Content: Slate Blackcurrant Watermelon Strawberry Orange Banana Apple Emerald Chocolate Marble
Background: Slate Blackcurrant Watermelon Strawberry Orange Banana Apple Emerald Chocolate Marble
Pattern: Blank Waves Notes Sharp Wood Rockface Leather Honey Vertical Triangles
Welcome to TerraFirmaCraft Forums

Register now to gain access to all of our features. Once registered and logged in, you will be able to contribute to this site by submitting your own content or replying to existing content. You'll be able to customize your profile, receive reputation points as a reward for submitting content, while also communicating with other members via your own private inbox, plus much more! This message will be removed once you have signed in.

  • Announcements

    • Dries007

      ATTENTION Forum Database Breach   03/04/2019

      There has been a breach of our database. Please make sure you change your password (use a password manager, like Lastpass).
      If you used this password anywhere else, change that too! The passwords themselves are stored hashed, but may old accounts still had old, insecure (by today's standards) hashes from back when they where created. This means they can be "cracked" more easily. Other leaked information includes: email, IP, account name.
      I'm trying my best to find out more and keep everyone up to date. Discord ( is the best option for up to date news and questions. I'm sorry for this, but the damage has been done. All I can do is try to make sure it doesn't happen again.
    • Claycorp

      This forum is now READ ONLY!   01/20/2020

      As of this post and forever into the future this forum has been put into READ ONLY MODE. There will be no new posts! A replacement is coming SoonTM . If you wish to stay up-to-date on whats going on or post your content. Please use the Discord or Sub-Reddit until the new forums are running.

      Any questions or comments can be directed to Claycorp on either platform.


  • Content count

  • Joined

  • Last visited

Posts posted by cevkiv

  1. Now that's just being stubborn lol, if they want a setting that allows no rogue block breaking, but allows seiges to break blocks, i like that. And nesissity is a funny thing, we don't NEED most any of these feautures, they are nice to have, and if you don't like them, don't use them. Have you never played a factions server before? and i realize it doesnt make them cooperate, but encourages it.

    Have you read anything I've written on the subject? It's entirely possible to create a mechanic that allows not only for block protection, but for bypassing of block protection by determined players without the need for some "war" mechanic. And if you can do it without the mechanic, you should do it without the mechanic.

    There is a principle in engineering called "KISS" -- Keep It Simple, Stupid. The more complicated and complex you make something the more points there are for failure of the system and behavior you don't intend.

    [edit] fixed spelling error


  2. Furthermore, in the second example, about faction relations, not having some hard-coded mechanic allows for more political intrigue. If two towns are in opposition to each other, you could have people from one town working behind the backs of their fellows as saboteurs, or engaging in clandestine trading with the enemy.

    The fewer restrictions you put on players, the more emergent behaviour you get. You don't want to needlessly restrict the player's options of interaction.

    The only thing absolutely needed is a form of block protection. A subset of block protection is protection of chests. I do not suggest making blocks completely invulnerable, nor making chests 100% secure forms of item storage. There needs to be a balance between security allowed and the freedom of other players to adopt other playstyles other than "I mine things and make things." You want players to have the options to be roving bandits. You want the option for players to raid others. This can be accomplished without needless mechanics and abstraction.

    Instead of forcing some model of government upon players via an in-game mechanic related to a town, you let the players decide on the form of government. It's an absolute dictatorship by default, because one person and one person only is in charge of dictating who can build in the protected area, but that does not preclude other forms of government. If the creator of the town wants it to be a democracy, it can be a democracy. If they want a communism they can enforce a communism. And so on, and so on. These things should not be hard-coded mechanics, but emergent behaviour that occurs when players band together for a common cause.


  3. War could allow for block breakage if not already enabled on a server

    If block breakage is disabled, it's disabled for a reason. Because the admin doesn't want people breaking the blocks to begin with. Therefore, there is no reason to add a function that will do absolutely nothing to begin with.

    , and would allow the leaders of factions to officially state their relations.

    I don't see how this is necessary. It adds needless complication onto something that doesn't need it. It's far easier and simpler to simply have the players themselves decide what the relations are. "Hey, Bob, are we trading with Larrytown?" "No, Steve, Larrytown is full of assholes."

    also to allow for unified attacks, rather than just rag tag groups of 4 guys fighting, instead 20-40 men all in one glorious battle.

    Once again, you don't need some complicated mechanic for this. Having this mechanic won't magically make people cooperate. They have to do that on their own. Therefore, the mechanic is still unnecessary. If people want to fight a "glorious battle" they will fight a "glorious battle". This "war" mechanic can't make them do that.


  4. What's to stop them, is needing to have an established town, Using NEI to spawn items requires OP. So really i don't understand what the problem is.

    Please explain exactly what the point of this "War" mechanic would be, and why it is necessary? It seems to me that it's an unneeded addition.


  5. Any formal, hard-coded system for declaring/initiating war between two parties should not be implemented. War should be purely a social construct. There is no need to add a mechanic for something that can be accomplished simply by player consensus.


  6. A summation of what I've seen so far is

    Some form of block protection, with the ability to dictate who can build within it, that makes it more difficult for lone individuals to break the blocks within the protected area

    A way to lock (and to pick the locks of) doors and chests

    Anything else?


  7. Another idea would be a Nest Box. This would be a craftable, placeable block. When placed, it would "suck up"/"vacuum up" all chicken eggs that are on the ground in a radius around it (so that they aren't lost due to someone not paying attention and picking them up). It would hold a single stack of eggs.


