Jump to content
  • Advertisement

Zaxx

Member
  • Content Count

    64
  • Joined

  • Last visited

Community Reputation

142 Neutral

About Zaxx

  • Rank
    Member
  1. Zaxx

    I was working on my operating system...

    Quote:Original post by joanusdmentia So, his system *can* be crashed..... I guess, but only by Him (God). So, I guess He wouldn't have to worry about a virus. And for the record, I bet God would code His OS in 100% assembly code. I don't know about Jesus though.
  2. For the few "games" I've made so far in java (they're really engine tests), I've been using Swing for the gui. I use Eclipse, and I've recently heard about its SWT library as well as its Visual Editor, and I was thinking of checking it out. I'm mainly interested in creating applications (not applets), and want my games to work on both Windows and Macintosh. So, has anyone here tried either one for java games?
  3. Zaxx

    Sleep Paralysis-OOBE

    This stuff about sleep paralysis reminds me of something that happens when I'm in bed sometimes. I don't know if this is actually sleep paralysis or not, but sometimes when I'm in bed and dozing off I have this sudden sensation of falling and landing. What I mean is that I don't feel like I'm falling, but that I feel like that I just landed in my bed after falling. This always shakes me fully awake, and is really weird. :P
  4. Quote:Original post by Vopisk ...someone who is very strong, and given that we're talking about "dodging ability" rather than a "health meter", we can assume that if they are able to "dodge/parry" the attack, they have a good chance of being able to repel or block an attack or even if the attack is successful, take a lickin' and keep on tickin'. Someone with high dexterity on the other hand, may be weaker physically and more prone to suffering from a powerhouse hit from the stronger guy, though they're heightened agility gives them a better and more frequently likely ability to dodge. That's a good qualitative description of what "stength" and "agility" can do when one's dodging. I probably couldn't have thought that up better myself. Thanks. :) Quote:Original post by Vopisk I don't think that there is any real way to pull this system off without having a randomized element, because otherwise, the guy with one extra point in any given area will win if both elements are balanced equally, and then you don't have a combat system, you have a ASCII-based movie playing in front of you. Sit back and await the outcome. Quote:Original post by John Kowawsky Removing de "dice rolling" will make the fight as simple as "who-have-more-attribute" and it is not very interesting, specially when it involves more than 1 attr. Some luck make the player have to gamble (knowing his own strength, and guessing the other one) if he can win that mean thugh over there. Isnt that what we make in real life? :) You're probably right. It would probably make things slightly easier to think about anyway: agility affects the chance that someone has of hitting someone else, while strength affects how much "dodgeness" is depleated from the target. Then I wouldn't have to worry about how to take both strength and agility into account with hitting and dodging. Quote:Iron Chef Carnage Your "critical hit" idea, Zaxx, brought that to mind. It does indeed seem like it would led to a spiral of doom as a slight disadvantage would reduce your effectiveness so you'd get hit again and be further crippled Hmmm. I never thought about it leading to a "spiral of death." Thanks for bringing it up. Quote:Original post by John Kowawsky Putting a dodge bar only would dupe the life bar so the player would have to worry 2 bars. I think thats not a good way. Play Star Wars Galaxies and youll understand what i mean. It have 3 health bars and 3 mind bars, and it gets very annoying. I was indending to get rid of the health bar and just have the dodge bar. In a way, my dodge bar is just like a health bar, except its recharge rate is a matter of seconds. Quote:Original post by John Kowawsky BTW why youre trying to avoid randomness? I was trying to make the combat (relatively) simple so it wouldn't have to get more attention than the other gameplay mechanics I'd implement in the full game. That's also why I just have two physical stats. I wanted to put some emphasis on player skill as well, considering that I was thinking of this being done in realtime.
  5. Removed link. It has probably served its purpose now. :( [Edited by - Zaxx on September 27, 2005 8:09:36 AM]
  6. You're right Iron Chef. It is a meter race I have in mind. The strength vs. agility thing was the first thing I thought of, and is probably the best system for my game's purposes. It's just that then I start thinking about how agility can have an affect on hitting an opponent as well; while high strength allows harder & more forceful blows, high agility allows quicker and more accurate strikes. One way to take into account both strength and agility when determining damage is to just add the two values together, but that seems kind of pointless to me (remember that I'm trying to avoid randomized elements). What I'm really trying to wrap my head around is what specific roles would strength and agility have in attacking and defending. Maybe if I thought about the extremes and picture how fights between two people with either high strength or high agility would look like (eg. The Rock vs. Jacki Chan, Jacki Chan vs. Jacki Chan, & The Rock vs. The Rock). I'm going on like this about the stats because I think my system uses them differently than one where hp is involved. Traditionally strength affects how much damage is done, agility/dexterity affects how likely damage is done with a hit, and sometimes there's constitution which determines how much health you get (for my system I assume strength and constitution is the same thing). The special moves you came up with are pretty good, Chef. Wanting to keep things simple, I wasn't thinking of introducing special moves (yet), but I did have an idea about physical damage. As I said, the dodge meter I have in mind does not represent any actual injuries. However, I've been thinking that each attack can sometimes inflict actual physical damage on the target, which can be thought of being critical hits. There's no hp loss involved; instead the target's strength or agility is reduced by some amount.
  7. ...And all I had in mind was a small zzt-style ASCII roguelike engine that I could get working in two or three months. [looksaround] It's just that I've been thinking about I guess we can extend the issue to rpg stats in general. What kind of qualities would one expect from someone with lots of physical "strength," as opposed to someone with lots of physical "agility" or "dexterity?" Can the two traits be treated equally or does one tend to be more important than the other, especially when trying to create "realistic" combat (where people don't just keep smashing axes into each other until someone runs out of hp)?
