quote: Original post by Odd the Hermit
Well, my rough idea has been to have the teams able to select the length of game they want (one hour -> a day -> many weeks -> etc.), and then have the actual player characters be unique to that game instance.

Having the player characters unique each time does solve a lot of problems with gameplay and even server requirements. The biggest issue with it is people like to get attached to their characters and develop them over a long time. Also, people could build their character over the span of the session, but each time they play they would be retracing the same ground of character development. I can see this working in a different game, but not in an RPG.

quote: This prevents one person from loading up the game and playing for ten hours, leaving everyone in the dust, but, of course, it introduces other problems, most of which could be labelled "the player''s". Obviously, you''d need to get the entire team together to play the game at all--if my wife is busy with her writing, that would affect my entire team. My solution? A) It''s up to the players to coordinate play times (which is true of online and offline games anyhow); B) add in a "Who''s online" feature; C) allow a quorum of the players to decided to go ahead with the game, potentially with the computer playing for the people who aren''t there.

Interesting, this is very close what I had proposed originally. It was soon observed that people don''t always want to play with the same group and most of all they don''t want to depend on another''s schedule for fun. I became convinced that the difficulty of getting your party members online would drag the game into obscurity. As far as letting NPCs play for players, well, the social aspect is a huge part of the game and NPCs would take that away. How long would you play Counter Strike with a team full of bots?

----
Darryl Long: Lead Coder The Pythian Project

quote: Original post by Oort
Having the player characters unique each time does solve a lot of problems with gameplay and even server requirements. The biggest issue with it is people like to get attached to their characters and develop them over a long time. Also, people could build their character over the span of the session, but each time they play they would be retracing the same ground of character development. I can see this working in a different game, but not in an RPG.

Yeah, my solution''s not perfect, I know. Perhaps an option to start a game with a semi-developed character from previous rounds, as long as all of the characters are within a certain threshold of each other (maybe selectable by the person setting up the game in the first place?) would work.

I will say that, in addition to the load benefits, this solution allows for much more dynamic game design freedom--you can introduce new character classes, new areas, quests, etc. without impacting existing quests; it could also allow for introducing quests with specific limits (for exapmle: a quest which can only be performed by dwarven characters, or some such; haven''t thought too much about this, and it''s somewhat offtopic...) Quests (and even entire worlds) could be dynamically created using a Diablo-esque randomizer, taken to a new level. Etc.

quote: Original post by Oort
Interesting, this is very close what I had proposed originally. It was soon observed that people don''t always want to play with the same group and most of all they don''t want to depend on another''s schedule for fun. I became convinced that the difficulty of getting your party members online would drag the game into obscurity. As far as letting NPCs play for players, well, the social aspect is a huge part of the game and NPCs would take that away. How long would you play Counter Strike with a team full of bots?

True. Of course, there''s no reason for the player to be limited to one quest at a time; they''d just need to start a different quest, probably with different people, with a different character. Thus, if I''m working my way through a Massive Quest with three of my friends on Saturdays, I could still log in on other days and play Small Quests (or even other MQs) with others, without impacting the Saturday MQ. Again, I''d compare it to an offline game--trying to find a time when all of your friends can get together to play D&D can be hard, but it doesn''t prevent you from playing other games when not everyone can be there. And if one of your players goes away for a two-week vacation, the players & DM can decide to continue on without him, or with the DM making his character go through basic motions. Requiring a quorum helps ensure that you''re never playing with a team full of bots.

-Odd the Hermit

quote: Original post by Odd the Hermit
Yeah, my solution''s not perfect, I know.

Not perfect, but a pretty interesting idea. However, I think we''ve strayed from the original direction of the thread and now we''re talking about a completely different game The problem that got us onto this was players leaving behind their party members. After some reflection and rereading parts of my design document, I don''t think this will be a problem. Players can be members of a guild of many players and they should always be find someone of their approximate level to quest with. The short game time of a quest should make it easy for players to sort this out themselves.

I''d like to redirect the discussion to the combat portion of the game and involving the player in the combat without making it a FPS or click-fest. What do you think of the "combat strategy" idea?



----
Darryl Long: Lead Coder The Pythian Project

quote: Original post by Oort
I''d like to redirect the discussion to the combat portion of the game and involving the player in the combat without making it a FPS or click-fest. What do you think of the "combat strategy" idea?

Well, personally, I don''t like giving up control of my character. My playing/fighting style tends to be fairly different from anything an AI would come up with--computers aren''t that good at creative innovation. I even have problems when I''m playing Final Fantasy Tactics, and have to watch the AI-controlled Guests go wandering off to die...

