I plan on breaking the golden rule of MMORPG''s and implementing a true death system, basically you play the game in a "school" environment, limited to an assortment of introductory adventures, death is a temporary problem and will simply result in a slight loss of time(while you recover your equipment) and a mild XP hit.(generally slowing your progress into the "real" world.) Once level 6(or perhaps another subjective number) is hit you get kicked out of your apprenticeship and now if your character gets killed then he\she is dead, roll up a new one. Of course there are potions and magicks that can resurrect your player, and at super high levels you are almost certain to have friends or perhaps guild members that will gladly bring you back. But on the darker side you will have anhilliation spells and certain scenarios(persona duels) that will render you dead without the chance of recovery. I''m sure a ton of people will pee on the idea so i won''t list the obvious cons. Some of the pros i believe that this system will deliver. Automatically balance the PK issue: If becoming a fugitive for murder might result in being killed or apprehended and permanently losing the ability to play that character then PK''ers will be either very smart about the way they PK or else very short lived. Death means something: If you give your characters life to hold a patch of land in a huge guild battle or you get smoked while trying to save a village from the marauding orcs it is an actual sacrifice, not an inconvenience. Taking a battle personally: If you are attacked by another player and your prize character becomes permanently Ghandi, you might conceivably chase that guy down with your next five characters seeking revenge. Would make for a bunch of great character related content.(sans code) "My name is Inigo Montoya, you killed my father, prepare to die!" lol Developing groups, guilds and companies: There is strength in numbers, and this will become very obvious when you actually have something riding on the line when you enter into battle. Not to mention the advantage of having allies that can heal and potentially raise you from the dead if everything goes wrong. Good and Evil: Alignment is generally just a tag added to your name in most RPG''s. Now evil will have it''s own areas of dominance(i.e. the Pirates cove, where anything goes and the law never shows.) as well as good having it''s own dominance. (i.e. the mighty Seven, who seek down and destroy murderers thieves and evil mages. The healer as a character class: Even though raise dead spells will be limited to all but the most powerful of characters, the ability to heal others and yourself in combat will become tantamount to victory. Player generated content: Lets say that a popular and powerful member of a guild is slain while trying to loot a dragon. Unfortunately noone can find the body. Now a huge quest is undertaken by the guild to find the "Brazier of Ressurection" a powerful artifact that can total rebuild and ressurect the fallen comrade. OK, enough rambling, i''m sure some of you hate the idea, and some see a bit of merit, any comments are welcomed even if you''re just telling me what a horrible idea it is. I''m mostly looking for additional twists or some slant i have not considered. thx Dredd Dreddnafious Maelstrom "If I have seen further, it was by standing on the shoulders of Giants" Sir Isaac Newton

I agree, although I think perma death should be enforced from the start of the character. Rather than letting them come back start them in an area safe from PK''s and the really dangerous things until they are confortable with the game. Not neccisarily a place that PKs couldn''t enter but someplace with NPC guards. They kill a newbie and could very well lose the character that they have spent time to build. I''m kinda torn on the resurection idea, mainly because I know it will be abused.

I don''t mean to criticize, but this isn''t really anything new as far as this forum, or indeed online games go: I suggest you browse through this forum for mentions of ''Permadeath'' and you should come across numerous discussions I seem to recall Silvermyst is at least one person in favour of such a system, for example. But you will also find that a lot of your proposed benefits may not necessarily be as successful as you may hope for.

[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost | Asking Questions | Organising code files | My stuff ]

yeah, i remember a post similar to this a few months ago, just trying to kick start the ol'' memory.

Dreddnafious Maelstrom

"If I have seen further, it was by standing on the shoulders of Giants"

Sir Isaac Newton

Actually, the permanent death idea is something I''ve been toying with for a few years now. My concept is similar, but clearly a system of checks and balances needs to be in place for such a system to ever work and be entertaining (as a game should be, right?).

My theory is that death should be a possible consequence of being a loner or nomad within the game.

To avoid permanent death, a benchmark system is awarded to those who form collective groups as a society. The higher the group or tech level, the higher the possible benchmark for each member. When a character dies, they suffer a consequence of loss, but never below the benchmark so long as they are within a thriving society centered around some place or physical item that is capable of being destroyed.

On the other hand, the nomads/loners/what-have-you have the added benefit of advancing much more quickly than those who are in groups. This adds an incentive to be on both sides of the fence, so to speak.

This lends itself to war-like scenarios within an MMORPG setting, where rival clans seek to destroy one another, or rogue warriors wreak havoc on society.

Of course my true design is far more in-depth than this, but I figured I could put out a little something to spark some creativity. Any thoughts?

Shadowin

Another convert.

Number one permanent death concern: technical glitches.

How do you plan to deal with a player losing his character after tens of hours of conservative play, due to a system crash, or a disconnect?

Another convert, lol...

Great point Silvermyst, let me think about this...

from the hip, without enough research done to qualify it as an acceptable way to go, I would say you could profile the connection parameters like a packet sniffer.(or a network sniffer, whatever you want to call it) Most connection losses are prefaced by "hiccups" in the packet delivery and the gap between sequential packets generally stretch just before a legitimate connection loss. (not fail safe but documentable)

The trick would be to write an algo that could qualify the nature of the connection loss, and resolve it to either an intentional disconnect or a series of packet loss culminating in a total disconnect. Of course, in the client you would code a tell all, last gasp message to the server that reported when the client was manually closed, negating the viability of turning off the game because you were getting smoked. which would leave you with the option of physically disconnecting the media in order to try to cheat the system.

