Wikipedia:Reference desk/Archives/Computing/2009 May 17

= May 17 =

Who would win in a fight generator
I'm looking for a program or something that (without any programming on my part) would predict the outcome of a fight between video game characters and other fictional creatures (I.E. Link vs. a Vampire). Me and my girlfriend are at a stalemate and would love to just know the outcome and figure SOME ONE had to have done this.  Rgood erm  ote    00:49, 17 May 2009 (UTC)


 * How do you expect a computer program to do that? The question isn't well defined. You would need to have information on their fighting capabilities in a way that can be compared - that information doesn't exist for characters that exist in different fictional universes. In most cases, it won't even exist for characters in the same fictional universe. --Tango (talk) 01:26, 17 May 2009 (UTC)


 * I expect some one to have already done it for me. Probably in Javascript or PHP.  Rgood erm  ote    01:29, 17 May 2009 (UTC)
 * But how? It's impossible. The information needed to compute the result doesn't exist. --Tango (talk) 01:33, 17 May 2009 (UTC)
 * Not a gamer? The stats exist. Open up a game guide and they have em right there. As for the stats of a vampire, those exist as well.  Rgood erm  ote    01:39, 17 May 2009 (UTC)
 * The stats only exist in a way that allows for comparison between characters in the same game (you may be able to compare height and weight, but not skill and certainly not any special powers). And what stats are there for a vampire? You need to be more precise about what you mean, do you want Dracula? A D&D vampire? A typical Buffyverse vampire? Only D&D has any real stats and, again, they are only useful for fights against other D&D characters. --Tango (talk) 01:45, 17 May 2009 (UTC)
 * They have specials in em. Plus, if that isn't enough. There are enough fan sites to make the hunt easy. As for vampire...the essential..which is Dracula.  Rgood erm  ote    01:49, 17 May 2009 (UTC)
 * Yes, there are stats for the specials, but not in a comparable format. You can't compare "4/10" in one game with "5/10" in another since they are completely difference scales and there is no meaningful conversion factor. I've read Dracula, there were no stats in it. You would have to base it on stats that someone other than the author has made up, so it isn't really Dracula. --Tango (talk) 01:56, 17 May 2009 (UTC)
 * The question and attitude indicate a probable troll. Tempshill (talk) 03:22, 17 May 2009 (UTC)
 * Gentlemen, the answer is obvious. The vampire would defeat Link.  You don't need a computer to tell you this.  Although, you can set up Smash Brothers to play by itself, matching AI characters against each other.  We even have the complete list of possible match-ups.  This may be the closest that modern technology can deliver.  Nimur (talk) 02:20, 17 May 2009 (UTC)


 * Obvious? I think not. The Light Arrows from Zelda would be a formidable weapon against a vampire, since you've basically got sunlight and a stake through the heart all in one blow. Even in the Zelda universe, the Light and Silver Arrows are supposed to be able to stop or kill immortals, so there you go. Link's definitely got a fighting chance. Indeterminate (talk) 07:33, 17 May 2009 (UTC)
 * The Internet never sleeps. Anyway, this goes to Link, no contest. He has a huge bag of tricks (let's not forget the Stone Mask), while vampires have, what, fangs and charm? And they can turn into bats? Link eats bats for breakfast. As if that weren't enough, if Link did die he would just restart the area with three hearts. Eventually the vampire is going down. -- BenRG (talk) 23:08, 17 May 2009 (UTC)

Fight generators are for pussies. Throw them all into a Blendtec, press the button; and at the end all you'll have will be Chuck Norris and bad guy smoke. RESOLVED -- Hoary (talk) 08:47, 17 May 2009 (UTC)


 * This video answers any questions you might have on the subject. No Javascript needed. --98.217.14.211 (talk) 14:12, 17 May 2009 (UTC)


