Mistakes You dont want to see Repeated

[Seytra]

Originally posted by RonHiler
A straight checksum won\'t work.  Checksums are quite trivial to fool.  You could maybe use an MD5 fingerprint perhaps.  Even so, whats to prevent the client from intercepting the sent fingerprint, writing it down, changing the level to their liking, then modify the client to return the original fingerprint (independent of the level contents)?  This is an open source game, after all, and you just gave a huge amount of trust to the client.  From a security standpoint, I don\'t think your idea would work.

I applied the term \"checksum\" very broadly, indeed. I was, however, thinking along the lines of MD5 or RSA fingerprints.

However, my description seems to have been misleading. I meant to say that the system should work like this:

1) Player exits the area (by crash / logout / moving into other area / whatever), server does:
- compress the item info
- checksum it
- store the checksum in the player\'s account (SERVERSIDE)
- send the compressed data to the client (without the checksum)

2) player reenters the area, server does:
- retrieve compressed data from client
- checksums it
- compares the resulting checksum to the one stored in the player\'s account (SERVERSIDE)
- accept or reject the data

Therefore, the client does not even need to know the checksum that was generated, and it does not matter one bit (literally *g*) whether or not it does, as the server checksums whatever junk it gets off the client.

Originally posted by RonHiler
Second problem with it is that what if I play the same toon from two different computers (e.g. home and work)?  The levels would be stored on one and not the other.  Kind of a problem.

Indeed this is a problem. However, I doubt that many players are going to use different computers, and if they do, they are likely to have networked them anyway. Therefore, the client needs only to support shared and freely relocatable data storage areas (i.e. folders), independant on the installed game\'s location, for the player data and most would be set by simply sharing the folder accross the network. Like mount a samba share to drive P:\\ in Windoze, and /mnt/p/ in Linux. The clients would be set to the appropriate paths once and the account is fine. Also, the path should be easily changeable in case the mountpoint has moved or another folder is to be accessed (like on a removable media for playing at your friends computer).
Therefore, the inconvenience would be mitigated, and if this method OTOH eliminates griefing, harvesting, level-grind and hunting zones, it may well be a good tradeoff.

Originally posted by RonHiler
I dunno, Seytra.  You just described, almost perfectly, the EXACT system in use in SWG.  Right down to the \"greater variation in time\" idea.  If I sit in the med center healing toons, at first I have very little capability (I use MedPack A\'s).  After doing this for quite a while, I gain more ability, eventually being able to create better medpacks and heal all sorts of different afflictions (bleeding, battle fatigue, etc.).  But what it comes down to is spending hour upon hour sitting in the med center.  Yeah, I could go do other things (like work on my dancing or mininig or artisan or even combat skills), but if you want to be good at your chosen profession, you have to use it for many hours/days/weeks.

Now, whether that works or not is a bit of a personal judgement, I suppose (after all they DO use it to this day, and people ARE playing), but the general consensus with SWG in the press was that it was a boring game, for exactly that reason.

Don\'t get me wrong, the *idea* is okay, but the implementation (at least in SWG) is flawed.  And if PS is using the exact same system, it will have the exact same flaw.

Well, this is only because one fundamental idea hasn\'t made it into MMORPGs yet: teachers. If there are teachers or even schools, you could invest money and gain (nearly) instant skill advance. This way will enable the players to choose whether they wish to quickly gain new skills / improve existing ones or do it the \"learning by doing\" way, possibly earning money as they go, spending their money on other things. Also, you yourself could teach others if your own skill is considerably higher than theirs and you got the \"teaching\" skill high enough. After all, where would we be without teaching?

Originally posted by RonHiler
Just out of curiosity, where will your equipment come from?  You say critters will only drop critter type items.  From where will come the swords and armor and wands and things usual to fantasy based MMORPGs?  Merchants?  Do the items just appear at the merchants (ala AC et al) or must they be made by players to be sold (ala SWG)?  If the latter, where do the raw materials for crafting come from?  Merchants again?  Do you have to go mine/farm/harvest them (you do mention the farming/mining areas, so am I to take it that the game\'s economy is entirely player determined?)

