Showing posts with label RPG. Show all posts
Showing posts with label RPG. Show all posts
Sunday, November 18, 2012
Last week rush
Alright RPG Roguelike prototype has had it's map size increased, A minor change added to map generation so it isn't painfully long for most floors, paladins added with working stances, fighter stances added, and initiative now works. So I guess it's time to add bloat features as the floors are actually boring with little on it. Granted the "floors" are simplistic 4x4 grids with rooms and such but I can still try and improve going around them.
Monday, July 4, 2011
Battle Formulater Scribbles
Suddenly, out of nowhere a horrendous formula. Technically not even all of it but I'm not encapsulating it in another font 60 set of parentheses and multiplying it by "attack stage".
That Formula up there is from as near as I can tell is the Pokemon damage calculation equation. A bit of Googling lead me to a certain web page and I'll take the benefit of the doubt on it. Math oddly enough is not my strong suit so excuse the horrific display of that, I figured it looked neater than the typed in one line my source provides. Though why this formula? Well I'm getting into damage mechanics for Combat Tester and Pokemon formula always comes up when I think about damage.
First a little story. My first RPG was Pokemon, kids in middle school played it and I was interested so after some grade and favor trading with grandparents I got a game boy color and red. There starts me maxing out my hours in the time played. Many years later and a few scraped projects ago I look into how Pokemon worked as well as a few various combat systems I have played only to find some of the most hard to follow and mind boggling decisions in design to my eyes. They just seem like needless obscuration of what your stats actually do. I may have been raised on computer RPGs but I honestly prefer the simplicity of Pen and Paper resolution of damage for the most part.
Now onto the here and now. So after evaluating the damage system of purely being attack successes - defense successes I came to the conclusion that I paint myself into a design corner with that. Instead using a commonly done solution of dice+modifier-damage resistance to get that job done. Specific details as to how I'm getting the numbers and such would be a little long for a post topped with a monstrous formula so I'll wait until later to go into those details. While not as unique as I like it'll work for now, I'll work on it more to see if I can go anywhere with it. Though it does open a few combat options I like, namely reducing or increasing your attack compared to your final damage.
In other news I'm going to be shooting for getting/starting a blog up every day around 5 o'clock central. With the current unemployment going on I could use something besides my Tabletop game schedule to set something up for me.
Edit: Happy forth to those of you in the United States. It's too hot and I'm without transport to care or get to the festivities.
Tuesday, June 28, 2011
Difficulty
One Shaw made a comment on my last blog about difficulty. Saying "It's important to have the AI scale with difficulty." which I agree with. Though on thinking on this I reached my old problem of "what is difficulty?".
And to illustrate as I have not added images to this at all, have a horrible MSPaint quality image that I should have been ashamed to make.
Once again I apologize, but it kind of gets the point across.
So the first thing to start off with is what a lot of games like to do with difficulty being big numbers. This comes in all manner of flavors both hidden and not so hidden. One is hit points or the number of times something has to be hit or damaged to take it down, this is the oldest of the old for difficulty. If you played a mario game that had all bosses take 4 more hits it would be more "difficulty". This happens a lot in shooters where difficulty just means increasing enemy damage and enemy health. For real time games there is always just how far the AI can "see" you. As it stands I do not believe there is an AI in a game right now that legitimately uses 'vision'. It just calls a method for the location of any player, checks if there are any solid objects in the way of 'seeing' them then goes about all the other functions. The higher the difficulty the more easily the AI can spot you, or even see through walls so it can plan a route around them to shoot you. That once again is more "difficult". RPGs have a notorious difficulty increase purely by numbers. Every number used by the enemies is increased which means ultimately the player just has to kill a lot more monsters to level up to continue than if it was on an "easier" difficulty.
For me this isn't the difficulty I am after. It's what I like to call "lotto" difficulty. I mostly use the term as a trap in Pen and Paper design where fights can only be won if the player rolls high consistently while the monster rolls low. Nothing the player does actually in the fight matters only what the Random Number Gods decides to give them. Any input on the fight the player had was decided before hand in how much they beat up lower level enemies or the quality of equipment they are using. This effectively just turns the bread and butter of the game into a math equation. For shooters and RTSs some of this is mitigated as there is a lot more things to do than click a button but at it's core it's not what people want. People want some challenge which comes down to their actions influencing the outcome, not some long formula of success that has no bearing on what they did during the time of conflict.
It isn't impossible but it is hard as a designer to address the problem. For shooters there is always changing up the map making the play have to choose the approach enemy, or by giving the foes tactics of their own while rudimentary can it more challenging for the player. RPGs have a lot harder time with their system not being twitch based but 'tactics' as some basic multi turn stratifies or nullifying players can prove quite rewarding if you're looking for a game that doesn't require a lot of grinding. Still it's a lot of work for what ultimately isn't what players really want, something akin person behind the scenes with forethought and a bit of an imagination. I fear it'll be a lot longer than we expect before we get such an AI and for now Pen and Paper games or ones with heavy Game Master involvement will be the closest we get to it.
