that probably would have made this a little easier. my setup isn't very necessary for what it does at the moment, but i'm thinking further down the line, for latency reasons, i'm going to have to, effectivley, gate off portions of my clock circuit so that only the blocks that have to be activated every tick are the only blocks activated on every tick.
what i'm working on is a game where lpayers spawn at the bottom floor of a tower. mobs spawn in at random intervals. each floor has a unique mob. plays start with a countdown timer which is extended for killing mobs and extended even more by killing other players. when a player does they respawn on the current floor. when the timer runs out: they are moved up a floor where different mobs spawn and a new timer starts. the rules for the higher floors are the same, except there are traps which affect the floor below. killing a player on the floor beneath you moves you back down a floor. when a player reaches the top of the tower they are out. once all players are out: the winner is rewarded and the circuit resets.
where the gating comes in is so that my clock only has to engage the parts of the circuit corresponding to the floors which are in use. or so that the higher floors aren't crowded wall to wall with mobs by the time players get there.
let's say objective=pigkills and objective2=timer. killing a pig adds 20 seconds to a depleting timer and decreases the count of pigs killed by 1.
this works well and fine so long as no player has a pig kill count higher than 2 since the block will continue to test positive and not cycle the comparator. so i suppose my question would be: is there a more elegant way of handling this than having a large tesfor bank at _min=2,_min=3,_min=4, etc?
maybe a clean way to force the comparator off?
hmm. i think i may have solved this already. i'm going to try replacing the comparator with air and creating a new one with two extra blocks at the end of the test. i'm going to go ahead and post this anyway incase someone finds it useful.
EDIT: confirmed
between the first repeater and block2 i have a split that runs to block4 and block5, separated by a repeater.
block4:
setblock x y z air
block5:
setblock x y z unpowered_comparator
this seems a pretty functional way to structure clock derived tests.
an important consideration seems to be that the testfor block needs to be triggered at least once while the comparator has been replaced with air. otherwise when the comparator is brought back in it will still be powered, and since the block in front is now underpowered: you can wind up with an unintentional clock loop. so for this to work there needs to be a delay between the removal and addition of the comparator that is greater than one tick of your game clock.
i've been having the same issue. but i do have a few insights.
it seems to have to do with how quickly minecraft can register inputs when opening chat.
i'm thinking an input registering too close to a T press will compete for priority and get bumped off.
for instance if i type "//paste"(using the command input not chat) really fast: sometimes it wont register the second forward slash.
now; if i press T manually to open chat and than press ^v to paste: the script will function properly.
i'm digging around trying to figure out how to set a delay between the scripted T input and the rest of the command, but it's looking like i'd have to rewrite my whole script in a different input mode to get that working as Sendmode Input sets an internal delay of Sleep 0 after every input. my thinking is that one of the patches for minecraft changed something and the chat bar took a hair more time to open making Sleep 0 no longer a sufficient wait peroid to recieve the rest of the command (competing for priority).
if i get this worked out i'll post a sample command.
i don't know if notch already has ideas for implementing this(or not) yet, but to me this seems like something pretty vital to the mod scene. especially as far as multiplayer is concerned.
my thoughts on this are:
allow severs to send content(mods) to clients on connect(content verification).
give each server with content which differs from the base game(not vanilla)its own folder or .jar file in which to store the modded content with a config file which tells the client which files need to be loaded(or instanced) for which servers.
this would allow use of a wide variety of content mods in a multiplayer setting without having to worry much about incompatibilities (such as you would with the current modding pipelines). it would generally translate into a more seamless user experience. not to mention the ability to run a server modded with something other than just server-side scripts.
again; i have no idea what notch has in mind for the topic. the framework for it is obviously there given how the game updates its self. one would think the inclusion of a server->client content pipeline wouldn't be much of a stretch from where things currently are.
bump. same issue. was getting some **** with my video card and the game, so i updated my drivers. now i get the same error.
i'm running an ATI 3870 HD
minecraft ran just fine until recently. so today i decided to do some dusting, as i noticed it was running a little hot and my display driver was seizing. dusting makes it run like it did brand new, but minecraft was still giving my display driver flack. thats when i decided on a proper driver upgrade hoping that would solve my issue. now i can log in and get an error log dumped onto my desktop.