Ron

AFAIK, it hasn\'t been decided yet. What I believe to be the current state of ideas, there will be NSCs to
- get things started and to
- help out in player-unpopulated economic nieches.
- stabilise prices

However, if there are no NSCs, the demand will eventually rise so high that players _will_ consider doing thetask, as the reward is big money. I expect that many ppl. will not mind, because in other MMORPGs there are all these level-grinders and hunters anyway, who in essence do the same thing over and over again to gain XP or money, so there really isn\'t much difference. Still, I do not know if this is such a good idea because it might well make it impossible for newbs to get basic stuff (because the prices are too high).

But I do not know of any definite statements yet.

Regards,

Seytra

[RonHiler]

Originally posted by Seytra
Therefore, the client does not even need to know the checksum that was generated, and it does not matter one bit (literally *g*) whether or not it does, as the server checksums whatever junk it gets off the client.

Ah, good point.  Okay, you\'re right, that\'s much better.  Dunno why I was thinking the \"checksum\" would be client end

Indeed this is a problem. However, I doubt that many players are going to use different computers, and if they do, they are likely to have networked them anyway.

Ug. No, I disagree with this STRONGLY. I consider that a completely invalid assumption. When coding, you must code for solutions for every reasonable case, or it\'s not a solution.  It is not unreasonable to assume some players will use multiple non-networked computers to run PS (it may not be a large segment of the population, but it will exist).  I might play from home and from work, for instance.  That\'s certanly not an unreasonable scenario.  Or perhaps I\'ve upgraded to a new computer.

And therefore your \"save to client\" approach has a problem.  Perhaps you could require transfer of a particular directory via floppy whenever a player moves computers, but that\'s not a solution I would use myself, too many potential issues (like what happens if they forget one day?  Now you have synch issues).

Another issue I might bring up with your scenario is the unexpected disconnect.  What happens if the player is dropped from the server due to no response?  In such a case, the server *couldn\'t* send the compressed data to the client (if it could, it wouldn\'t be dropping the client in the first place).  The only thing you could do to ward against this is send periodic compressed level data to the player as they are playing.  But then you are running into bandwidth issues again.

Well, this is only because one fundamental idea hasn\'t made it into MMORPGs yet: teachers. If there are teachers or even schools, you could invest money and gain (nearly) instant skill advance. This way will enable the players to choose whether they wish to quickly gain new skills / improve existing ones or do it the \"learning by doing\" way, possibly earning money as they go, spending their money on other things. Also, you yourself could teach others if your own skill is considerably higher than theirs and you got the \"teaching\" skill high enough. After all, where would we be without teaching?

Perhaps, although I\'m skeptical.  Sounds like you\'d be giving a HUGE advantage to guild members (who could essentially insta-level from newbie to uber by learing from thier guildmates) and leave soloers out in the cold.  I guess there are ways of getting around this, but it seems pretty problem prone in terms of game balance to me.

BTW, SWG also has teaching (although you must earn the \"xp\" by doing the task until you have enough to \"level\" the skill, then you can be taught (by another player) to get the actual advance).  So it\'s not a COMPLETLEY new concept

AFAIK, it hasn\'t been decided yet. What I believe to be the current state of ideas, there will be NSCs to
- get things started and to
- help out in player-unpopulated economic nieches.
- stabilise prices

Ah, okay.  I was just curious, thanks for the answer.  It\'s an area I\'m struggling with for my own game as well, and I wanted to see how PS was handling it.  It\'s a very tricky area.  On the one hand, you want equipment to be reasonably easy for toons to get, at least for the common stuff (if they have the cash) and prices to stay reasonably stable, but on the other, if you go overboard with the merchants selling pre-genned euipment, you put your tradeskill professions out of business.  

It\'s a fairly tough thing to balance.  I\'m not sure I\'ve seen a game YET that has gotten it entirely right, and I\'ve seen quite a few of them, heh.

Ron

[Murais]

ok, a mistake that I will never ever want to see in PlaneShift is an impossibly dificult, completely unbalanced leveling system. I used to play FFXI, and let me tell you, I loved the game at first, absolutely loved it!
But then it came time to level up...

    All the fun jobs had to have a quest completed for them, and all the quests were impossible until about lvl 30. At first I figured \"Hey, it\'s only level 30, how long could that take?\", let me tell you, I was not impressed. There is nothing more annoying than sitting on a spawn point, waiting for the same damned monster to appear. That, is the most annoying. Then you spend hours, and hours, and hours, only to find out you only went up 2 levels. Now this would be ok if you were like lvl 25. I played that game for 2 months, and only got to lvl 11. Now that, is ridiculous.