  8. I would probably add pigeons in as a mob that you would have to capture, but I agree, once the pigeon flies off with your message, it would disappear and then arrive at it's destination at the correct time.

    I would suggest that, instead of having them be a capturable mob, that during the generation process of a tree there is a very small (1 in 200, at the most) chance that one of the leaf blocks will instead be replaced with a bird's nest block. A bird's nest block would be a container with a 4 slot inventory that would contain between 0 and 4 bird's eggs. These eggs could then be hatched to create the required pigeons. Perhaps by carrying them around in your inventory for a while. Breaking the bird's nest before "opening" it to remove any eggs that might be inside it would destroy any eggs inside it. So there would be an incentive to look carefully at that tree before you start whacking at it with your scythe.

    The odds of generating a bird's nest probably needs to be lower, as it would be "relatively" easy to farm them with, say, your standard willow tree farm. Or, they could be more common, but have the probabilities distribution for the number of eggs in a nest be skewed towards 0. Say, 50% chance of 0 eggs, 30% chance of 1 egg, 15% chance of two eggs, 4% chance of 3 eggs, and 1% chance of 4 eggs.

    Detaching the message from the pigeon carrying it could cause the pigeon to "fly away" and be lost. Maybe not all the time, but something like a 50/50 odds of it happening, so that you have to keep going out and looking for more eggs, or keep growing more trees to hope more birds build more nests and lay more eggs.

    Or nests could only appear during the spring.

    [edit] added idea in italics


  9. I would go look at the beginning of this thread, as it mentions nutrition already.

    and quite detailed Cevkiv, i like it. and hold on, oil isn't needed for basic bread, and unleavened bread is also yeast isn't needed either (as noted), nor the sugar in that case. but, it is more like crackers like that, but it would be something yeah? and oil can come from a vareity of sources, including animals, rapeseed, soybean, and just about every plant... but yes, traditional bread would require most of what you said.

    I was working with things that were already in the game. The bread that one makes with three wheat is quite clearly leavened. And since olives are already in the mod, they would be an easy source of oil.


  10. Regarding the bread suggestions:

    Of course, we need some sort of a bread-making GUI, like smithing.

    The better you do at cooking minigames, like making dough, the more "durability" the bread would have.

    Then, when baking, it'll, at some point, start to lose "durability", and eventually burn to a piece of coal.

    I used to be a doughmaker at a pizza place. I would easily make something like 600 pounds of dough a week, sometimes more. This is what it involved:

    Fill a pitcher full of ice and add water. Toss it into a mix. Do this twice. Add yeast (amount varying based on temperature and weather. Yes, weather). Add some sourdough culture. Add two cups of sugar and two cups of oil. Let mix until ice melts. Add seven scoops of flour (very precise, eh?) Then, quickly add one to two cups of salt. Then you have about thirty seconds to determine if you got the amount of dough right, and if not, whether to add more flour or more water. Too much water and the dough is sticky. Too much flour and instead of a big clump you get lots of little ones. Let it mix until it gets to a good consistency, and then take it out, cut it into portions, put it in bags, and put it in the walk-in refrigerator until needed. Every morning we would take a tray or two of dough out of the walk-in, and let it warm up on top of the ovens. Over the hour or so it took us to get all the food prepped, the heat would have caused the dough to rise sufficiently.

    We kept a bucket full of sourdough to use as a culture for the regular dough (and when we were hungry and it was slow we'd make ourselves loaves of sourdough and eat it with oil and vinegar and wine). You'd use up about a cup of sourdough for each batch of dough. When you get low on sourdough, you pull what's in there out, leaving the juices in the bucket, cut the sourdough up, cut up some fresh down, mix it together, and throw it back in the bucket and let the bucket sit out (with the lid on) next to the ovens to rise. Every couple of hours you go and stab the fuck out of the sourdough to stop it from rising out of the bucket.

    I approve of the idea of a bread-making minigame. Making good bread consistently is very difficult and requires a great deal of practice to do so. Not only do you have to have a good understanding of what ingredients to add, but when to add them, how much of them to add in relation to each other, how long to let it mix, and how long to let it rise, but making a loaf of bread isn't as simple as tossing a ball of dough in the oven. You have to shape that shit, and, depending on what you're making and how fancy you want it to look, do other things to it, like cutting it. Furthermore, you don't just shove dough into an oven (at least we didn't. Our ovens were stone slab over natural gas burners). You have to spread some form of grit, like semonila, to avoid the stuff sticking. And even then you have to take a peal and work it under the stuff every now and then just to make sure it doesn't stick.

    To make dough, you would need







    Which means you would need

    Rock Salt to make table salt

    A grindstone, to grind wheat into flour

    A press, to get oil from olives

    A bucket of water

    Sugarcane, to get sugar

    And some way to get yeast, otherwise you're making unleavened bread.

    So "real" breadmaking would require salt mines, olive trees, two new kinds of blocks (grindstone and olive press), and a new type of resources, yeast.


  11. Also, using pigeons that don't actually have a model, and are just used as an abstraction, removes the need to make a model, to animate it, and to program the pathing for an AI. Also, what do you do if the pigeon has to travel over unloaded chunks?


  12. Mileaos: NPC's won't be playing that big of a role in cities at most they will become guards it would seem. Also we don't want all buildings to built off plans, we want them to be built the way the people want it and by the people.

    cevkiv: Pigeons sound like a good idea to me but what if any animal could be used for a messenger just pigeons would be faster and more dependent than say a pig.