note: using the latest catalyst drivers for the 3xxx hd series.
caught word of this server from mr.fribbles. been looking for something a little more populated than the one i've been hosting. how does one qualify for your whitelist?
this is my first time actually using java for a personal project and not an assignment. i haven't touched it since college. i'm making pretty good progress considering i'm still getting familiar with the code base. personally i like to just jump in and start out totally lost.
the link is a good un. it got me to my current mod in a day never having seen the code before. eclipse is really helpful too. the black and item tutorials are packed absolutely full of useful examples. though if it is your first time with code it will take you a few days to know to read it. i'd do a search for statements you don't understand while you follow the tutorials. most of it will be easy to grasp. then you'll come across something that looks totally alien to your eyes, and have to bang your head against the desk until you get it. then it gets simple again until you **** something up without realizing it. which is promptly followed by the desk receiving 30 lashings from your face. suddenly: everything works exactly like you want it to, you get +1 happy. . . until you run into some more alien statements. this goes on until eventually you break your desk or knock yourself unconscious. if that sounds like an appealing way to spend every waking moment, then programming is for you.
no joke though; it is seriously one of the most frustrating and rewarding things you can do with your time.
i managed to get the damage working, but broke my item pickup in the process. though i did manage to get the trajectory right before breaking the damage trying to restore item pickup.
i managed to get a hacky way together to make testing less frustrating for now (no more tree punching).
both pickup and damage are working, lets see if i can sneak my rebound and movement in without breaking anything this time; now that i've figured out enough of what is happening inside the arrow code.
java is an easy language compared to just about everything aside from qbasic and pascal. you just need to be able to teach yourself, and know how to do research. good critical thinking skills don't hurt either.
if you're smart enough to figure it out on your own from there you'll make an alright coder. if you're not. . . i'd take omega's advice. there's enough resources out there to get you started, but no one is going to hold your hand though it. being a coder is all about being resourceful.
much appreciated. the only one i've seen so far is SDK's mod, but he hasn't released a pack yet, so i cant peek through it. when i'm finished with this i'll comment things out a bit and post the source.
right now i'm trying to hack the EntityArrow code down to the hit calculations. basically i dumped half of the arrow code into my hammer entity, and am slowly stripping movement, inground, arrowshake, and various other bits away just so i can get to the embedded hit calculation. kind of a bugger to do the way its put together.
EDIT: to any interested check out Vec3D.java and MovingObjectPosition.java. was really hoping there was an easier way to do it than to do localspace detections.
0
what i'm working on is a game where lpayers spawn at the bottom floor of a tower. mobs spawn in at random intervals. each floor has a unique mob. plays start with a countdown timer which is extended for killing mobs and extended even more by killing other players. when a player does they respawn on the current floor. when the timer runs out: they are moved up a floor where different mobs spawn and a new timer starts. the rules for the higher floors are the same, except there are traps which affect the floor below. killing a player on the floor beneath you moves you back down a floor. when a player reaches the top of the tower they are out. once all players are out: the winner is rewarded and the circuit resets.
where the gating comes in is so that my clock only has to engage the parts of the circuit corresponding to the floors which are in use. or so that the higher floors aren't crowded wall to wall with mobs by the time players get there.
0
first: the setup
game clock -> block1 -> comparator -> repeater -> block2 -> repeater -> block3
block1:
testfor @a[score_objective_min=1]
block2:
scoreboard players add @a[score_objective_min=1] objective2 20
block3:
scoreboard players remove @a[score_objective_min=1] objective 1
let's say objective=pigkills and objective2=timer. killing a pig adds 20 seconds to a depleting timer and decreases the count of pigs killed by 1.
this works well and fine so long as no player has a pig kill count higher than 2 since the block will continue to test positive and not cycle the comparator. so i suppose my question would be: is there a more elegant way of handling this than having a large tesfor bank at _min=2,_min=3,_min=4, etc?
maybe a clean way to force the comparator off?
hmm. i think i may have solved this already. i'm going to try replacing the comparator with air and creating a new one with two extra blocks at the end of the test. i'm going to go ahead and post this anyway incase someone finds it useful.