   And primly, there were next to no quests suitable for the lower levels. That just makes me angry.

     So, pretty much I am saying is, DO NOT FALL INTO THE MMORPG DEATH TRAP!!! That is pretty much do not make the game inhospitibol to beginners, make it interesting, make many other things to do other than chatting and leveling. Make it interesting, make it fun. Make it unique

[DrunkenPimp]

I was playing knight online wh/ile waiting for this game. On websites it said it was free. They trick u into playing for a month, then a little notice comes up that says that the game really isnt free andif u want to keep playing  u have to pay.

i think its terrible that they pretended it was free then waited till u were hooked and charged u. i hope planeshift doenst try to trick us into paying.

[RonHiler]

Originally posted by Murais
ok, a mistake that I will never ever want to see in PlaneShift is an impossibly dificult, completely unbalanced leveling system. I used to play FFXI, and let me tell you, I loved the game at first, absolutely loved it!
But then it came time to level up...[...]

Well, keep in mind, game like that are meant to be played for months/years at a time.  They\'ve scaled the leveling system to account for that.

On the other hand, you are absolutely right.  If they are going to force you to level so slowly, you have to have content appropriate to every level beyond mindless hacking.  Sounds like they need to add in some lower level quests.

[Murais]

exactly what I was thinking, all I am saying is I would hate to see that mistake here...

[Seytra]

Originally posted by RonHiler

Originally posted by Seytra
Indeed this is a problem. However, I doubt that many players are going to use different computers, and if they do, they are likely to have networked them anyway.

Ug. No, I disagree with this STRONGLY. I consider that a completely invalid assumption. When coding, you must code for solutions for every reasonable case, or it\'s not a solution.  It is not unreasonable to assume some players will use multiple non-networked computers to run PS (it may not be a large segment of the population, but it will exist).  I might play from home and from work, for instance.  That\'s certanly not an unreasonable scenario.  Or perhaps I\'ve upgraded to a new computer.

And therefore your \"save to client\" approach has a problem.  Perhaps you could require transfer of a particular directory via floppy whenever a player moves computers, but that\'s not a solution I would use myself, too many potential issues (like what happens if they forget one day?  Now you have synch issues).

OK, the \"play at home and at work\" scenario is a problem that I didn\'t think about, true. I\'d also hate the floppy method for various reasons. We\'d need to auto-synch here. One option would be to store the _current quests_ data serverside so that this issue will most likely never come up. However, this can involve too much disk usage as well. OTOH, maybe we could make the players provide some sort of webspace for themselves? They could then have the server or client upload the data to a web / ftp server automatically and then retrieve it. Of course, this implies that everybody has access to webspace of sufficient size.

We _could_, of course, add the option to not store anything for these unlucky people, but I\'d really like to see a way to get around this.

Under no circumstances must the server use old data, because you could deliberately desynchronise to duplicate items (like take item from ground, quit, reload old level with item still on ground, pick item up again, etc...). The server must only accept the most recent checksum for each level. If the level data is lost, then the server would revert to an empty level, after asking the player for confirmation.

However, except for the _current_ quest, there is no real need to go back to the previously completed ones, it\'s just nice (and you\'d have safe item storage there), so if the current quest would be kept on the server, there would not be much inconvenience for the multiclient player.

Maybe if the current quest would be stored on the server, nice things like holes in walls, blackened walls, whatever, would not be stored, so I\'d pick client-side storage if I can help it.

Maybe we could make this an option rather than the norm? Modem users might benefit from this, as well.

Originally posted by RonHiler
Another issue I might bring up with your scenario is the unexpected disconnect.  What happens if the player is dropped from the server due to no response?  In such a case, the server *couldn\'t* send the compressed data to the client (if it could, it wouldn\'t be dropping the client in the first place).  The only thing you could do to ward against this is send periodic compressed level data to the player as they are playing.  But then you are running into bandwidth issues again.