    Because, historically, pigs weren't used as messengers. Pigeons were. Pigeons were used thousands of years ago, and even in World War 1.


  13. I really hope you and Bioxx do consider adding some sort of Pigeon Mail System in with the Kingdoms segment of development. It would really go a long way to making a period-appropriate method of communication between settlements. Except I think they actually used ravens. I can't remember. Anyway, it would be invaluable to have this system if you do end up adding OOC and RP modes of chat, which I would also really like to see implemented, and I don't think those would actually be that hard.


  14. I think you failed at that reference right there.

    (and seeing as how humans can sprint at roughly 30 km/h, although in mc its closer to 5.3 m/s, pigeons can fly at 80 km/h)

    If a "real world" pigeon can fly at 80km/h, and a MineCraft human can run at approx. 1/6th of the speed a normal human can, then a MineCraft Pigeon should fly at approx 14km/h. So at that rate a Mail Pigeon would take about 4 minutes 15 seconds to deliver a message that a person running on a completely flat surface with infinite stamina would take 11 minutes 19 seconds to deliver.

    [edit] So you could send a pigeon and get a response before you would even get there yourself.


  15. I forgot to add, and I'll say it here so it doesn't get lost in the Great Wall of China Text, that the Pigeon Mail System was made in mind with the thought of limiting the range at which players can read another player's text message. I think that there should be two chat modes, which can be switched between by players. The first mode would be OOC (Out of Character Mode). This would be the default mode for chatting, and would function exactly as chat does now: Everything anyone says (in OOC or RP Mode) can be seen by everyone else, and you are able to receive /tells.

    The second mode would be RP (Roleplaying) Mode. In this mode, you cannot see anything entered in the OOC chat channel, and other people in the RP chat mode can only see what you say if you are in physical proximity to them. Furthermore, in RP Mode, you can only receive tells from, and send tells to, a player in very close physical proximity to you (to simulate whispering).

    If this isn't added, the only purpose of the Pigeon Mail System is allowing players to send messages to others who are offline. If these changes are made to chat, then they won't affect people who choose to opt out of it (people who continue to use normal chat functionality), but will be invaluable for people who choose to roleplay, as it would be the only way for communicating between settlements.


  16. One question about the birds...

    How fast will they go, [...]

    Two to three times as fast as a person can run, at least. Furthermore, the time they take to travel is calculated "as the crow flies" -- they don't take into account hills and valleys and other bumps. But if my settlement is a hour's travel away from yours, a pigeon should be there in 10 to 30 minutes. Probably closer to 10.

    [...] and are they african or european?

    I don't knoooooooooooooooooooooooooooooooooooooooooooooooooooo



  17. Sorry for double posting, but I want to clarify my ideas in a more readable format, modified by the ideas of others I have read in this thread

    Town System

    Hearth Stone (placeable craftable block)

    The Hearth Stone is a craftable block. It should require at least medium tier items, but not extremely rare items to craft. For example, wrought iron and average gems. When placed, the Hearth claims a 5x5 chunk area, centered on the chunk the Hearth Stone is in. The Hearth Stone remembers the person who placed it (hereinafter referred to as the owner), and when activated opens a GUI. For anyone not the owner, the displayed GUI shows the owner's name, the current area claimed, and people who have permission to build in the area. For the Owner, this GUI will list the owner's name, the current size of the area claimed by the hearth, contains a list of players allowed to build in the claimed area and the ability to add more people to the list (after a certain point, adding more people should require the expenditure of resources), and displays the resource cost to upgrade the Hearth Stone to the next level, which increase the size of the claimed area by 1 chunk in all directions. This should be at least as expensive in resources as the Hearth Stone itself, and go up substantially as more area is claimed, up to some limit, which could be defined in a server configuration file.

    Dirt, Sand, and Naturally occurring rock blocks below the level at which the Hearth Stone is placed are automatically reinforced to the maximum reinforcement amount, to prevent people from simply tunneling under walls.

    The GUI further contains two buttons for both the Owner and for people whose names are entered into the list by the Owner.

    One button creates a tool that allows blocks to be reinforced (via expending other blocks of the same type from inventory), which dramatically increases the number of times someone not on the list of approved players has to break a block in the protected area to actually get rid of the block (for example, reinforcing a rhyolite brick once would require an additional rhyolite brick block and would make force an unauthorized user to break the block twice to actually break it, reinforcing it a second time would require two rhyolite brick blocks and would make the reinforced block have to be broken four times to actually break it, etc, etc, up to a maximum reinforcement amount that could be changed via server configuration files)

    The other button creates a tool (a pair of hand cuffs, say), which may be used by any player on the authorized list of the Hearth Stone on any non-authorized player within the protected area. Doing so would teleport the unauthorized player to a jail area. The Jail Area would be defined by the placement of a special Jail Block, and the area itself would be a 5x5 cube with the Jail Block at its center. Blocks and doors within this 5x5 cube are not breakable under any circumstances by players within the Jail Area -- they must be released by an authorized person of the protected area within which the jail is situated.

    Jail Block (placable craftable block)

    I have no idea what the items used to craft this would be. Probably lots of metal bars. It should be roughly as expensive (resource wise) to create as a Hearth Stone block. The jail block would provide a way to take care of griefers who are caught in the act. It creates an area, which is linked to handcuff items produced by the Hearth Stone block whose protected area the Jail Block is situated in. This area is where people who have handcuffs used on them are teleported to. Blocks within the Jail Area are not breakable by persons within the Jail Area -- they must be released by someone from outside the area who is on the authorized list of persons in the Hearth Stone of the settlement where the jail is located.

