I know there are 2 similar threads going on right now, but I wanted to talk specifically about a specific kind of system. What if to get spells the only way was to research them through trial and error and hints and clues you could get from adventuring? No buying of spells. No stealing of spell books. No looted spells. Pure and simple: research only. Furthermore, there is no "forum surfing" for spells either. Every single character is unique. The "pieces" one character needs to create a spell effect are different from another. There would be potentially hundreds or thousands of combinations of the "pieces" which could lead to a particular spell effect, but only one that would work for a particular mage. What makes a mage different than any other character is the ability to Read Magic. Through months to years of training, their teachers find a way to teach an apprentice mage to Read Magic. This skill is a different innate ability for different people. Once apprentices can finally Read Magic, they are set free from the master. Therefore, the system wouldn''t be totally random. When magical "clues" are found, they would read differently for different characters. All items that you use Read Magic on are immediately consumed (the writing disappears) as an image is implanted into the mages head (a vision is seen). The images can later be recalled through meditation (this is used as a tool for spell research later, so players don''t have to write stuff down outside the game). Obviously it will be a LOT of work to come up with a system that has sufficient combinations and not totally random, and with good enough hint system so that it''s a challenge and not dumb luck (dumb luck trial and error wouldn''t be fun). Essentially, the images present both a puzzle and a solution, and it''s up to the mage to figure out which is which and how everything fits together. Also, the system needs to be tunable, some spells should be harder than others to research. But, this shouldn''t be a lower percentage or a roll of the die, it should simply be that the puzzle is harder and possibly more hints/clues need to be collected before the spell can be researched correctly. I think this is buildable. The question is, do any of you think it would be fun? Has it been done in a game I''m not familar with (chance is it has)? The goal is turn mages into a thinking class, not an action class. They will adventure, but not for gold or jewels, but for these arcane bits of knowledge. These are what will lead them to true power over time, but I could see it being boring in the beginning. Like mages in a lot of tabletop RPG systems, mages will start out weak (as poor fighters who can''t do much other than fight poorly and Read Magic :-), and it will take a LOT of dedication for a player to get his mage really strong (a 30th level mage with poor spellbook is pretty much worthless). But, overtime I think the rewards of building such a character would be very appealing, but it probably wouldn''t be for all people. In fact, I would design the game such that without high level mage (or two), there is simply stuff that a party can not do. Thoughts? Rhino

Daggerfall had this sort of spell-researching system. A lot of MUD have that also.

This sounds like a nice idea, but every game I''ve played so far that has included "Spell Research" has been the worse for it. If you could implement it in a way that makes the game more fun , it would be quite an accomplishment.

"Let''s see...

Water crystal, L rune, eye of newt, iron ore... nope.
Water crystal, L rune, eye of newt, silver ore... nope.
Water crystal, L rune, eye of newt, copper ore... nope.

(15 minutes later)

Water crystal, E rune, dragon''s fang, aluminum ore... HEY! Alright! I learned.. .. . Minor Ice Bolt I."

Well, since most people don''t play games to read books, that is kinda out. I think that spell research should be done by the player. And here''s what I think it should be like! =D You should get like some parts of a puzzle to form some kind of icon that gives you the knowledge of the spell. (well ok maybe a book would be useful here) But the book details the construction for this in a little bit of a riddle form. Like the book says something cyptic that the player will have to figure out themself. (But dont make most spells of this type, just the super cool ones) It could say something like ''combine the hair of the animal that breys with the blood of the monster that deepest in the caves. Once that is complete, combine this with the stone that lays where the sun watches you.(a place where the sunset at an exact time could look like an eye.)''

-=Lohrno

Asherson''s Call uses a system similar to this. In it though, it''s not exactly research, more like trial and error. You combine herb A with stone B, say the magic words, and wait to see what happens. The downfall to this system was, I started into the game late, and everybody had every spell. How? Even though they tried to randomize it for each person, people still figured out the pattern (those hackers are clever!). In a single player environment, there would be less reason to cheat (ie. no "everybody else does it" mentallity). In theory, this system is amazing. In practice (in AC''s case), it was more of a bother. You didn''t hear people saying, "Cool, I can research all these spells!". Instead, "ahh man, I have to download that spell guessing application". Personally, I tried not to cheat. But, everybody else did it, so it put me at a disadvantage. Eventually I cracked.

I''m all for this type of system though. If you can figure out a way so it''s hard for people to cheat the system, that would be awesome. I''m kind of interested in this idea, so when I have more time I may be back to discuss implementation details.

I think that basically all good ideas boil down to the design/implementation. It may be that you have to just make 100 possibilities for component 1, 100 possibilities for component 2, and 100 posibilities for component 3, and none of them are dependent on eachother. 100x100x100~=1,000,000 So it would take people a long time to do it by guesswork. (I know, that isnt totally trivial, but =D ) That would also mean serious writers work.

-=Lohrno

PS: I don't know, when you have 100 things to write riddles for, and you have to test them out....it could get ugly =(
And if there are only 100 items, and not 300, it needs to make them all unique in the creation of some spell.

Why not instead of making wizards research spells, make them do a dance routine for them using the num pad? =D (duck)


Edited by - Lohrno on March 10, 2002 3:47:52 PM

I think a good way to do this would be to include hints in some old forgotten tomes, or in various books through old libraries. The player would actually have to read and comprehend the books to understand the principles of the spell, and they can use a little clever thinking to put it together. This may seem tedious work, but it would make magic more rare and spectacular, and only the truly dedicated mages would excel.

Thanks for the replies.