And to illustrate as I have not added images to this at all, have a horrible MSPaint quality image that I should have been ashamed to make.
Once again I apologize, but it kind of gets the point across.
So the first thing to start off with is what a lot of games like to do with difficulty being big numbers. This comes in all manner of flavors both hidden and not so hidden. One is hit points or the number of times something has to be hit or damaged to take it down, this is the oldest of the old for difficulty. If you played a mario game that had all bosses take 4 more hits it would be more "difficulty". This happens a lot in shooters where difficulty just means increasing enemy damage and enemy health. For real time games there is always just how far the AI can "see" you. As it stands I do not believe there is an AI in a game right now that legitimately uses 'vision'. It just calls a method for the location of any player, checks if there are any solid objects in the way of 'seeing' them then goes about all the other functions. The higher the difficulty the more easily the AI can spot you, or even see through walls so it can plan a route around them to shoot you. That once again is more "difficult". RPGs have a notorious difficulty increase purely by numbers. Every number used by the enemies is increased which means ultimately the player just has to kill a lot more monsters to level up to continue than if it was on an "easier" difficulty.
For me this isn't the difficulty I am after. It's what I like to call "lotto" difficulty. I mostly use the term as a trap in Pen and Paper design where fights can only be won if the player rolls high consistently while the monster rolls low. Nothing the player does actually in the fight matters only what the Random Number Gods decides to give them. Any input on the fight the player had was decided before hand in how much they beat up lower level enemies or the quality of equipment they are using. This effectively just turns the bread and butter of the game into a math equation. For shooters and RTSs some of this is mitigated as there is a lot more things to do than click a button but at it's core it's not what people want. People want some challenge which comes down to their actions influencing the outcome, not some long formula of success that has no bearing on what they did during the time of conflict.
It isn't impossible but it is hard as a designer to address the problem. For shooters there is always changing up the map making the play have to choose the approach enemy, or by giving the foes tactics of their own while rudimentary can it more challenging for the player. RPGs have a lot harder time with their system not being twitch based but 'tactics' as some basic multi turn stratifies or nullifying players can prove quite rewarding if you're looking for a game that doesn't require a lot of grinding. Still it's a lot of work for what ultimately isn't what players really want, something akin person behind the scenes with forethought and a bit of an imagination. I fear it'll be a lot longer than we expect before we get such an AI and for now Pen and Paper games or ones with heavy Game Master involvement will be the closest we get to it.
Ai and choices
Well back at home after house sitting. Got a lot of programing done over there, the basics of the combat tester engine are pretty much complete. You can hit a guy, aim better, and guard. The very very basics are here and it works out which leads to some problems if I want to get to my goal of having something
interesting in game play.
So I have the basic mechanics down, you choose dice to assign to differnt aspect in the combat round and the system calculates everything and then goes to the next round until someone falls down. This is working but not quite what I want it yet. After some brain storming I decided to try some of the basic ideas that are common in some pen and paper games.
I'm unsure if D&D started the whole idea but I know it best from D&D 3.5. A little feat called power attack. Effectivly power attack reduces our chance to hit for some extra minor damage. I'm unsure if from there other X attacks spawned but in mutant and masterminds 2nd edition a few more were made allows if your to sacrifice damage pottental for the chance to hit.
Why this matters for my system? Because as it stands it doesn't work. The main problem is that I'm flatly using how much attack beats defense as damage. So a -2 on one side and a +2 on the toher just reduces the average damage overall, and how many times it actually does damage. Using something more drastic such as *2 on one side and /2 on the other leading to the more accurate but less damageing attack somehow able to deal more damage than normal attacks and other such crazy things. So the most basic ideas in the hit them till they fall down to try and increase the complexity of this system sort of fails so I will have to think harder on this.
Now for the other part of combat, the enemies side the AI. AI is decently hard. For me I don't have to worry about a lot of things, position, movement actions, and all the sorts for things that make games annoying to program. For me I have it easy as I just have to choose what ability it can do so I have a few routes to go. I can go through a gauntlet of variable comparison and spit out an action. I could do what no RPG has ever done and do breadth or depth first searches taking into account what the players and random number generator into acount and make one of the hardest RPG games out there by virtue of that. Though I could also do what I usualy end up doing and go with my "randumb" model of RPG enemies for the mooks at least, where it just randomly picks from the list of abilities and goes with it.Not the most imaginative but I have little work with applied AI and would rather have it work or use some scripts to act though it is something I could work on.