    Reinforcement Tool (item dispensed by a Hearth Stone to people on its Authorized User List)

    Use of this tool on a block inside a Hearth Stone's protected area by someone on the Authorized User List would allow the Reinforcement of that block by using other blocks of the same type within the reinforcing player's inventory. Each level of reinforcement requires 2^X blocks of the same type as the block being reinforced, where X is the number of times the block has been reinforced (For those of you who can't into math, the first reinforcement requires 1 block, the second 2, the third 4, and so on).

    Each reinforcement of a block requires it to be broken an additional number of times before the block will actually be destroyed, where the number of times a block must be broken to be destroyed is equal to 2^X, where X is the number of times the block has been reinforced (again, if a block is not reinforced, it need only be broken once, if it is reinforced once it must be broken twice, if it is reinforced twice it must be broken four times, etc, etc). I suggest a maximum of 6 reinforcements per block. This would require a single grief to break a block 64 times before it will be destroyed, while a group of 8 griefers (who I guess would call themselves a "raiding party") working together would only each have to break the block 8 times. This would have the effect of deterring lone individuals, but would not stop a determined group of players working together in a coordinated fashion.

    Furthermore, using a Reinforcement Toll on a block with the left-click (as though trying to break it), would display its current reinforcement level, and how many times it must be broken to degrade to the next lowest level. For example, if a block has been reinforced 5 times, using the info aspect of the Reinforcement tool would give a message like: "Reinforcement Level 5. Time to be broken to downgrade reinforcement level: 16".

    So, if a griefer is trying to break through the town wall, is caught in the act and killed, a citizen could then check the area of the town wall around the griefer to find the damaged sections of the wall, and then could use the reinforcement tool to "repair" them. As in the above example, if a block had been reinforced 5 times, and had been broken 8 of those times, it could be repaired by to full be expending the number of blocks required to reinforce it to its current level from the previous level (in this example, 16 blocks to repair it fully).

    Handcuff Tool (item dispensed by a Hearth Stone to people on its Authorized User List)

    Lock (craftable item, ingredient)

    A lock would be a craftable item, made out of metal, and requiring plans to make. They could be made out of various metals, creating various tiers of lock (e.g., a copper lock is worse than a wrought iron lock, which is worse than a steel lock). A lock could be combined with a door to create a lockable door, or with a chest to create a lockable chest.

    Lockable Door(craftable, placeable block)

    A lockable door would have the normal door graphic, but with a lock on it. A lockable door, when first placed, remember the player who placed it (its owner). There are a few possible ways to go about dealing with the locks

    Option One:

    When first created, a Lockable Door is "unregistered". Using a key (craftable item) on the door registers that door, causing it to remember the key which was used on it. Keys should be renameable (via shift+right clicking while holding one and not pointing it at a door (to allow players to distinguish between keys more easily). From then on, the door can only be opened by right clicking on it with the appropriate key. Keys can be copied by placing one in the plans section of the anvil, and an ingot in the item-to-be-worked slot, and crafted just as a key is normally crafted. This will create a duplicate of the original key with a denotation that it is a duplicate (for example, if I have a key named "cevkiv's front door", a copy of the key will have an unchangeable name of ("copy of cevkiv's front door"). Duplicates can be given to other players and allow the player to open the door to which the key is linked to. Anyone can close a lockable door, however, without a key, they will not be able to exit.

    Option Two:

    When first created, a Lockable Door is "unregistered". Shift+Right clicking on the door will bring up a GUI that contains several dials (3 to 5, representing low to high tier locks), which form a combination lock. The first person to use the door enters a combination and then clicks the GUI button "Set". They are also remembered as the Owner of the door. The door now remembers the combination that unlocks it. To open the door later, one must shift+right click on the door, bringing up the GUI, and must enter the correct combination on the dials. Upon closing the dialog, assuming the correct combination was entered, the door will be able to be opened and closed. The combination entered will persist between opening and closing of the door, requiring someone to scramble the digits entered after closing the door to prevent unauthorized entry.

    Option Three:

    When first created, the person who initially places the Lockable Door is considered its Owner. The owner may Shift+Right Click on the door to bring up a GUI where they may enter the names of other people (with a limit on the number of names based on the quality of the lock used to make the door) who may freely use the door. Anyone else who attempts to use the door whose name is not on this list will not be able to open the door.

    Lockable doors themselves are reinforced by default to a number based upon their lock's quality, which can be set in the server configuration files. If a maximum of 6 reinforcements is set in the config files, I recommend a low tier lock cause the door to be reinforced twice (requiring it to be broken 4 times before actually being destroyed) while a highest tier lock would be reinforced to the maximum allowed amount by default. I suggest that the level of reinforcement provided a door be the same as the level of reinforcement provided to blocks. The door should not be, at its maximum, harder to break than the blocks around it.

    In all cases, when a lockable door is broken by anyone except the door's owner, it drops a regular door of the same material -- the lock is broken and not obtainable by the person breaking the door.

    I also suggest that doors must be made of the same material as their locks -- copper locks require copper doors, wrought iron locks require wrought iron doors, etc. Metal doors could be made following the normal pattern for a door, but with metal sheets instead of wood planks.

    Lockable Chest(craftable, placeable block)

    A lockable chest would have the normal graphic for a chest, but with a lock upon it. A lockable chest, when first placed, remembers the player who placed it (its owner). There are a few possible ways to go about dealing with the locks, just as there are with the locks on doors. See Lockable Door, above for the various methods I thought of.