Yeah, I played Daggerfall and AC. I *hated* the AC magic system for the very reasons pointed out, yet it is what inspired me since I think it had potential but suffered from poor implementation.

I think Lohrno has got it right. The idea could work, but it depends on if it''s implemented well.

The system I''m planning is less trial and error and more puzzle/riddle. There are enough combinations such that there several million combos, and yet there will be at most a few hundred spells. Even if the game grew to 1000 spells, you are talking a very low chance of hitting one randomly. Plus, the cost of attempting and rarity of the components will prevent trial and error. You won''t want to try until you have a real good idea that it''s the right combo. I figure trial and error isn''t fun, so why encourage it?

What has made me think about this so much is the idea of Spell Research in every game that has had it got me all excited. This is probably because I used to LARP and play in a system where the magic research *ruled*, I could spend hours on end researching and never get tired of it. Unfortunately, I''m not sure this particular excitement can be reproduced in a CRPG since it''s "hands on". What I would consider menial and boring in front of a computer is for some reason fun when I''m "in character" doing the same kind of stuff at a LARP event. The level of immersion needs to be a lot higher than it is today for a system like that to work I think.

Yes, indeed hackers are clever, any system that has patterns (rules) will in time be reverse engineered. But, I think I have a way to make it MUCH harder that it was in AC. AC falls short because the pattern is set at creation time, essentially like you have one random seed that is used forever. Therefore, once you get a success, then two successes, then three, it''s easy to spot the pattern (in fact the more successful you become, the easier it is to guess).

The system I''m coming up with will work around this by re-seeding itself every time there is a successful research, but it has two pretty big problems. First, ideally past failures should''t read as successes (this minimizes the re-seeding pool). Second (and more importantly), the puzzles/hints have to be such once you reseed, they still make sense and still work. I''ve gotten solutions to both of these, but they aren''t really that pretty.

One thing I am struggling with related to the my solution to the above two problems is which patch to take wrt research. I''d like a vote from you folks is what would you find more appealing:

a) when you are doing research, the system supports you knowing exactly what you are trying to create. When researching you are looking for a spell that does a specific thing. For example, you know you are trying to find Fireball.

b) when you are doing research, the system supports you having a vague idea of what you are researching. For example, it''s a fire-based range attack spell. You don''t know if you''ll come up with a Spark, a Magic Missile, a Lightning Bolt, a Fireball, or what.

c) you have no clue what you are researching, you have some hints which lead you to think the following will prepare a spell. You don''t even know if it''s going to be a "storable" spell, or an immediate spell which immediately affects the caster.

There are of course different shades of option b. for example, you may only know it''s a fire-based spell, making it closer to option c.

Option c (and b to a lesser extent) leads to some really interesting gameplay. For example, let''s say you just research the Immediate version of Invisibility, turning you invisible. Or, the Immediate version of Flesh to Stone! I want to make Spell Research fun, but also risky.

But, I think c could be less fun than b. Having no clue what you are going after seems too random. The idea is that at any given time the player if actively adventuring will have several paths of research to follow. If they really need an ice based attack spell, it would be nice if they could eliminate a few paths.

Unfortunately, this kind of forward-knowledge in the system makes it tough to re-randomize. Option c works well with this though.

Rhino

Asheron''s Call used this, as someone else said.

The problem is, sitting around doing nothing is not most people''s idea of fun. If you''ve ever meditated while staring at your spellbook in EverQuest, it''s not exactly a great reason to pay $20 a month to play. You sit there, staring at the screen until you get done or are attacked.

The way I was thinking is to base your ability to cast the spell on how you were trained. For example, in Robert Jordan''s Wheel of Time series, some of the Aes Sedai (spellcasters) have to cast their spells by doing gestures (i.e. throwing motions for a fireball) and some don''t. What if you keyed the system to a person''s training? If you were trained to "throw" your fireballs, then tried to do a shimmy dance, it wouldn''t work. It''d provide some individuality, and not be so involved.

Ahhh yes, I can say I''ve stared at an EQ spellbook screen for a good bit. From March 1999 through February 2000 I logged approximately 150 man days played.

As I mentioned, I think AC had a good idea. As it turned out, it wasn''t fun though, since it was all trial-error (and then the cheats ruined it for even those who could have fun with trial-error).

The big "problem" I see is with the design I''m thinking of is if you want to play an uzi mage and don''t like puzzles or riddles and just want to adventure and blast stuff, then you won''t like mages in this game. Play an archer instead, it''s closer to what you really want.

I''m trying to provide something that breaks the "kill, loot, sell, buy spells, kill, level, loot, sell, buy spells..." cycle in other RPG games. This spell research system is just one small aspect of that.

I''m trying to fold in the stuff that I''ve learned from years of LARPs across a dozen different systems (what I found fun, what I didn''t) and trying to make RPG games less like simulations, and have real role-playing by forcing you to think and act like the character would have to if it were real.

This is hard to do as you all know. Realism for the sake of realism is not fun, it has to be there for a reason. At the same time, it''s hard to get a decent level of immersion without a certain amount of realism. And it''s hard to stay in character when not immersed. Therefore, it''s hard to role-play, and you are back at playing a simulation, not ROLE-PLAYING game.

I''ve played LARPs with folks who absolutely LOVE them and will drive 6 hours and pay $50 to play in a 4-hour event. The same people wouldn''t walk across the street for a free CRPG. I tried to get many of them to play EQ, AC, whatever...the drop out rate was over 80%.

They had tasted the real thing, and nothing else interested them. Maybe i''m a fool for trying to design the game that could offer something anywhere near the equivalent of this.

Rhino