Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Felblood]

Hm. Does turning Urist's body into a could of butterflies kill him, or allow him to be in many places at once, and how good of a friend do you want to test it on?

How do we deal with prepositions? That's going to make word order important.

"Gently Attach Sword To Target Arm" could be used to make cyborg killing machines out of wounded vererans, but what should "Sword Arm Target Attach To" do, besides causing the laws of physics to explode?

[alfie275]

The targetting system would have to be able to recieve conditions on the conditions, for example Burn All Enemy Flying Arrows

[Felblood]

Narrowing targeting and effect parameters with additional adjectives and doing likewise with effects and adverbs is simply, though most modifiers should get expensive fast if you stack them up. Making them cost multipliers would account for this.

--but there comes a time when you want to specify a second target, in a slightly different way. Glue Goblin Foot To Floor, is different from Glue Goblin Foot Floor. You're liable to glue the whole floor, the whole goblin and your own feet, without some grammar to codify the relationships between the words.

However, grammar means there will be some combinations that simply don't work. Glue Goblin To Floor Foot? Green Goblin Hair? Putrified Wiggling Vermin Heads? And Self Lack Spoon? You broke the universe, tearing a hole that causes the source of the paradox to be sucked out. Have fun with the formless horrors, out there in the twisting nether.

We can account for some of these. Someone suggested dropping subtargets that make no sense. However, sometimes you're going to end up in a situation where none of the targets is viable. Kill Wall, for instance. Do non-targeted spells simply fail to cast, waste energy but do nothing, or strike the first viable target(usually the caster)?

A lot of people hate mana, either because it's quantifiable magic("Twenty-eight thousand thaums!", "It's over Nine thousand!" "She is a class five psychic" etc.) which sucks some of the mystery out of mystics, or because they saw MP in Final Fantasy, which clearly makes it a product of Game Design Satan, as a part of his evil conspiracy to make all games identical and boring (OMG! They're after my brain! Where's the tinfoil?!).

I don't want to see the idea go away, because of all the cool ways it could interact with spheres, but I'd like to see what can be done to keep it palatable, to those who like their magic arcane and ookey.

[Granite26]

cause that won't be buggy at all...

[Eagleon]

Hehe, I'm one of those people that hate MP, simply because it turns magic into bullets in most games.

I'd interpret "Sword arm target attach to" as start attaching things to the target's sword arm. All sorts of things. Anything the magic can lift and move to the arm, it tries to attach. Horrifying o.o

But that's moving the words around a lot into weird grammar. I guess there needs to be some control over that. Some order needs to come into play of what's possible and what's not. I don't really know how you'd handle it, other than having an artificial fixed ordered interface, with slots for <adverb> <verb> (adjectives) <target 1>, (adjectives) <target 2>.  It doesn't have to be poetry, it's being translated into something incomprehensible to normal beings already. When you cast it, and when other things casts it, you hear an arcane language, not a rigid and artificial pattern.

Hmm. Butterfly example, using simplified input: "store <Urist's truename> mind, <truename> body", "turn <truename> parts, butterflies", "turn butterflies, chimps". Would be three seperate spells, but it's doable.

[Felblood]

I'd like to see lot's of hidden variables affect the exact cost and effectiveness of most magics, so you can't just check your clip and see that you've got enough juice for six more fireballs left in your staff. It needs to be mystical, not just plasma weapons.

I'd also like to see magic feared by people. Urist McGuy should look at the prospect of knowing a single word of power the way modern people would look at driving a plutonium fueled car. What if something goes wrong?

The gift for magic should be relatively rare, like most people have very little natural mana and little real connection to their more spiritual aspects. Only the gifted should be trained, and only the suicidally curious should try to learn magic without training.

Dwarves could even get annoyed if they are exposed to magical knowledge without their consent. Keeping conscripted talents from murdering the entire city with some kind of apocalypse spell should be challenge you have to overcome in order to produce good mages. Teach them the harmless stuff first, because you don't want them to be able to burninate people, until they've gotten over their distrust of magic.

cause that won't be buggy at all...

Care to be more specific, Oh Sarcastic One?

[Granite26]

You're talking about writing and interpreting a full language, one of the few things computer science is still bad at.  If normal people could write algorythmically, I'd be out of a job...   Seriously: It's hellah complicated to write clearly, and even then, people would have trouble using it.


As a helpful hint, keep power source separate from effects... makes the discussions easier.

[Felblood]

Yeah, text parsers suck. We've been over that. That's the problem we're trying to help solve.

And it's not a bug, it's just another danger of using magic. Plaid beards for everyone!

Do you have anything more constructive to add?

Power source and effect are critically linked. Yes, we probably should limit this discussion to how words shape the magic, and not worry about where the magic comes from, right here. --but it's a legitimate concern when mana in numerical quantities looks like the best way to complete some of the concepts that were proposed.

If we don't address these points [NO_MAGIC] guys are going to come in and derail our thread with some redundant argument over flaming swords.

[Mel_Vixen]

Ok it hasnt to be "Mana", exhaustion etc. would work also. An elf can by the way say "ok i have 15 Arrows left" why shouldnt be an mage able to say "ok i feel like could throw another curse"? Anyway your are right let ud delay the energysource for an moment.

Side-effects: Give any Sphere "sideffects" which ocur by "casting" level and legendary mage would have an much lower chance to have (fatal) side-effects. An spell with murder sphere could for example poison our mage-apprentice slightly.

The power of the spell and the sideeffects also could correlate with the surrounding and the astrology. Firespells get an plus in firelands and another one with an good Starconstellation. This might be dangerous for the apprentice cause his spell would be migthier but by casting an firespell he could set himeself in flames.