    Lockable chests would be made by combining a chest and a lock and 7 extra wooden planks in the crafting cube, with the chest in the middle, the lock above the chest in the top middle, and the other squares forming a U shape of wooden planks. A lockable chest would be considered reinforced by default, to a level configurable in server configuration files.

    Furthermore, instead of creating a lockable chest with wood used to reinforce the chest, the wooden planks in the above example could be replaced with metal sheets. A metal Lockable Chest would have the same graphic as a lockable (wooden) chest, but color-shifted based on the material it is made of. Metal Lockable Chests should be reinforced to a higher degree than Wooden Ones, based upon the material they are made out of. I also suggest that lockable metal chests have a higher maximum level of reinforcement that other blocks, to make them worth while, and to make tools designed to pick locks on chests worth the investment. This will serve to further deter casual griefers, while an organized "raiding party" that spent the time and resources to come prepared would be rewarded for their extra effort.

    As an example, a lockable wooden chest would need to be broken 8 times, by default, to break it, while a copper one might need to be broken 32 times to actually break it, a wrought iron one 128 times to actually break it, and a steel one 512 times to actually break it. To those of you who say that that is far too many times, that's the point. Spending the considerable amount of time and resources required to be able to make a steel lockbox would assure that no one, short of a determined individual or group of individuals, who have devoted a considerable amount of time and effort, approaching or exceeding the amount of time and effort required to create the steel lockbox, would be able to open it in anything approaching a reasonable amount of time, via the use of special tools used to overcome lockboxes.

    Furthermore, just to clarify, a lockable chest, even when placed in an unprotected area should always be reinforced to whatever level it normally would be reinforced to, to give people who do not live in a town some protection against having their things stolen (and to those of you who say, "Just hide your things," I respond that a griefer is likely to use an X-Ray mod, and hiding your things does not help against an X-Ray mod.)

    Trading Post (craftable, placeable block)

    A trading post would be an item, created from a lockable chest, which functions as a lockable chest, but with an addition to its GUI. Upon placement of a Trading Post block, the block registers the placing player as its owner.

    The owner, upon opening the Trading Post, is presented with a GUI that contains the same amount of storage space as the lockable chest from which it was made (if higher tier chests contain more storage space), and two slots: One for the Item to be Traded Away, and one for the item the placing player was to receive in turn. The placing player then fills a certain number (but not all) of the storage slots within the Trading Post with the item he wants to sell.

    Any other player, upon opening the Trading Post, is presented with a GUI that displays the item and quantity of said item that they may buy at one time (for example, if the owner of the trading post put 8 charcoal in the Sellable Item slot, the Trading Post would inform the prospective buyer that they will receive 8 pieces of charcoal), and will also see what item is wanted by the seller (for example, if the owner placed 8 copper coins in the receiving slot, the prospective buyer would be informed that they need to pay 8 copper coins (or some other material. I suggest that if Bioxx wants to add coinage to the mod and have a use for it, Trading Posts be restricted to accepting coinage only as payment). There would be a slot into which the prospective buyer could place the payment, and, upon placing in the required payment, assuming there was still 8 coal (or whatever else was being sold) remaining in the Trading Post's inventory, the buyer would receive that item.

    I suggest Trading Posts be at least as reinforced as the kind of chest they are made from.

    For the Griefers

    No, I have not forgotten about you.

    Lockpicks (craftable item)

    Lockpicks would be a craftable item, requiring a metal ingot and plans. The final "durability" achieved in the crafting process determines the number of lockpicks received upon completion. Lockpicks would be in tiers, just as locks would be in tiers. Depending on the kind of lock scenario chosen from above, the lock picks would work in different fashions.

    Scenario One:

    If the kind of lock used is that that is attuned to a key, using a lockpick on that door or chest successfully with cause it to open. After it shuts, another attempt to pick it must

    be made to open it again.

    Scenario Two:

    If the kind of lock used is a combination lock, then, instead of a lockpick, some sort of stethoscope would be used. The quality of the stethoscope would determine how many digits of the combination would be revealed to the user, starting with the first digit of the combination. For example, a lowest tier stethoscope would reveal only the first digit of the combination lock, cutting the time required to guess the combination down to ten percent of what it would normally be. For the lowest tier of combination locks, this would mean having to try 100 combinations instead of 1000, and that would be a significant return on the investment in making the stethoscope. However, for the highest tier of combination locks, while you're still cutting the number of combinations to be guessed to 1/10th of normal, that would still leave you with 10000 combinations to guess -- this would reward the player who devoted the resources required to make such an excellent lock for his doors or chests.

    The highest tier of stethoscope should reveal X digits of the combination, where X is equal to the number of digits in the combination of the best combination lock - 2. For example, if the highest tier of combination lock has 5 digits in it, the highest tier of stethoscope should reveal 3 of those five digits. This would have the effect of a given tier of stethoscope used on the same tier of lock always having the same effect: giving 100 possible combinations to guess. So someone using the highest tier of Stethoscope on the highest tier of Locked Chest would (in this example) only have to guess 2 out of 5 digits of the lock. Furthermore, someone who has the highest tier of stethoscope would, in this example, be able to find the entire combination to the lowest tier of chest (as the highest tier stethoscope would reveal 3 digits of a combination, and the lowest tier of locked chest would only have 3 digits in its combination). This rewards people who spend the time and resources required to get the best possible tools for safe-cracking.

    Scenario Three:

    If the kind of lock used is one wherein the owner of the door or chest may specify users who are allowed to open the door or chest, use of lockpicks on the door or chest would cause the door to switch to its open state, or the locked chest to open. After closing the door or chest, another use of lockpicks would be required to reopen the door or access the chest again.