Basically we could profile the nature of the packet loss and see if we could spot a pattern consistent with the various styles of disconnection, and then decide which styles of disconnection merit putting the player in an invulnerable stasis, while we save his state, location, etc...

Once again, this is an untested idea, but are my uneducated plans going in. I am just now getting to the point where i am writing my network code so i have plenty left to discover. The packet sniffing idea is based on my real world experience of supporting my employers network, and profiling bottlenecks in the network traffic, and monitoring the latency between our local loop and our server.

I should have a full time server running in the next 30 days, and i am going to distribute 3 clients to a few friends of mine and let them play with what i have so far, so i should be a more qualified poster with a bit more experience here in the next month or two.

@ shadowin77

Sounds similar to my idea but our implementation is a bit different. I have been working on the alignment metric for a bit.(you know, good and evil, law and chaos) My intentions are to allow evil and chaotic players to gain a small experience bonus over the lawful and good players for the same actions, this will be a gradiated scale based on an alignment number(undisclosed to the player) On the good side, you have the distinct advantage of being able to trust your companions. so while the evil side will be quicker to rise to power, the good side should have a true advantage in teamwork and consistency. (not to mention that one betrayal could permanently cost your character his life, so if you keep the company of evil, you really never know if anyone has your back.)

i think that players will group together out of instinct and intelligent play. Letting the law of attrition weed out those that lone wolf it too much.( or else they will quickly gain power and become able to fend for themselves.)

Grief players, as discussed in another post, might be the target of an assassins arrow. (hired by their previous victim perhaps)
and would then be resigned to creating another player, choosing a class, and completing yet another apprenticeship before they were allowed back into the "real" world.

Oh yeah, just a final point, with PermaDeath a GM has a very simple way of dealing with grief players. He can simply repeal any karma or alignment punishment for killing a particular player. Thereby giving the entire world a free license to kill and loot that player at will for a specific period of time, without retribution. This handles the administration of grief players in-game, by simply making grief players "persona non grata" and letting the local thugs have a field day.

@ tstrimp

My reasoning for the safe zone and apprenticeship is that my battle system is a combination of complex, skill based moves(like a first person shooter) and mathematical probabilities(like a typical RPG) so in order to play the fighter decently you have to master the movement and attack patterns, in addition to simply leveling up, and having good stats. you would be able to "test out" of your apprenticeship if you can pass the graduation test at a lower level, but then you would be a lower level dude in a dangerous world. My wizard class. (which i haven't even begun yet, will be composed of a small library of "known" spells, and will be fortified by original spells that you script in your own "wizard client" (after hunting down the necessary components of course.) I've been told that this idea isn't viable because only programmers would be able to script their own spells, but if that is the case then good, wizards will be rare and enigmatic, just like i want it. (not to mention if you kill one and take his spell book you will now be adding a bunch of unique spells to your own library)

As for the resurrection being mistreated, i plan on making the necessary components and side effects on the caster vicious enough so that it will be a true act of selflessness to cast such a spell.

Anyway, i am ultimately just another donk trying to write an MMORPG. But this has been a good exchange, feel free to keep the comments coming!


Dreddnafious Maelstrom

"If I have seen further, it was by standing on the shoulders of Giants"

Sir Isaac Newton

[edited by - Dreddnafious Maelstrom on December 5, 2002 12:30:16 PM]

Okay my position: Neither for or against as long as the game is fun.

Now the problem with your solution to Silvermyst's problem. (Ex. Two players are fighting each other, one sees he's about to die and kills his connect before he's killed in game.) See what I'm getting at? The sniffer will see this as a d/c not the kill it should have been. I've played pvps who did this and it was very frustrating. It also allowed us to d/c while fighting an extremely tough monster and not lose the items/experience. Then they were forced to implement d/c = kill. The sniffer will have a hard determining whether it was a valid d/c or not. If you can think of a better way around this please voice it. I'm interested in hearing a solution.

(Edit: typed before I read your entire post. How would the server differentiate between a physical line going down and me pulling the plug out the computer? People will do anything if they have a lot of time and energy invested.

[edited by - TechnoHydra on December 5, 2002 1:42:53 PM]

quote:

---

The packet sniffing idea is based on my real world experience of supporting my employers network, and profiling bottlenecks in the network traffic, and monitoring the latency between our local loop and our server.

---

A system crash that results in a disconnect would be analyzed as an intentional disconnect by this type of packet sniffing, right? (if it''s successful at determining a difference between packet loss disconnect and intentional disconnect)

My personal solution is to allow players to construct customized auto-behavior patterns for their avatars. You wouldn''t have to bother to check why a player gets disconnected (packet loss or intentional) but only IF a player gets disconnected.
Player gets disconnected -> triggers pre-constructed auto-behavior cycle for situation ''disconnect'' -> avatar runs away as fast as possible OR avatar stands and fights till death OR avatar stands and fights until 50% health, then runs away as fast as possible (etc)

Also as far as the disconnect issue goes, there is currently a game in development called Atriarch, or somthing like that, which is a truely persistant online rpg. Your player never leaves the game. When you log out you can choose to have you character practice skills while you are gone. In order to keep you character safe you''d need to invest in some body guards etc. But no one will be able to d/c to avoid death. I really like this system, mostly because the casual players wont be left quite as far behind as the hardcore ones.