[Granite26]

I'm just saying that 3-4 guys on a game forum aren't going to solve through kubitzing what a generation of computer scientists and linguists have failed to do.  It's a warning about the complexity level and a statement that I don't think an open ended language parser is a good way to take this.  If you want something that's readible by people who don't have a noun-verb native language, I'd take a look at the old Changelings system.  It's got effect words, target words, and cost words.  If you structure magic as a formal language where each sentance has three phrases (the noun, or target, the verb, or effect, and the cost or power), you should be able to simplify the parser significantly.  Either the noun will be specific and abstracted (select the person, and the noun becomes their string), specific and random (type in a string and a match is guessed at) or generic and specific (type in the string and all matches are selected).  Effects are the classic spells and the spell effect phrase could contain other modifiers.  Cost is... well cost.  how much juice you're willing to put into it.


As far as cost is concerned, it's not really that important.  Either a unit has the power to do something, or it doesn't.  How the acquisition of that power is modeled has nothing to do with the effects.  At the end of the day, there will either be a skill check, a mana pool, or a threshold.  Any system you devise will be reduced in code to one of those three situations.  (or a combination thereof)

[flabort]

if the discussion is shifting to sources of power, i have something to add.

mana is a good idea, but more as something in the air, not from the wizzard. three invisible values are added to each tile:
mana
distortion
delay

if a wizzard casts a spell, it will tire him out a little, but it sucks mana out of the surrounding tiles.
his spell is delayed by complexity (number of words) times average delay value in the tiles he removes mana from. the spell will increase the distortion in the same tiles, as well as delay.

when the distortion value in a tile reaches a certain value, it will cause unintended effects around it, like turning all beards near it purple, and changing the effects of spells.

mana value increases over time, distortion decreases over time, and delay can be decreased with certain spells only, other wise.

[Eagleon]

I'm just saying that 3-4 guys on a game forum aren't going to solve through kubitzing what a generation of computer scientists and linguists have failed to do.  It's a warning about the complexity level and a statement that I don't think an open ended language parser is a good way to take this.  If you want something that's readible by people who don't have a noun-verb native language, I'd take a look at the old Changelings system.  It's got effect words, target words, and cost words.  If you structure magic as a formal language where each sentance has three phrases (the noun, or target, the verb, or effect, and the cost or power), you should be able to simplify the parser significantly.  Either the noun will be specific and abstracted (select the person, and the noun becomes their string), specific and random (type in a string and a match is guessed at) or generic and specific (type in the string and all matches are selected).  Effects are the classic spells and the spell effect phrase could contain other modifiers.  Cost is... well cost.  how much juice you're willing to put into it.


As far as cost is concerned, it's not really that important.  Either a unit has the power to do something, or it doesn't.  How the acquisition of that power is modeled has nothing to do with the effects.  At the end of the day, there will either be a skill check, a mana pool, or a threshold.  Any system you devise will be reduced in code to one of those three situations.  (or a combination thereof)

We're not talking about making a language that can communicate anything in an aesthetic way. The goal of magic is to do things. A generation of computer scientists have made languages that do things many times over. It'd require a fairly rigid syntax if we didn't want to spend years working on it, but I don't think it's impossible to get it working in an entertaining and novel way.

The nice thing about this system is that you can have it draw from multiple sources of power, or have the power determined by skill, etc. There could be other systems too... I dunno how they'd interact. That really would depend on the source of both powers.

[Faces of Mu]

if the discussion is shifting to sources of power, i have something to add.

mana is a good idea, but more as something in the air, not from the wizzard. three invisible values are added to each tile:
mana
distortion
delay

if a wizzard casts a spell, it will tire him out a little, but it sucks mana out of the surrounding tiles.
his spell is delayed by complexity (number of words) times average delay value in the tiles he removes mana from. the spell will increase the distortion in the same tiles, as well as delay.

when the distortion value in a tile reaches a certain value, it will cause unintended effects around it, like turning all beards near it purple, and changing the effects of spells.

mana value increases over time, distortion decreases over time, and delay can be decreased with certain spells only, other wise.

OMG that's a really awsm idea, flabort! 

[alfie275]

I dont think language parser would be that complicated, the spell system we are describing is just a very basic scripting system, just instead of 'Set Urist.pos(gobx,goby,gobz)' its 'Teleport Urist to Goblin' the parser just finds the word and replaces them with parts of code.

[flabort]

...quote flabort...
OMG that's a really awsm idea, flabort! 

thanks. the only problem with it is dettermining what type(s) of distortion would be caused, which would make over 50 invisible values labled "distortion" per tile, because of all the spheres. actually, that would be a lot of lag, but it would be an OK system until toady can find another, if he even uses it.

to summarize my idea:

pros: power source, side effects, miscasts
cons: lag, side effects, misscasts.

Edit:
oh yeah, wizzards should be able to work together. if two (or more) wizzards are in adjecent tiles (or further, provided there are wizzards in tiles between) cast spells with overlapping delay values, and atleast one agreeing spere, the spells can combine without causing as much mana drain. they will cause slightly less distortion and delay, and be more powerfull. esentially, if wizzards hold hands and cast at the same time, they will act as cattylists, meaning they use less power together.

if the spells are EXACTLY the same, then they combine fully, doing the effect stronger, with more range, using the mana for only one of the wizzards. (so if 40 wizards are in a line, and all say "freeze goblin invader" at almost the same time, they might end up freezing not only the attacking goblins in their fortress, but on another world entirely, if thier good enough. they'll also only use like 32 of 400 mana in the five tiles beside the end wizzard)