  8. I'm trying to think of a simple combat system for a theoretical rpg-ish game. It's real-time and simplified, but it deviates from the standard "health" system. Mainly I want actors to have a "dodge bar," which represents how many hits they can fend off before succumbing to a fatal blow, though it doesn't mean they actually get hit. Gameplaywise I imagine it would be a lot like health as we know it now, except the dodge bar will always be rising. An attacker is trying to deal "damage" faster than the defender's doge bar can recharge. My real question is how to involve the various stats I want to have involved here. The thing is, I want my game to be as straightforward and streamlined as possible. Because of that, I came up with only two physical traits something can have: Strength and Agility. I want them to be of equal importance to a character, so a player can choose which stat to invest in. I want a character that's all strength to be equal to a character to be all agility, but for both characters to have to employ different fighting styles. Meanwhile, a character with both high strength and high dexterity would be the most powerful. I want this system to be as deterministic as possible; no dice rolls if I can get away from it. It's just that, from what I've been able to come up with, agility tends to trump strength in combat. I decided agility would determine how fast one's dodge bar refills. When I think about it, agility could also affect how often an actor can make an attack and how accurate its attacks are. That doesn't seem to leave much stuff for strength to do. Strength would affect how powerful one's blows are; stronger blows would take away more from the target's dodge bar. But agility can have the same affect, since one attacks would be more accurate with higher agility. I suppose to determine how much the dodge bar gets depleted I'd sum the actor's strength and agility, but I could do that with just an "attack" stat instead. So, what roles do you people think Strength and Agility can have in combat while being equally important?
  9. Your "Quake 3" system sounds the most reasonable so far, so I'll work on that. That looks a lot like the final collision algorithm I read in the article General Collision Detection for Games Using Ellipsoids (did you write that, oliii?). It's probably time I worked on the collision response anyway. I just wanted to clarify one last thing: as I calculate the response on a body do I actually update it as I go, or leave that until after similar to what I've been doing? I'm asking because I feel that if I don't update a body then when I move on to the next one I wouldn't know if it would ever hit the former as it (the former) went along the path I calculated before.
  10. Quote:Original post by oliii if all was perfect, you would never get overlaps. Are you talking about my collisionTest method? As if my problem could actually be in that? Quote:Original post by oliii Also, don't consider a collision if (entity0.Pos - entity1.Pos).Dot(entity0.Vel - entity1.Vel) > 0.0f since the entities are moving away from each other, this will stop them getting glued together in case of an overlap. Don't worry, I do check this. My collision test is based on the Gamasutra article Pool Hall Lessons: Fast, Accurate Collision Detection between Circles or Spheres (login required). My algorithm, for anyone that's curious, goes like this: public double collisionTime(Actor A, Actor B) { Get relative velocity from A to B (treat B as stationary). -> vel if (vel.magnitude == 0.0) return 1.0; // Actors stationary to each other. Get relative position from A to B. -> pos Get minimum possible distance between A and B: sum of their radii. -> minDist if ((vel.magnitude + minDist) <= pos.magnitude) return 1.0; // Actors would not reach each other. Get unit vector of vel. -> unitVel Get dot product of unitVel and pos. -> dotprod if (dotprod > 0.0) return 1.0; // Actors either moving parallel to or away from each other. /* dotprod is also the projection of pos onto vel. dotprod is the distance along vel that's closest to B. Therefore, the shortest distance between B and vel is the square root of the squares of the magnitude of pos and dotprod. */ Get shortest distance between B and vel, (distance between B and its projection onto vel). -> F // F is shortest distance A would ever get to B as they travel. if (F > minDist) return 1.0; // A wouldn't touch B. // F makes right triangle with minDist and a third value of T. Use pythagoras to get T. // If no such right triangle exists, T^2 would be negative. if (T^2 < 0.0) return 1.0; Get the real distance, which is dotprod - T. -> realDist if (vel.magnitude <= realDist) return 1.0; // Final distance greater than relative velocity. double time = realDist / vel.magnitude; if (time < 0.0) return 0.0; else if (time > 1.0) return 1.0; else return time; } Anyway, thanks for your help oliii. I'll try out your threshold method to see if that does anything.
  11. I'm testing my engine's collision and movement code. I've set up a simulation where I have 15 or so entities start at random locations on the screen, facing at random angles, and have them all move straight ahead until they bump into each other. There's no real collision response; if an entity is blocked it just doesn't move. My problem is that after all the entities are clumped together and can't move anymore some overlap with each other. As far as I know the collision test itself is working. The entities are represented as circles and are given a movement vector. When two entities are compared the collision test returns a value between 0.0 and 1.0, which represents the time within one frame that the two entities would collide (0.0 means they've already been touching, 1.0 means there'll be no overlap). The movement vectors of the two entities are multiplied by this value to get the respective translation vectors for the next update. The test worked well when I just had two entities, so I think the problem lies in when I test all the entities against each other. My algorithm looks like this: Actor entity[] = new Actor[15]; // All the entities. ... double time[] = new double[entity.length]; // Keep track of translation times for all entities for collision. ... update { // Set velocity vectors for all entities. for (int i = 1; i < entity.length; i++) entity.updateMovements(/* various variables */); // Collision tests. for (int i = 0; i < entity.length; i++) { for (int j = i + 1; j < entity.length; j++) { double tempTime = collisionTest(entity, entity[j]); // Entities' velocities multiplied by smallest calculated time. if (tempTime < time) time = tempTime; if (tempTime < time[j]) time[j] = tempTime; } } // Actual movement. for (int i = 0; i < entity.length; i++) { entity.move(time); time = 1.0; } } I test all the entities against each other, keeping track of the times of collision between them. But I always keep track of the smallest times, so if the collision time for entity A and entity C is kept over the time for entity A and entity B; thus A is translated by the lower time (as it was to collide with C). I wonder if this is where I'm getting my overlaps. If I kept the time for A and C over the time for A and B, meaning A hit C before B, what do I do with B? If C hit A before B did, would I have to calculate the translation time for B again? I would be appreciative if someone could help me sort this out. Thank you.
  12. Zaxx