Until later I'll be working a bit more on trying to make this system interesting as is.
interesting in game play.
So I have the basic mechanics down, you choose dice to assign to differnt aspect in the combat round and the system calculates everything and then goes to the next round until someone falls down. This is working but not quite what I want it yet. After some brain storming I decided to try some of the basic ideas that are common in some pen and paper games.
I'm unsure if D&D started the whole idea but I know it best from D&D 3.5. A little feat called power attack. Effectivly power attack reduces our chance to hit for some extra minor damage. I'm unsure if from there other X attacks spawned but in mutant and masterminds 2nd edition a few more were made allows if your to sacrifice damage pottental for the chance to hit.
Why this matters for my system? Because as it stands it doesn't work. The main problem is that I'm flatly using how much attack beats defense as damage. So a -2 on one side and a +2 on the toher just reduces the average damage overall, and how many times it actually does damage. Using something more drastic such as *2 on one side and /2 on the other leading to the more accurate but less damageing attack somehow able to deal more damage than normal attacks and other such crazy things. So the most basic ideas in the hit them till they fall down to try and increase the complexity of this system sort of fails so I will have to think harder on this.
Now for the other part of combat, the enemies side the AI. AI is decently hard. For me I don't have to worry about a lot of things, position, movement actions, and all the sorts for things that make games annoying to program. For me I have it easy as I just have to choose what ability it can do so I have a few routes to go. I can go through a gauntlet of variable comparison and spit out an action. I could do what no RPG has ever done and do breadth or depth first searches taking into account what the players and random number generator into acount and make one of the hardest RPG games out there by virtue of that. Though I could also do what I usualy end up doing and go with my "randumb" model of RPG enemies for the mooks at least, where it just randomly picks from the list of abilities and goes with it.Not the most imaginative but I have little work with applied AI and would rather have it work or use some scripts to act though it is something I could work on.
Until later I'll be working a bit more on trying to make this system interesting as is.
Saturday, June 25, 2011
Weapon of the gods and combat testing
Haven't been posting as much due to house sitting. Something about having access to cable and more food than I could buy in a month has made me as lazy as the cats I'm here to watch. Either way on with the blog.
So I've been playing game designer again and trying to come up with a more complicated than usual combat system. So as it usual goes with my designs I wanted to see what else was out there and did some digging through a few rule books and dug up a gem called Weapon of the Gods. A self proclaimed wuxia role-playing game with special effect martial arts which include a technique that will break apart your armor and cause you to lose limbs if you're unlucky. For how campy most martial arts films are this is taking itself decently serious, lots of 'historical' fluff in there and a lot of techniques. The system is a dice pool that uses matching, no successes or cumlative adding though. It takes the interesting approach of using d10s and having the number you pick be the 1s (a zero in 10s case) and the amount of dice matched being the 10s. So rolling 6,3,3,0 lets you have the choice of 10 (set of one 10 die), 16 (set of one 6 die), or 23 (set of two 3 dice). So the craziness is a little toned down from things like frameworks matching dice pool system for the number it can generate. Beyond this intersting take on dice pools is what is called the river.
The river is a very crunchy and somewhat narrative idea where you can save dice for latter. In our pool we rolled before we had a set of pair of 3s. After we roll we go through the "phases of the river" first phase is to wash out what we have in the river, also known ans plainly discarding the dice we have saved there. Each scene starts with no dice in the river so you can't keep your 9s sticking around to save you when needed. To get dice into the river after you "wash" them you can float in anything from a set of 2 or more dice. You can just put one in or both in. With the pool above we could float in the two 3s as insurance for later, though this would have to be a legitimate roll not something the game defines as a 'style' roll which will come up in a minute. After this you enter the "flowing" phase where you can move dice out of the river and into your dice pool so that you may use them to generate the dice pools results. So for example later when we roll nothing but 1 sets we could flow in the two 3s and have at least 23, or we could be waiting for 3s to be rolled and roll a much higher possible number.
Now I mentioned style above, style is generally skill checks for mundane things, or pointless things that you will succeed at regardless of your skill at it. Getting the attention of a friend across a room with a few people, climbing up the ladder, other silly things you want to roll dice just to see how awesome you are. So for giggles you order noodles at restaurant and stay you want a style roll for it, GM rolls his eyes and asks for the attribute and possible skill in question and you pull out something around 30s just to say you made the cook weep with your love of the noodles. Silly idea but style checks apply to something else, damage. Damage in this game is determined by how much you pass the targets defense. For every 5 over you roll you get another die in the damage pool. From there it's standard pool reading though you can flow in from your river some 9s to put a serious hurt on.
