:
Seriously Ts, the play-once idea is indefensible.

This comment made me laugh off and on throughout the day as my thoughts drifted to and from the idea.

Fine.... I'll tuck the idea away in my back pocket and save it for later.

While thinking about this some more today I came up with two very specific features that I would like to contribute to a project.
- Map Overview
- Casting System
- Traps

The first two are pretty unique and would add good flavor.

Aside from those features, I would love nothing more than to add gross atmospheric detail and sounds. I might even be capable of a score.

What features of a Dungeon Crawler would you guys be primarily interested in doing? I'm not suggesting that you are volunteering for this project by responding either. I think this is a pretty neat question actually.

I may try and post some info about my previous efforts on a Dungeon Crawler and get the game online for folks to see. It's a pile of dung for the most part. It was too much with too little at the time. It does sport the overview feature I'm talking about. I'd actually be willing to use it as the starter project but I'm afraid of the lashings I would get when people see the code. :)

On the project management side of things, here are some thoughts I had. I have little experience with managing any sort of project over the internet so take these thoughts lightly. This might just be drivel...

If we were to do a shared public project. It would be managed and protected in some place by several trusted (enough) file managers.

Permissions to edit will be granted to folks who meet criteria (x). Perhaps the project has no criteria.

- Any file with Core_ appended must be "checked" out by one person at a time. By checked out, I mean that you could edit that file but you won't be able to submit it.

- People can work freely provided they are working out of their own file (tsfreaks.dm)or if they are adding a new .dm file.

- Permissions must be asked for any edits of existing code (tsfreaks.dm or Core_)If you need to make changes to core, you have to wait your turn.

- When you plan to add a feature, you post a pretty detailed description of the feature so others can discuss, adjust, and collaborate appropriately.

A voting system could be used for features and edits. Example, I want to add a new boss. I would post the details and description as a vote and ask? "Can I add this boss to the game?". 51% of the "team" will need to vote yes before your feature will be allowed into the project. Crappy descriptions will be "No'd" and each member should comment for any "No" and contributors will be encouraged to provide feedback for any Yes's as well.

We would expect most feature adds to be relatively small and complete for each check in. The build can never be broken and the file managers make sure by building the project before replacing. As part of the check-in system, perhaps the editor posts a new poll, "Can I check in feature: X?" and the post includes a link to a hosted session with the feature.

ts
Tsfreaks wrote:
In my vision of a rogue-like, its not about crazy hard luck. It's about visual experiences and wow factor. I want to wow a person with detail.

Have you ever played a rogue-like?

#######+##
#........#
#......f.#
#........#
#........#
#..@.....#
#........#
###+######
Tsfreaks wrote:
Lummox JR wrote:
Seriously Ts, the play-once idea is indefensible.

This comment made me laugh off and on throughout the day as my thoughts drifted to and from the idea.

Fine.... I'll tuck the idea away in my back pocket and save it for later.

No really, throw it away. It's a terminally dumb idea; there's no saving it. No game will ever succeed at being fun that can only be played once. Heck, even taken as a rich story experience, people like to re-experience stories they find appealing. It's okay to come up with a bad idea, but not to use it.
Page: 1 2