    In scenarios one and three, where actual lockpicks are used to pick a lock, I suggest two possible methods for lock picks functioning

    Lock Pick Functionality Scenario One:

    For a single ingot, up to 4 lock picks can be produced. The number of lock picks created is determined by the final durability achieved in producing

    the item on the anvil (half durability would only produce 2 lock picks, etc). Each use of a lock pick has a chance to consume that lockpick. A lock pick has a 50% chance of successfully opening a door or chest of the same tier (copper lockpick on copper lock, iron lockpick on iron lock, etc). A lockpick on a lock one tier lower would have a 75% chance of success (iron lockpick on copper lock), and a lockpick being used on a lock two tiers lower than it would have a 95% chance of success (since lock picking is a skill and is hard, 100% success rate should never be possible. Even experts fuck up sometimes.) Using a lockpick on a lock one tier above it has a 25% chance of success, while using a lockpick on a lock two tiers above it has only a 5% chance of success (or less. Personally, I think using the shittiest lockpick on the best lock should not have a chance of succeeding). If the lock is successfully picked, there is a chance, based on the tier of the lock pick, that the lockpick will not be expended (high tier lockpicks have a better chance of not being expended). The lowest tier of lockpick might be expended 50% of the time, while the highest tier might be expended only 10% of the time. On failure, the lock pick is always consumed (because you accidentally broke it when you screwed up).

    Lock Pick Functionality Scenario Two:

    For a single ingot, 1 lockpick is produced. Its final durability is determined by what you achieve in producing the item on the anvil. Higher tier lockpicks (more durable kinds of metal used) have a higher maximum durability. Each time the lockpick is used, it has a chance to fail or succeed, as determined above based on what tier of lockpick is used on what tier of lock. Upon a successful picking of the lock, a small amount of durability is expended. Upon failing to pick a lock, a large amount of durability is expended, with a greater loss the lower the tier of lockpick. For example, if a copper lockpick can be used successfully 10 times before breaking, you might need to fail only 3 or 4 times to break it, where as a steel lockpick which could be used 100 times before breaking might be able to withstand 30 to 50 failures before breaking).

    Scenario 1 would, I think, require more resource investment on the part of the griefer.


    I know gunpowder is currently in the game as a drop from creepers. Should this be removed as a drop, and gunpowder be craftable from saltpeter, sulfer, and charcoal (which are all in the game current, unless I am quite mistaken), then I think crafting this now much rarer and harder to obtain explosive should have a use in breaching walls and breaking open doors and safes.

    If gunpowder is removed from creeper drops and is craftable from items available in game, I recommend a (formless) recipe of 7 saltpeter, 1 sulfur, and 1 charcoal, as this as as close to the correct proportions as we can get with 9 crafting slots (although 6 salt peter, 1 sulfur, and 2 charcoal might be more "realistic" -- it's a matter of opinion, and I personally think that something as easy to obtain as charcoal shouldn't be a larger part of the formula than 1 part in 9).

    Gunpower could then be used to craft a crude explosive. This most crude explosive could have a "power" of 16 -- that is, when it is set off, every block within its blast radius that is protected would be considered to be broken sixteen times for the purposes of overcoming protection. This would allow someone to overcome thin, highly reinforced walls, or thick, low reinforced walls, very quickly, as a reward for expending the considerable amount of effort required to find these rare ingredients.The "power" should diminish for each block it has to go through. E.g., if the first block breaks from the 16 breaks the explosion causes, a block behind it should only receive 8 breaks, and if that one breaks, the one behind it should only receive 4, etc.

    Using an explosive against a door would have the same effect as on any other protected block, and, upon destroying the door, would leave no door behind -- you blew it to bits.

    Using an explosive against a locked chest would have the same effect as on any other protected block, except that there would be a chance for each object stored inside the chest to be destroyed due to the explosion when the chest breaks. This could be a flat chance for all objects (easiest to code), or each object could have a hardness value which makes it more resistant to explosions (a flint spear might only have a 5% chance of surviving, whereas a steel ingot might have a 90% chance of surviving, and a ruby or sapphire might have a 95% chance of surviving, and a diamond might have a 100% chance of survival, while paper or books would have a very low [1%] chance of survival.)

    Gunpowder should be made in a Mortar and Pestle (a 3x3 crafting grid used for the sole purpose of making gunpowder). Furthermore, this item would have additional functionality if, in the future, guns are added to the game, as it could be used, in combination with a bucket of water placed in the crafted grid along with raw gunpowder used to make explosives, to simulate the process of corning, which would render the "only good for bombs" quality gunpowder into something usable in a gun.

    Other suggestions

    Guard Dogs

    Guard Dogs would be created by applying a collar to a standard dog. Upon right clicking on a guard dog, they would follow you around. While in follow mode, right clicking on the guard dog again would cause it to enter guard mode. A guard dog only responds to clicks from its master. A guard dog will guard a 9x9 area, centered around itself. If in a protected area (where it can only be placed by a "citizen" of that protect area), the guard dog will attack anyone who enters its protected area who is not also a citizen of that town. A guard dog will have more HP, and deal significantly more damage, than a regular dog. An unarmed and unarmed person should have no hope of defeating a guard dog. A guard dog will also attack someone who attacks it from outside of its guard range (such as with a bow or a spear), provided they are within the 9x9 area it guards (assume they're on a leash and can't get out of that area). A pack of guard dogs should be able to take down even (lightly) armored opponents. If a guard dog is set to guard an area that is not inside a protected area, it will simply attack anyone who is not its owner while it is in guard mode. A guard dog who is attacking a player can be stopped by its owner right clicking on it (removing it from guard mode). A guard dog not in guard mode will not attack its attackers (to avoid cases of "Shit your dog is attacking me! "Here, let me right click on it!" "Shit, I hit it after you right clicked on it" "PLAYERGARY has been slain by a guard dog!")