Now there are other interesting things that I could talk about concerning such wonderful euphemisms such as yin-yang warrior but for now onto my little project.
The actually straight up manipulation of the dice pool got me thinking a bit for at least the basis of what I'm thinking of playing with. Every rpg system I can think of besides Baten kaitos(which used a card based dice pool like system) just has you select an action and just watch the results. So after reading through the weapon of the gods system I wondering if assigning what you roll could add an interesting element to
this. I don't know how valid it is but that's why I want to try. To follow my computer science roots thinking I'm following the KISS rule (keep it simple stupid) and just going to have you able to assign dice to your initiative, action, and defense pool. From there your chosen actions play out as normal, so far I'm seriously lacking in them but I have to see if the current system pans out.
And as a bonus my first image, it didn't feel right at the start so here it is. I'm programing in python cause I haven't been bothered enough to get oracle or C family compilers on here.
Got most of the tools finalized and am starting to play with actual combat mechanics. Learned more things in python recently, list comprehension as I'm showing off a bit can't believe I wasn't taught that in college. All console based, my philosophy is if I do it right I can just use another shell to output it correctly in a GUI anyways.
So I've been playing game designer again and trying to come up with a more complicated than usual combat system. So as it usual goes with my designs I wanted to see what else was out there and did some digging through a few rule books and dug up a gem called Weapon of the Gods. A self proclaimed wuxia role-playing game with special effect martial arts which include a technique that will break apart your armor and cause you to lose limbs if you're unlucky. For how campy most martial arts films are this is taking itself decently serious, lots of 'historical' fluff in there and a lot of techniques. The system is a dice pool that uses matching, no successes or cumlative adding though. It takes the interesting approach of using d10s and having the number you pick be the 1s (a zero in 10s case) and the amount of dice matched being the 10s. So rolling 6,3,3,0 lets you have the choice of 10 (set of one 10 die), 16 (set of one 6 die), or 23 (set of two 3 dice). So the craziness is a little toned down from things like frameworks matching dice pool system for the number it can generate. Beyond this intersting take on dice pools is what is called the river.
The river is a very crunchy and somewhat narrative idea where you can save dice for latter. In our pool we rolled before we had a set of pair of 3s. After we roll we go through the "phases of the river" first phase is to wash out what we have in the river, also known ans plainly discarding the dice we have saved there. Each scene starts with no dice in the river so you can't keep your 9s sticking around to save you when needed. To get dice into the river after you "wash" them you can float in anything from a set of 2 or more dice. You can just put one in or both in. With the pool above we could float in the two 3s as insurance for later, though this would have to be a legitimate roll not something the game defines as a 'style' roll which will come up in a minute. After this you enter the "flowing" phase where you can move dice out of the river and into your dice pool so that you may use them to generate the dice pools results. So for example later when we roll nothing but 1 sets we could flow in the two 3s and have at least 23, or we could be waiting for 3s to be rolled and roll a much higher possible number.
Now I mentioned style above, style is generally skill checks for mundane things, or pointless things that you will succeed at regardless of your skill at it. Getting the attention of a friend across a room with a few people, climbing up the ladder, other silly things you want to roll dice just to see how awesome you are. So for giggles you order noodles at restaurant and stay you want a style roll for it, GM rolls his eyes and asks for the attribute and possible skill in question and you pull out something around 30s just to say you made the cook weep with your love of the noodles. Silly idea but style checks apply to something else, damage. Damage in this game is determined by how much you pass the targets defense. For every 5 over you roll you get another die in the damage pool. From there it's standard pool reading though you can flow in from your river some 9s to put a serious hurt on.
Now there are other interesting things that I could talk about concerning such wonderful euphemisms such as yin-yang warrior but for now onto my little project.
The actually straight up manipulation of the dice pool got me thinking a bit for at least the basis of what I'm thinking of playing with. Every rpg system I can think of besides Baten kaitos(which used a card based dice pool like system) just has you select an action and just watch the results. So after reading through the weapon of the gods system I wondering if assigning what you roll could add an interesting element to
this. I don't know how valid it is but that's why I want to try. To follow my computer science roots thinking I'm following the KISS rule (keep it simple stupid) and just going to have you able to assign dice to your initiative, action, and defense pool. From there your chosen actions play out as normal, so far I'm seriously lacking in them but I have to see if the current system pans out.
And as a bonus my first image, it didn't feel right at the start so here it is. I'm programing in python cause I haven't been bothered enough to get oracle or C family compilers on here.
Got most of the tools finalized and am starting to play with actual combat mechanics. Learned more things in python recently, list comprehension as I'm showing off a bit can't believe I wasn't taught that in college. All console based, my philosophy is if I do it right I can just use another shell to output it correctly in a GUI anyways.
Subscribe to:
Posts (Atom)