Well, I wouldn\'t count that as a problem, because the data that must be kept will in this case be exactly one area, nothing more. This could IMO be tolerated to compress and store serverside until the client reconnects. Also, while the world data is loaded down to the client when it enters a new area (not a new quest), the old data could be downloaded along with it, reducing this problem even further. Of course, if you create numerous chars, go to a level, drop millions of items there and then crash the clients, the server would be out of memory, but this is tedious so it\'ll not happen since plain old DDOS is easier to achieve (and less traceable).

Originally posted by RonHiler
Perhaps, although I\'m skeptical.  Sounds like you\'d be giving a HUGE advantage to guild members (who could essentially insta-level from newbie to uber by learing from thier guildmates) and leave soloers out in the cold.  I guess there are ways of getting around this, but it seems pretty problem prone in terms of game balance to me.

BTW, SWG also has teaching (although you must earn the \"xp\" by doing the task until you have enough to \"level\" the skill, then you can be taught (by another player) to get the actual advance).  So it\'s not a COMPLETLEY new concept

OK, if the guilds would provide internal teaching for free, this would be unbalanced. We\'d maybe need a teachers guild, or the requirement that the teacher is paid regardless of the standing of the pupils. OTOH, it might suffice to allow free teaching only for guild matters, and also to put a cap on the amount that can be learnt in one session. This way, even guild members would need to actualy _use_ the skills they just learnt for some time until any teacher will re-admit them for the same area of teaching. Not entirely realistic, but better than to make teaching take days or weeks real-time, during which you\'d not be able to use the char. You could, however, allow teaching only once per week, artificially slowing down the process.

[Monketh]

If I read what you posted correctly, then I have one thing to say.
Do NOT, under any circumstances, make players responsible for their character data.  This can only be bad.

[Raiben]

Something I think should is completely stupid is if random monsters carry gold and weapons and stuff.  Seriously why would a wolf or something have an axe and three hundred and four gold?  Did it eat it or something?  I mean sure its alright if the enemy is a thief or something.  I think monsters should carry materials like fur or something then you can make it into stuff.  That would be better...  Harder, but better...

[Seytra]

Originally posted by Monketh
If I read what you posted correctly, then I have one thing to say.
Do NOT, under any circumstances, make players responsible for their character data.  This can only be bad.

Heh, \"the user is an idiot\", eh? Well, that is true, and if only by accident.
I was, however, not talking about the character data (which would still be stored serverside), but only about so-called \"quest spaces\" (instances of quests on a per-player basis), and how to act on them once the quest is solved.
This is meaningful if the player at one time decides to revisit one of the places _after_ having solved the quest. The effect would be that the area is exactly like when it left, while commonly it\'s just re-generated by the server, if it\'s not shared among all users.

Therefore, the loss of this data would only hurt the immersion a little, the character itself is untouched (if the player stored valuable items there, however, these will then be lost forever).

[tygerwilde]

ok, a mistake I\'d really love to see eliminated is dev apathy...
this is a problem in most MMOs I\'ve played
somewhere along the developement process, devs seem to lose interest in the experience the players are having in their game. they don\'t care that there are a hundred people out there looking for an actual role play experience that can\'t stand people talking like 7]-[15. all they begin to care about is processing the content.

they need to monitor the game, make sure that if nothing else, people aren\'t acting like total n00bs in game. they need to make sure people are obeying naming conventions.

they also need to have some kind of connection to the gamers in general

care about the game, not just the programming

[Seytra]

This isn\'t the devs job, it\'s the GMs job (unless the two are the same ppl.). However, the devs obviously still need to know about the game as it\'s being played to actually bring in new content that is desirable. The problem is that most MMORPGs are profit-driven. Therefore, every player that is lost is money lost. Therefore, if the GMs go around removing accounts of ppl. who destroy the game\'s RPG aspect actually loose the game makers money. Therefore, they are told to stop doing this somewhere along the lines, when the _management_ stops seeing the game as a place to have a good RP experience and start seeing it as money-generating-machine, with TCO, capital investment, risks, investor incentive, value-added services, product, ammortisation, value return and all these other marketroid and finance department bullsh$t that _really_ messes up _everything in our world_.
Of course they need to finance the freaking game servers and bandwidth, but once they start to \"maximise revenues (or profit)\", the players are starting to become screwed with no end in sight. Most marketroids probably don\'t even know what RPG is (they probably think it\'s these training programmes for customer relations and team management that companies increasingly have employees take (I\'m not saying these are a bad thing)).
That is a fate that I wish not to see PS facing. Actually PS has better chances to prevent this because it started free and open, therefore there is no marketing/financial department that decides what to do.