EDIT: confirmed
between the first repeater and block2 i have a split that runs to block4 and block5, separated by a repeater.
block4:
setblock x y z air
block5:
setblock x y z unpowered_comparator
this seems a pretty functional way to structure clock derived tests.
an important consideration seems to be that the testfor block needs to be triggered at least once while the comparator has been replaced with air. otherwise when the comparator is brought back in it will still be powered, and since the block in front is now underpowered: you can wind up with an unintentional clock loop. so for this to work there needs to be a delay between the removal and addition of the comparator that is greater than one tick of your game clock.
0
i've been having the same issue. but i do have a few insights.
it seems to have to do with how quickly minecraft can register inputs when opening chat.
i'm thinking an input registering too close to a T press will compete for priority and get bumped off.
for instance if i type "//paste"(using the command input not chat) really fast: sometimes it wont register the second forward slash.
now; if i press T manually to open chat and than press ^v to paste: the script will function properly.
i'm digging around trying to figure out how to set a delay between the scripted T input and the rest of the command, but it's looking like i'd have to rewrite my whole script in a different input mode to get that working as Sendmode Input sets an internal delay of Sleep 0 after every input. my thinking is that one of the patches for minecraft changed something and the chat bar took a hair more time to open making Sleep 0 no longer a sufficient wait peroid to recieve the rest of the command (competing for priority).
if i get this worked out i'll post a sample command.
0
my thoughts on this are:
allow severs to send content(mods) to clients on connect(content verification).
give each server with content which differs from the base game(not vanilla)its own folder or .jar file in which to store the modded content with a config file which tells the client which files need to be loaded(or instanced) for which servers.
this would allow use of a wide variety of content mods in a multiplayer setting without having to worry much about incompatibilities (such as you would with the current modding pipelines). it would generally translate into a more seamless user experience. not to mention the ability to run a server modded with something other than just server-side scripts.
again; i have no idea what notch has in mind for the topic. the framework for it is obviously there given how the game updates its self. one would think the inclusion of a server->client content pipeline wouldn't be much of a stretch from where things currently are.
0
i'm running an ATI 3870 HD
minecraft ran just fine until recently. so today i decided to do some dusting, as i noticed it was running a little hot and my display driver was seizing. dusting makes it run like it did brand new, but minecraft was still giving my display driver flack. thats when i decided on a proper driver upgrade hoping that would solve my issue. now i can log in and get an error log dumped onto my desktop.
note: using the latest catalyst drivers for the 3xxx hd series.
0
0
0
the link is a good un. it got me to my current mod in a day never having seen the code before. eclipse is really helpful too. the black and item tutorials are packed absolutely full of useful examples. though if it is your first time with code it will take you a few days to know to read it. i'd do a search for statements you don't understand while you follow the tutorials. most of it will be easy to grasp. then you'll come across something that looks totally alien to your eyes, and have to bang your head against the desk until you get it. then it gets simple again until you **** something up without realizing it. which is promptly followed by the desk receiving 30 lashings from your face. suddenly: everything works exactly like you want it to, you get +1 happy. . . until you run into some more alien statements. this goes on until eventually you break your desk or knock yourself unconscious. if that sounds like an appealing way to spend every waking moment, then programming is for you.
no joke though; it is seriously one of the most frustrating and rewarding things you can do with your time.
0
i managed to get a hacky way together to make testing less frustrating for now (no more tree punching).
both pickup and damage are working, lets see if i can sneak my rebound and movement in without breaking anything this time; now that i've figured out enough of what is happening inside the arrow code.
0
0
0
if you're smart enough to figure it out on your own from there you'll make an alright coder. if you're not. . . i'd take omega's advice. there's enough resources out there to get you started, but no one is going to hold your hand though it. being a coder is all about being resourceful.
0
right now i'm trying to hack the EntityArrow code down to the hit calculations. basically i dumped half of the arrow code into my hammer entity, and am slowly stripping movement, inground, arrowshake, and various other bits away just so i can get to the embedded hit calculation. kind of a bugger to do the way its put together.
EDIT: to any interested check out Vec3D.java and MovingObjectPosition.java. was really hoping there was an easier way to do it than to do localspace detections.
0
0
that's the joke.