    Scent-Enabling ...

    <my $0.02> Just so you know, some people (like me) are allergic to artificial scents. A "smell pack" is probably not a good idea, unless you made the smells entirely from all natural ingredients like fruit juices and real flower nectar. [rolleyes] As for Moe's idea; unless smells were important to the game, I probably wouldn't care. </$0.02>
  13. Remember City of Heroes? Remember how much you could customize you appearance, allowing you to create almost any avatar you want? Remember how players took this customization and created thousands of Wolverine clones? [rolleyes] Will Wright's Spore promises to allow you to create any kind of creature you want. Your creations are also supposed to be automatically transmitted to an online database then retransmitted to other Spore players, so you can meet the creations of others. What if Spore gets its own Wolverine; a certain creature that people will create over and over again. If this happens, what do you think the creature will be? My vote is on any of the pokemon.
  14. Quote:Perhaps it would be an adea to allow your vehicles to accelerate up to the max velocity rather than just applying this all the time. It would look nicer. I plan on doing this, but not right now. I need to clean up my physics engine, and I'd rather figure out most of the ai logic before having to take acceleration into account. Quote:I'd suggest using a method where you are always steering toward some point which may be calculated via a number of inputs. Would you mind elaborating on this a bit? From what I can tell you're proposing that for my ai I calculate a point or list of points that an agent needs to traverse, then have it steer itself toward these points. This might be good since I could incorporate pathfinding algorithms into my ai. However, I'd also like some of the dynamic behavior exhibited by Reynolds' simulations such as seek and evade, obstacle avoidance, or wandering.
  15. So I'm wanting to implement steering behaviours for the ai in my game. It's a top-down 2D arcadish game with a similar control scheme as Asteroids, where the ai agents track the player through a maze (they're hostile, of course). The ai has the same movement controls as the player: thrust, reverse, turn clockwise, and turn counter-clockwise. I've looked at Craig Reynolds' web site, an excellent resource on steering ai. However, I'm having a little difficulty figuring out how to adapt Reynolds' flocking/steering behaviours to the steering behaviours of my own ai. Reynolds' steering seems to depends on putting forces on an agent and have it accelerate along the resultant vector. My bots on the other hand have a fixed velocity they can thrust or reverse and a fixed rotational velocity for turning. Are there any resources online that deal with this kind of steering?
  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

GameDev.net is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!