Oh, of course, GMs can be overwhelmed by the sheer number of players who act n00bish, so it might be apathy following resignation. But PS has better chances of dealing with this, too, because it\'s open: as more ppl. join, so will aspiring GMs. They don\'t need to be hired (and then kicked out to \"maximise profit\").

[RonHiler]

Originally posted by tygerwilde
ok, a mistake I\'d really love to see eliminated is dev apathy...

Don\'t confuse dev apathy with publisher apathy.  They\'re two different things.  Devs put theiir blood, sweat, and tears into a game for several years of their lives.  They are intimately connected to the game and they believe in it (or it would never get finished).  You won\'t ever convince me that a dev doesn\'t care about his/her game (unless they got sick of it and released it early, and such releases are VERY obvoius and infamous (BC3KAD, Outpost, et al)).

The *publishers* are the ones you most likely have a problems with.  They push games out the door before they\'re ready, they handle customer support (or lack thereof), and very often, they make the play-affecting decisions, usually bad ones.

For instance, one of my very favorite games of this genre, AC, was developed by Turbine and published by Microsoft.  AC had (for a very long time) a severe playbalance issues with one of their systems (monarchies and xp chains).  While the players abusing the system weren\'t technically breaking any rules, they were also not using it in the spirit it was intended.  Turbine wanted to fix the problem, but was forbidden to do so by MS (probably due to fear that such a change would cost subscribers, as you say Seytra).  What MS failed to understand was that the problem unbalanced the xp portion of the game, and thus you had toons levelling at an insane rate.  As a result, content was increasingly created only for the higher ups, leaving the people who played as intended out in the cold (and as a result, they lost subscribers anyway).  At one point that sticks out in my mind, it got so bad that a formerly decent level toon (which is to say levels 30-40) couldn\'t even leave the towns without getting immediately mobbed and dropped by critters.

Turbine recently bought out MS, and I don\'t know if they did or intend to in the future finally fix the issues, but it\'s probably way too late now, the damage has been done.

Anyway, the point being don\'t blame the developers, it\'s almost certainly not their fault if bad design decisions are make due to a \"bottom line\".

Fortunately, PS won\'t have this problem.  Neither will my game

Ron

[rifft]

I\'m new to this, so I may not understand much, I\'ve been reading the boards and haven\'t posted much, in fact I think this is my first post, so I might as well make it worth while eh?

Anyway, onto my point. I\'m not entirely sure that having *quest space* is desirable in a persistent world. Though I do think that making quests unique is desirable. I think that one of the most common mistakes devs make is game *balancing* before I get flamed, let me explain what I mean. By balancing I mean introducing unnecessary complexity into the rules in order to handle all special cases. I think that in most cases this unbalances the game. As an example take some small portion of an economy, in which all items are generated by players.  If you only have on weapon smith, and he has made only one sword. Then we have a problem, there is no set of rules that will fix this situation. Now if we make it even more extreme, for example that weaponsmith has only made one sword and refused to make any other weapons. Again we have an economy that\'s completely messed. So what can we do? Force the weaponsmith to produce X amount of swords before being able to sell any? I don\'t think that\'s a viable solution. But things of this nature exist, not the particular example I pulled out of my behind, but in general things like this occur in all RPGs (I mean both MMO, Single, and Pen and Paper) and in most cases it\'s dumb. Those rules get thrown out and player find other ways to do things. So how would one balance the game? I think common sense is a good guide combined with past experience. I think every description of an MMO, or any RPG for that matter assures the player that it isn\'t a hack and slash game, it has no leveling, etc. And in most cases it\'s a blatant lie. Don\'t get me wrong I\'m not flaming PS in any way. I\'m just saying that everyone thinks that they aren\'t going to fall into the same trap. Anyway I think I lost track of what I was talking about. So back to game balancing. Common sense and past experience. For example, teaching, in real life it just means that the average skill level of people is increased overall, but that can\'t happen in an MMO (or can it? should it?). Anyway, you have to come up with some minimal set of rules and let the players balance it out.