    Guard Dogs would also make an audible noise (barking), when someone enters their guard area (who, if this is in a protected area, is not on the authorized user list). This should be able to be heard a good distance away (so if you're in town, and you hear a dog barking on the other side of town, you can go check it out).

    Since Guard Dogs would not require a Protected Area (town) to be usable and useful, I did not include them in that section.

    Guard Dogs would also need to be fed at regular intervals or they will starve. I suggest that the MINIMUM amount of beef dropped from a single adult cow be at least enough to sustain one guard dog for at least the amount of time it takes for an adult female cow to become pregnant, birth the calf, and then the calf grow into adulthood. At least if this were single player. Since there's a SMP focus, A guard dog should "need" to be fed once every 24 hours, and should "starve to death" after three (earth) days with no food, just so people who can't get on every day don't constantly lose their guard dogs.

    Mail via Passenger Pidgeon

    This idea would require crafted blocks and crafted items to work, and might work best if delayed until 1.3, which introduces writable books.

    Pigeon Stand (craftable, placeable block)

    A pigeon stand would be a craftable block (probably some sort of horizontal support on the bottom of the crafting cube with, I dunno, a chest on top and feathers on either side of the chest). When placed, right-clicking on the Pigeon Stand opens up a GUI. The GUI has an inventory of, say, 8 slots, where incoming pigeons are stored. It also has an Outgoing slot, capable of holding a single pidgeon. Placing a Pidgeon in the outgoing slot and then click "send" would cause the pidgeon to fly off to deliver its message. Upon arrival at the destination, it would be put into the "incoming pigeon" storage area.

    A pigeon can be used to retrieve the message it carries, which could then be read by a player.

    There are three options I can see for how to address letters:

    Option One:

    Each Pigeon Stand, when crafted, is given a Unique ID Number that is displayed in the Pigeon Stand's GUI. When sending a pigeon, the player inputs the Destination Pigeon Stand's number, and the pigeon will be sent to that stand. This is probably the simplest method.

    Option Two:

    Each Pigeon Stand is nameable when placed. When sending a pigeon, the player inputs the Destination Pigeon Stand's name ("Joe's Town", or "Gary's Summer Shack", etc), and the pigeon will be sent to that Stand. This would be slightly more difficult than Option One.

    Option Three:

    When a Pigeon Stand is placed, it registers its placer as it's owner. When sending a pigeon, the sending player inputs the name of the player to whom them are sending the pigeon, and it will arrive at their stand. Only the player who placed the stand can open it.

    For Options One and Two, I suggest that, if placed in a protected area, only citizens of that Protected Area be allowed to remove pigeons from the "incoming pigeons" area.

    In all cases, sending a pigeon takes time. The time is based on the distance between the two pigeons stands. I would say let it be two to three times as fast as a person can run, to represent the pigeon being much faster than the player. During this time, there would be a "cooldown" displayed on the sending pigeon stand, representing how long it will take for the pigeon to arrive, and disallowing further deliveries in that time. Furthermore, if the destination stand's receiving slots are full, any newly arriving pigeons, and their messages are lost. The sender will be unaware of this.

    A Pigeon Stand must have access to the sky, just as a forge.

    Passenger Pigeon (craftable, usable item)

    A Passenger Pigeon would be created from an egg (I know they're chicken eggs. I don't care at this point) in some kind of nest box. A passenger pigeon could be combined with a letter to create a "ready to send" Passenger Pigeon. This "ready to send" Passenger Pigeon would then be placed in the "Outgoing" slot of the Pigeon Stand and could be sent.

    Using a "Ready to Send" Passenger Pigeon would return the Passenger Pigeon to its default state, and recreate the letter that was originally attached to it, allowing it to be read.

    Letter (craftable, usable item)

    A letter would be craftable in a crafting cube using a piece of paper, an inc sac, and a feather. Upon picking up the letter from the output slot, a dialog would be created where one would be able to enter some amount of text (I suggest 128 characters, max). At the bottom of the dialog is a "finish" button, which finalizes the letter, after which it cannot be edited further. Prematurely exiting the dialog (via escape, or some other failure) wastes the letter (turns it back into a piece of paper). Using a Letter in the hotbar will display the contents of the letter.

    One option is that the final line of the letter could be "signed" by the player who wrote the letter. That is, their name is attached to the letter as a way of proving that it was written by them, to avoid people forging letters. I personally, would suggest the final line of all letters be the name of the player who wrote the letter.

    A letter would be combined with a Passenger Pigeon to create a "Ready to Send" Passenger Pigeon.

    Next Box (Craftable, placable block)

    I got nothing here, other than you'd need one to turn eggs into passenger pigeons.

    Currency (Craftable item)

    If I recall correctly, the graphics for coins are already in the resources for the mod. I suggest that a block be created to mint coins, probably requiring a die to be created on an anvil. One bar of copper, silver, electrum or gold would render 9 coins of the same material. Furthermore, 9 coins of the same material could be melted down to create 1 bar of said material. Coins should stack up to 64 (or 99, if that's possible). This would, in combination with the above suggestion of a Trading Post, if it were only able to accept coins in payment, would provide a more advanced form of economic exchange for larger, more advanced settlements, as they would be able to move away from a barter economy. Furthermore, it would allow the creation of banks -- A player could set up a building which contains Trading Posts which will exchange, say, 10 coins of a given kind for 1 coins of a better kind, or one coin of a given kind for 9 of the next lower value.