So, having said that, you can probably appreciate that I have some reservations about the "You encounter an enemy; now sit back and watch the computer fight for you" system. Adding in some options for customizing the AI would help, but it''s still not the same as actually controlling your actions yourself. Can''t say as I''m really sure what good alternatives would be--all of the RPGs I''ve played are either Diablo-esque "Click on the monster and watch it die", Zelda: OoT "Dodge, attack, and roll out of the way" button mashers, or menu based systems where you can take your time trying to dig the appropriate selection out of the menus, or even leave the game for an hour without the monster getting bored and mauling you. (PSO was somewhere between Diablo and Zelda.) The first two options favor fighters, since they only have one option to select (Attack!), while the third could really annoy the other players as a mage spends five minutes looking for his "Pretty Colors" spell...

Wish I had something better to suggest.

-Odd the Hermit

quote: Original post by Odd the Hermit
So, having said that, you can probably appreciate that I have some reservations about the "You encounter an enemy; now sit back and watch the computer fight for you" system.

That''s basically the way NWN worked, as I recall, and it was pretty boring to sit there and wait for your character. I remember I would sit there and click repeatedly on the bad guy because the delay between attacks was so slow. At the same time, I remember my finger would get sore in Diablo because as fast as the attacks might be, I would still be mouseclicking 3 times for every attack

quote: The first two options favor fighters, since they only have one option to select (Attack!), while the third could really annoy the other players as a mage spends five minutes looking for his "Pretty Colors" spell...

I think most games have handled this by allow the mage to equip spells in the "quickslots". Making a really easy magic interface would help a lot as well (as opposed to working your way through the spellbook menus).

I''m thinking of having the player click on the enemy to melee attack, and the character will continue attacking until you disengage from the enemy. For every time you click on the enemy again, your character will attack immediately regardless of the peril this will place him in. When you''re not clicking, the character will try to defend and attack based on the level of aggression you''ve given him. I''ll also have the character auto-track the clicked enemy (keep facing him at all times). To circle your enemy, use the strafe keys. To disengage, press the "back" button.

What do you think?


----
Darryl Long: Lead Coder The Pythian Project

> I''m thinking of having the player click on the enemy to melee
> attack, and the character will continue attacking until you
> disengage from the enemy. For every time you click on the enemy
> again, your character will attack immediately regardless of the
> peril this will place him in. When you''re not clicking, the
> character will try to defend and attack based on the level of
> aggression you''ve given him. I''ll also have the character
> auto-track the clicked enemy (keep facing him at all times). To
> circle your enemy, use the strafe keys. To disengage, press the
> "back" button.

I like this idea, and think that it would work well as a balance between the two extremes ( click fest & menu controlled madness ). But maybe you could bring it a little further along, by allowing the player to have 4 distinct controls over thewir player while fighting ( this is all secondary to the AI controlled fighting of course. basically all these do id override what the ai was doing with what the player wants to do ).

1st - strafeing ( circleing ) around the other fighter
2nd - attack regardless
3rd - block regardless
4th - disengage from enemy and run away.

These should all have natural / obvious key bindings so they come easily to the player. I think that this combined with the AI attacking ( as defined by what ''mode'' the player has put it in ( discussed in earlier relpy ) ) would be enough to satisfy all diffrent types of players.

- n30n

quote: Original post by Anonymous Poster
1st - strafeing ( circleing ) around the other fighter
2nd - attack regardless
3rd - block regardless
4th - disengage from enemy and run away.

That''s a good idea. I can only see one problem and have a solution. The player would get frustrated if he was trying to carefully time an attack and the AI kept attacking for him. Therefore, we''ll allow the player to hold down the block button and the AI won''t attack as long as block is held down.

Borrowing an idea from "Blade of Darkness", we should use left-mouse for attack and right-mouse for block. It worked very well in that game.

Now, going back to the Human "Tools" concept. This was a really interesting idea, but I need some more ideas for tools. One was the grappling hook. I''ve added lockpicks and setting traps, although I had these as an ability of the Arbag Burglar before so I''m not too happy with these.

One idea is to allow the Human to carry literal tools, like a hammer and screwdrivers. With these, the Human can build things for his party to use (but these things would be one use only). For instance, the Human could build lockpicks for the Arbag Burglar, a grappling hook so the Berman can get on top of something to glide off, a wand that the Valun can imbue with a powerful one-use spell, and so on. Based on this, I could change the Human class to "Human Tinker". Are there any good ideas for what the Human could create?

----
Darryl Long: Lead Coder The Pythian Project

One reply to bunches of fragments...

quote: Oort : Each race has special abilties, except the Humans. As in most games, when we designed the races we just ended up with "Humans are average." I don''t like this very much.

How about item usage? For example, in many FRPGs, fighters can use all armour/weapons, mages can''t but only mages can use most magical items. Perhaps there''s a class of useful items that only humans can use? (kind of cheap, and reminiscent of how Bards/Gadgeteers are implemented in several games, but...)

(reading more carefully, I see this was already mentioned. oops...)

Okay, another idea: maybe Humans are the most literate, or are able to learn to read other languages, whereas the other races can only learn their own (and a few words in others).