 * There's no way you could make a computer program that does this in any useful way. You're probably thinking of all the hit-point stats available on the internet, but it doesn't work. How do we know the hit points are compatible?  If I base a game on a hitpoint scale of 1 to 10,000 then all my characters will slaughter characters from a game that runs from 0.0 to 1.0!
 * The more fundamental problem is the incompatibility of the "universes" the fictional characters come from. In a Zelda game, Link is virtually indestructible as long as he is holding up his shield. Does this mean that Master Chief doesn't have any weapons that could punch through Link's medieval shield? Even if Master Chief's weapons can't pierce the shield, can he use a sniper rifle to shoot Link's head? (It's never fully behind the shield in the animation of Zelda games.) Even assuming that Link survives Master Chief's attack and launches a counter attack, does the Master Sword effect MC? Zelda fan's will say "yes" because it's magical and can kill anything. Halo fans will say "no" because it's preposterous that the MC's armor is vulnerable to such ancient weapons wielded by such a small person.
 * There's no way a computer program could decide this sort of stuff on its own. If a program like this exists, it's only because someone has put in his opinion for all the variables. APL (talk) 16:03, 18 May 2009 (UTC)

http://www.googlefight.com/ CaptainVindaloo t c e 16:18, 18 May 2009 (UTC)
 * Thanks CaptainVindaloo, that's close enough to what I was asking.  Rgood erm  ote    03:15, 20 May 2009 (UTC)

I am not a gamer, but if scores already exist, then could he not just try 1) converting the different scores into a z score or failing that, a percentage, then 2) decide which of the scores from the different websites are more or less equivalent, then 3) decide what the weighting of the various scores would be in relation to the characters fighting prowess, then 4) calculating a weighted average fighting score for each character? 89.242.109.25 (talk) 12:46, 20 May 2009 (UTC)
 * Unfortunately, such a simplified scoring system would not work. The set of all things which can fight each other do not form a total ordered set, which is why all of these games have such complex scoring statistics to begin with.  (In plain english, consider this simple example, common in many games: infantry can defeat rockets; rockets can defeat armored tanks; armored tanks can defeat infantry.  Which has the highest score?)  Complex interactions between combatants' unique abilities result in the need for a nonlinear combination of capabilities which cannot be mapped to a single linear statistic.  In the pathological case, N different statistics are required for N different participants, with a unique match-up outcome for every possible combination.  If time-variation is permitted, or if there is an element of indeterminate (random) outcome, then those must also be parameterized.  The more realistic the model, the more sophisticated such parameters must be.  Nimur (talk) 14:30, 21 May 2009 (UTC)


 * A lot of academic assessments are done in the same way as I've suggested: does it mean they are nonsense? 78.146.18.113 (talk) 18:37, 23 May 2009 (UTC)
 * You don't generally use academic assessments to predict which of two individuals will do better at a specific task and they really aren't very good at that. They give you a broad idea of where in the population a person is, but you can only compare individual scores if they are either very far apart (that's why you generally just look at grades rather than actual percentages, the percentages aren't particularly more meaningful than the grades) or the task you are trying to predict the result of is very similar to the task they were assessed on. But none of that is really relevant - you can't compare z-scores from different populations without directly comparing several members of one population with several members of the other. The only way to compare them would be to get a character from one game to fight a character from another game and you simply can't do that. --Tango (talk) 18:56, 23 May 2009 (UTC)


 * There's always Googlefight.  Sp in ni ng  Spark  19:16, 22 May 2009 (UTC)

Using Google
I think there are ways I can improve my google searches by using speech marks and plus signs and other stuff, but I can't find a page to teach me these things. Any help?91.109.232.157 (talk) 08:45, 17 May 2009 (UTC)


 * Google search basics: More search help (it's linked on the "Advanced Search" page). -- da pete 08:58, 17 May 2009 (UTC)

This cheat sheet (http://www.google.com/help/cheatsheet.html) is also very useful, I use it myself. ny156uk (talk) 12:46, 17 May 2009 (UTC)

Inkscape banknote tutorial
Hi, guys. I'm interested in finding the full version of this tutorial, but it appears it costs! Does anyone know of a free version, alternative (or does anyone have it and is willing to email it...)? Thanks! ╟─ Treasury Tag ► contribs ─╢ 09:13, 17 May 2009 (UTC)


 * They are called spirograph images colloquially. Google "Inkscape spirograph" or "Illustrator spirograph" and you can find lots of different ways to make them (like this one or this one for Illustrator, but I suspect Inkscape can do similar things). --98.217.14.211 (talk) 14:17, 17 May 2009 (UTC)