    Summation (tl;dr)

    In closing, the highlights are: The suggestions above provide a way for a player or group of players to "claim" an area of land to allow them to build defenses which provide security against griefers, without forcing any one particular model of government on the places, excepting for the fact that it would be a despotism. The suggestions also provide a way to keep your valuable safe from other players, while still making it possible, though difficult, for a griefer to steal them. The suggestions also provide a way for a determined individual, or a coordinated and resourceful group of individuals to bypass protections put in place. The Guard Dog suggestion gives a way to provide extra security for players when they are logged off, and for individuals who do not have the resources to establish a town of their own. The Mail via Pigeon system would be useful in allowing players to communicate with players who are not online without having to worry about the message getting passed through the grapevine, and, if letters are forcibly signed and "mailboxes" are only usable by the person who created them, allows for secure communications. The "Trading Post" provides a way of establishing shops, which allow players, or groups of players, to specialize in the production of certain goods which can then be sold to other players, and the option to make the only form of payment a form of coinage or other fiat (as decided by Bioxx) currency would form a standardized system of economic exchange for more advanced and "sophisticated" cities, which would simulate the move from a barter economy to a more advanced system.

    [edit] Fixed some spelling and grammatical errors. I'm sure I didn't get them all


  18. Of course not. They consider building adequate defenses to stop their stuff getting stolen as fun. Losing a game isn't fun, but playing is.

    It is impossible to build adequate defenses.

    Someone can simply build a dirt column and hop over your wall.

    Someone with a high enough tier pick can just go through your wall.

    The only way I could see this being remotely, remotely tolerable, is if when an item is taken from a protected chest by a user who is not on the protected listen, the item becomes flagged as stolen by the thief from the owner of the chest (that is, if JoeGriefer steals a diamond from cevkiv, the item's tooltip will now display, below "Diamond," "Stolen by JoeGriefer from cevkiv"). Furthermore, in addition to that, there would have to be some message that could be read by the person being stolen from identifying who did the thieving. A server-wide message would be the bare minimum, but that won't help much if no one else is on. Having the owned container be able to spit out the last X items that were removed from it and who removed them would be nice, except that a griefer could easily just fill the chest with dirt and remove the dirt until what items have been taken from it are out of the last X items taken.

    I like the idea of being able to claim an area for a person or group of people, and excluding others from being able to build inside that area.

    I like the idea of being able to "Reinforce" blocks within that area to make them more difficult for an individual to break.

    I like the idea of having tiers of lockboxes, either with keys, and locks that need to be picked (with various tiers of lockboxes and various tiers of picks, so that the highest tier of lockbox will always be the hardest to pick, while the lowest tier of lockboxes could be opened quickly with the highest tier of pick), or combination locks where a 3 to 5 or 7 digit combination must be entered, but there is a device (stethoscope or similar) that allows you to determine the first X digits of the combination). Lockboxes with keys and locks that must be picked require an expenditure of time in the form of gathering the resources necessary to pick the lock, and combination locks where parts of the combination can be guessed require an expenditure of time in the form of gathering the resources necessary to make the item that tells you what some of the digits of the combination are, and a further expenditure of time in the form of actually having to guess the rest of the combination.

    Griefing, if allowed, should not be something that can easily or quickly be accomplished against a settlement, no matter how small or crude.

    Also, to prevent people from just tunneling under walls, and to prevent the people of the town from having to reinforce every god damn block of dirt, if there is a placeable "Hearth Stone," or similar, then all dirt/gravel blocks below its level within its effective range should be automatically highly reinforced. Because if they're not someone can just go under the wall to get inside. And there needs to be a "no build zone" around the claimed area, or people need to leave a buffer between the end of the claimed area and their walls, to avoid people just pilling up dirt and getting over the wall.

    As far as trading goes, some sort of craftable block, similar to one in Industrialcraft2, whose name I can't remember, I think it's Trade-o-mat, would be nice. The way the block works is you place the block, and then you place a chest (lockable, or not) next to it. The GUI for the trade-o-mat would have two slots: One for the item you're willing to sell, and another for the item you will accept in exchange (if you want to enforce a currency, this slot could be restricted to using coins only). A player other than the player who placed the Trade-o-mat, when accessing it, would be presented with a GUI that would tell them what is being exchanged for what, and would have an input slot where they could deposit the item that the owner wants, and an output slot that would spit out the item that the buyer wants. The trade-o-mat could either have an internal storage capacity, or could place/retrieve items to be sold/items that have been bought into an adjacent chest. This would allow people to be able to buy and sell goods even when offline, and could be used to enforce a specific currency by accepting only coins as payment.


  19. Ultimately, the way to completely solve this is to give sever admins the ability to just outright disable people from lock picking, but to out right have it disabled or not even an option I think would be a worse option and would stifle a lot of creativity and game play options.

    Read: It will make assholes have a harder time being assholes, and I, being an asshole, don't find that fun.


  20. That's the fun part and usually where war comes into place if the server can't hold itself together.

    You're failing to understand this: This is not what I consider fun. This is not what a lot of people consider fun. I dare say that the majority of people, if asked, would not say they consider their things being stolen to be fun.