Though really I am speaking out of my behind as I have only been a concept artist on an MMORPG.

Anyway, I just wanted to make some comments. I may be entirely off base. But I would really love it if the game did not devolve into doing some small activity for hours on end in order to do anything else. Oh and please, oh please don\'t let the devs  implement anything in the same way FF XI did. Having nothing in common with that game is a good thing!

Back to the *quest space,* it may not be possible to implement quest space, since it may require a significant redesign of how world data is stored. Also, there might be quests which occur in the general game world (i.e. wilderness) where it is impossible to generate a single instance of a location. I think in general having persistent locations is a good thing, and the biggest problem with current MMO\'s as far as I understand is the DROP/XP lineup.  And simply restricting the quest to one person won\'t solve that problem, I don\'t think so anyway, or it wouldn\'t solve it any better than allowing the player to complete the quest only once and therefore get the item only once.

Damn, my brain just liquefied.

Oh and as for the lack of numbers. I think that\'s the most brilliant idea I have ever heard. Picture this: In your character sheet all that\'s listed are your skills, and how much you\'ve improved since last time you gained experience using them. So for example instead of showing how good you are overall, you simply know how much you improved since last time you used that skill. So every time you use that skill a new number, or perhaps icon of some sort is generated. In addition, you would never know how much HP you have, only a bar of health. Again the same thing can be done for MP. In addition the concept of a level should be eliminated. In which case players would have to come up with some new way of gauging skill. I mean underneath it all, there are still percentages and numbers involved, but they are never reported.

Though, with an open source game, it guess it would be *trivial* to change the UI in such a way that it would report numbers. Anyway, that\'s something that I don\'t have to worry my little brain with, but it would be pretty cool to see it implemented properly. I think that I\'ve ranted far too much, so I\'ll stop here.

Cheers

[Seytra]

Hmmm, welcome to the discussion!
I\'m going to comment on the quest spaces specifically since it seems that I failed to get them accross right.
They are not meant to get rid of the problem of XP lineup, but they are meant to get rid of the problem od dropped loot while still maintaining coherence. A good example is diablo, in which you, once you re-start the game, have to kill all the monsters again, open all the doors again, etc.. In a singleplayer game this can be overcome easily by saving the \"solve state\" of a quest and even savin the monsters that are left, but in a multiplayer game it can\'t that easily.
In an MMORPG, the quest needs to be there for each player. To allow it t be done only once is sufficient to inhibit XP lineup and harvesting. It doesn\'t solve dropped loot and immersion, hoever. The problem is that it simply is totally unrealistic that you just killed the evil / good wizard, and while still looking at the corpse (and looting it), another player enters the room and waits for it to respawn. It simpy can\'t be there after you killed it, so it hurts immersion _if_ it is there. If you can\'t kill it more than once, you can\'t level-grind, but still it\'s not good for immersion. So this is what quest spaces can achieve: you kill it, and it stays dead forever (unless there is a followup later), and nobody else who shouldn\'t be there can be there. Furthermore, when you reenter the place, it is exacrly like you left it, not reset to the default. This makes it more realistic. The questspaces would take up storage, therefore it\'s not feasible / desirable to store them serverside.

If the quest takes place in a general area, however, it cannot contain special monsters. It might contain special items, but you shouldn\'t be able to see them if you got them once, and also they should be still there if you haven\'t. But this does then mean that the quest isn\'t fir tof a general area. The only quests fit for general areas are IMO time-restricted ones, like a hunt for something / someone, who is there exactly once, and there is exactly one winner (like a tournament).
Any other quest is, by definition or common conception, meant to be solved by exactly one hero (or one group thereof), not all over again by all sorts of groups.

Also, it would encourage RP, because except for your groupmates and NSCs there just isn\'t anybody around to talk to. Even if you talk 1337 on end, it wouldn\'t bother anyone since you\'re in your own space. One is much more likely to stay IC if they don\'t encounter things that can\'t really _be_ IC (like the wizard example). After all, the quest experience is meant to be unique to the player (like in the sagas). Without questspaces, Odysseus would maybe have stopped at an island, just to see Odysseus sail by, or blind the giant again, while other Odysseus\'s would get in line to blind it. _That_ is what questspaces are intended to correct in the first place.