Obviously you need a "common speak" language so the party can... be a party. ;-) But for reading, Humans could have ancient rune lore, read/decypher other races'' languages, etc. It would make the Human a powerful member of a party exploring ancient ruins, libraries, or places where things might be labelled.

quote: Back to puzzles, one idea I had was to make the puzzles depend on a races special abilities.

But then if you make, say, a Bermen-specific puzzle, and a party goes in but doesn''t have one of them... you want to make sure that either it''s solvable, or designed for several races, or there''s some notice up-front. (or I suppose you could force the party to give up, exit, find a Bermen, and return... but that''s cruel. ;-))

Still, without those sorts of puzzles, you''d be left with the user-based variety (logic, patterns, learn the password, throw switches, etc). Which are all fine and good, but you do want other sorts of puzzles.

quote: They are cooperative. Players form a party then embark on the quest together. XP is usually given out to the player who performs a task (e.g. killing a monster). In the case of major quest goals (e.g. reaching an important waypoint), the XP is divided evenly between the players.

I agree with the problem stated by someone else (combat-based characters get more XP).

Have you read/skimmed "The Complexity of Cooperation", by Axelrod? It goes into various multi-person models of the Prisoner''s Dilemma. An idea that might spring from that could be:

- you get some small XP for specific things (monster kills, assists, etc).
- most XP gets put into a party pool
- at the end, the players themselves help distribute the XP by voting.

Basically everyone votes on how "useful" someone was during the quest. The more positives someone gets, the more of the pool''s XP they get. And the way negatives are balanced is: you can vote to "punish" someone (negative rating) so they''ll get less XP -- but in doing so you also surrender some of your XP.

In general, though, you''ll still get some XP. If there are five people, say you might get between 10-30% of the party pool. It''s not perfect (a gang of six could still screw the one outsider who joined their group), but with some testing and balancing, it doesn''t seem unworkable.

quote: However, you might be on to something with the "characters attack in different ways". Maybe a weak connection between the player and the character can be established by having the player set the strategy of the character.

I like this, too -- choose berserk/normal/vulnerable/parry sorts of attacks, or even a slider for aggression level. I don''t mind sitting and watching combat, especially if those nuances have a significant effect. (vulnerable spot attacks may do an extra 8pts of damage, but they occur sporadically, so you might not want to go that route)

The player can also be entering long-term tactical goal/commands, such as "keep falling back until my Thur and I are back-to-back", or "edge over to the portcullis so I can lower it". Depending on items, it might take a while to get enough of an opening to pull a potion out of a pouch.

quote: Having the player characters unique each time does solve a lot of problems with gameplay and even server requirements. The biggest issue with it is people like to get attached to their characters and develop them over a long time. Also, people could build their character over the span of the session, but each time they play they would be retracing the same ground of character development. I can see this working in a different game, but not in an RPG.

Yes -- I would want to build my character. That''s one of the reasons I prefer D&D campaigns to "afternoon games" -- one afternoon isn''t even enough for me to learn what my own character is like!

Another quest thought: why must every player have the same quest goals?

Perhaps there is a general goal ("raise the flag on the citadel"). But each race might have distinct subgoals. A Golmek might want to "find the X that you have heard is in the basement and fix it". An Arbag might want to search for treasure. Some goals would be racially implicit, others might come in the form of "you recall an elder telling you that X was lost in battle, but was last seen in Khatovar".

This could add nice branches to the quests, and increase replay value (mostly when replaying as a different race, but still). If players knew that many quests had race-specific subgoals, it would promote diversified parties, which might be good anyway, as you could design puzzles without having to make them so every race could solve them.

Hi Merle. Thanks for your posts!

quote: Original post by Merle
Okay, another idea: maybe Humans are the most literate, or are able to learn to read other languages, whereas the other races can only learn their own (and a few words in others).

That''s a pretty good idea. Humans can develop language skills while other races can''t, so if you want to read a scroll to find a side-quest, you need a human.

quote:

quote: Back to puzzles, one idea I had was to make the puzzles depend on a races special abilities.

But then if you make, say, a Bermen-specific puzzle, and a party goes in but doesn''t have one of them...

Yes, the intention was to only make side-quests, for bonux XP, dependent on race skills.

quote: Have you read/skimmed "The Complexity of Cooperation", by Axelrod? It goes into various multi-person models of the Prisoner''s Dilemma. An idea that might spring from that could be:

- you get some small XP for specific things (monster kills, assists, etc).
- most XP gets put into a party pool
- at the end, the players themselves help distribute the XP by voting.

Basically everyone votes on how "useful" someone was during the quest. The more positives someone gets, the more of the pool''s XP they get. And the way negatives are balanced is: you can vote to "punish" someone (negative rating) so they''ll get less XP -- but in doing so you also surrender some of your XP.

That''s a really interesting idea! Very innovative. I found the document here, but I haven''t read it yet. Even without that, I can see your idea has creative merit and it''d be really interesting to see how it works. I''ll discuss it with some of my project members and see what they think.

----
Darryl Long: Lead Coder The Pythian Project